It can also own companies from several countries while depending on one industry.

These may both look diversified.

They are diversified against different risks.

Sector diversification spreads exposure across different kinds of businesses.

Country diversification spreads exposure across different economic, political, regulatory, and currency environments.

Neither one replaces the other.

A resilient portfolio should understand both.

Sector diversification asks what the companies do

Companies in different sectors make money in different ways.

A bank earns money through lending, deposits, fees, and financial services.

A miner extracts and sells natural resources.

A retailer distributes products to consumers.

A software company sells digital products or subscriptions.

A utility provides essential infrastructure.

These businesses respond differently to economic changes.

Higher commodity prices may benefit miners while increasing costs for manufacturers.

Higher interest rates may help some banks while hurting indebted businesses and expensive growth stocks.

Weak consumer spending may hurt retailers while essential-service companies remain more stable.

Sector diversification reduces the chance that one industry-specific problem damages every holding.

Country diversification asks where the system operates

Countries create different environments for companies and investors.

They have different:

• Currencies

• Interest rates

• Tax systems

• Regulations

• Political risks

• Economic cycles

• Natural resources

• Consumer markets

• Industry structures

• Accounting and reporting practices

A Canadian bank, German manufacturer, Australian miner, British insurer, and American software company operate under different national conditions.

Their businesses may still be connected globally.

But they do not depend completely on the same government, central bank, currency, or domestic economy.

Country diversification reduces dependence on one national environment.

Different sectors in one country still share national risks

Imagine a portfolio containing:

• A Canadian bank

• A Canadian railway

• A Canadian utility

• A Canadian retailer

• A Canadian software company

• A Canadian energy producer

The sectors appear diverse.

But every company remains exposed, directly or indirectly, to Canada.

They may share exposure to:

• The Canadian dollar

• Canadian interest rates

• Domestic taxation

• Federal and provincial regulation

• Canadian consumer spending

• Canadian housing

• Canadian capital markets

• Domestic political decisions

A national recession may affect all of them.

A currency decline may reduce their value for a foreign investor.

A major regulatory change may influence several sectors simultaneously.

The companies do different work.

They still inhabit the same national ecosystem.

Different countries in one sector still share industry risks

Now imagine a portfolio containing banks from:

• Canada

• The United States

• Germany

• Britain

• Australia

The country exposure is broader.

But the portfolio remains heavily dependent on banking.

The companies may all suffer from:

• Credit losses

• Deposit instability

• Falling loan demand

• Regulatory pressure

• Funding stress

• Property-market weakness

• Interest-rate shocks

The exact effects will differ by country.

A banking crisis may begin in one market and spread through global financial connections.

The portfolio crossed several borders.

It did not escape the economic biology of banks.

Two dimensions of diversification

Country and sector can be imagined as two separate axes.

One axis describes the company’s business model.

The other describes its national environment.

A portfolio with companies from one sector and one country sits in a narrow corner.

A portfolio with several sectors but one country spreads across only one axis.

A portfolio with one sector across several countries spreads across the other.

A stronger structure attempts to diversify across both.

This does not require owning every industry in every country.

It means recognizing that company type and national environment create different sources of risk.

Sector labels can hide shared national exposure

Companies from different sectors can still depend on the same domestic condition.

Consider:

• A homebuilder

• A mortgage lender

• A furniture retailer

• A building-material supplier

• A residential utility

• A real estate website

They belong to several sectors.

Yet all may depend heavily on one country’s housing market.

If domestic housing activity weakens, the entire group can struggle.

The sector labels did not reveal the shared national economic driver.

This is one reason a portfolio should be understood as a system rather than a collection of classifications:

Diversification should be based on causes, not merely labels.

Country labels can hide shared global exposure

Country diversification can also be overstated.

A Canadian miner, Australian miner, British-listed miner, and American mining company may trade in different markets.

But their revenue may depend on the same commodity.

A global decline in copper, iron ore, or gold prices can affect all of them.

Similarly, software companies listed in different countries may depend on the same technology spending cycle.

Automobile manufacturers from several countries may depend on similar supply chains and global consumer demand.

The flags are different.

The economic root may be the same.

Listing country is not always operating country

A company’s exchange listing does not necessarily describe where it makes money.

A company listed in Canada may earn most of its revenue in the United States.

A British-listed miner may operate mainly in Africa or South America.

A German manufacturer may sell products around the world.

An American technology company may generate more than half its revenue outside the United States.

This creates several possible country exposures:

• Listing country

• Headquarters country

• Revenue country

• Asset country

• Customer country

• Regulatory country

• Currency exposure

A simple country label captures only part of the picture.

The system should avoid assuming that a company is purely Canadian, American, German, British, or Australian merely because of where its shares trade.

Domestic companies can provide clearer country exposure

Some companies are primarily tied to one national economy.

A regional bank may lend mostly within its home country.

A domestic retailer may depend on local consumers.

A utility may serve one regulated region.

A railway may move goods across a defined national network.

These companies can provide clearer exposure to local economic conditions.

That can be useful when deliberately diversifying across countries.

But the concentration should remain visible.

A Canadian regional bank may provide Canadian financial exposure, not broad global diversification.

Multinational companies complicate the map

Multinational businesses can reduce dependence on one domestic market.

They may earn revenue from dozens of countries.

But this does not automatically make them perfectly diversified.

A company may operate globally while relying on:

• One production region

• One major currency

• One critical supplier

• One regulatory approval

• One technology platform

• One consumer market

Global revenue can create resilience.

It can also create geopolitical, currency, and operational complexity.

The company’s true geographic exposure should be understood rather than inferred from the number of countries listed in its annual report.

Countries experience different economic cycles

National economies do not always expand and contract at the same time.

One country may benefit from rising commodity prices.

Another may struggle with high energy costs.

One central bank may be cutting interest rates.

Another may still be raising them.

One housing market may be booming.

Another may be correcting after excessive borrowing.

Owning companies from several countries can reduce dependence on predicting one economic cycle correctly.

A weak period in one region may be offset partly by strength elsewhere.

The protection is imperfect.

Global crises can synchronize markets.

But national differences remain meaningful.

Industry composition differs by country

Stock markets do not contain the same mixture of businesses.

The Canadian market has substantial exposure to financials, energy, materials, and other resource-linked companies.

Australia also contains significant banks and miners.

The United States has a much larger technology sector.

Germany has major industrial and manufacturing exposure.

Britain contains financial, consumer, healthcare, resource, and multinational businesses.

Buying a broad market from one country is not economically identical to buying a broad market from another.

Country diversification can therefore create indirect sector diversification.

It can also accidentally increase exposure to sectors that dominate several national markets.

The portfolio must examine both dimensions together.

Currency creates a separate source of return

Foreign investing introduces currency movement.

A stock may rise in its home currency while producing a weaker result after conversion into the investor’s home currency.

A foreign currency can also strengthen and improve the final return.

This means country diversification often creates currency diversification.

That can reduce dependence on one home currency.

It can also increase volatility and introduce conversion costs.

The effect must be measured honestly:

A global portfolio cannot simply combine incompatible local prices and call the result diversified.

Currency exposure may differ from company exposure

A company may trade in euros but earn much of its revenue in U.S. dollars.

Another may trade in Canadian dollars while selling globally priced commodities.

The investor’s immediate currency exposure comes from the security and the portfolio’s conversion policy.

The company’s economic currency exposure comes from its revenue, costs, debt, and assets.

These can pull in opposite directions.

A weaker home currency may increase the translated value of foreign revenue.

It may also raise the cost of imported equipment.

Country diversification introduces these relationships whether the investor intends to forecast currencies or not.

Regulation is a national risk

Governments influence companies through:

• Taxes

• Labour laws

• Environmental rules

• Competition policy

• Financial regulation

• Foreign-ownership restrictions

• Trade rules

• Industry-specific approvals

A company can be financially strong while facing a sudden regulatory change.

A concentrated national portfolio depends heavily on one legal and political framework.

Country diversification spreads exposure across several systems.

That does not eliminate regulatory risk.

It prevents one policy decision from automatically controlling every holding.

Political stability matters differently across industries

Political risk does not affect every company equally.

A domestic retailer may be influenced by taxes, wages, and consumer policy.

A miner may depend on permits, land rights, royalties, and local relationships.

A bank may be deeply connected to financial regulation and national credit conditions.

A software company may face privacy rules, antitrust policy, or restrictions on data movement.

Country and sector exposures interact.

The same political event can affect industries differently.

The portfolio should therefore avoid treating geographic and business-model risk as independent checkboxes.

They form a matrix.

Tax systems affect shareholder outcomes

Different countries tax:

• Corporate income

• Dividends

• Capital gains

• Foreign investors

• Resource extraction

• Financial transactions

A company may look attractive before considering the taxes that apply to the investor.

Dividend withholding can reduce income from foreign holdings.

Tax treaties may change the result.

Certain account types may treat foreign income differently.

Tax should not be the only reason to own or avoid a company.

It is part of the real return.

Country diversification is not complete if the backtest models share prices but ignores how foreign ownership actually works.

Accounting practices and disclosure differ

Companies across countries may follow different reporting standards, filing schedules, and disclosure traditions.

Even when accounting frameworks are broadly comparable, data availability may differ.

One market may provide detailed quarterly reporting.

Another may rely more heavily on annual and interim statements.

Some historical filing dates may be difficult to obtain.

This creates research risk.

A global system should not interpret stronger data coverage as stronger company quality automatically.

It should recognize when the infrastructure understands one country better than another.

Home-country bias feels comfortable

Investors often prefer companies from their own country.

The businesses are familiar.

The currency is familiar.

The news is easier to follow.

The investor understands the political system and consumer culture.

This familiarity can reduce some forms of uncertainty.

It can also create hidden concentration.

Living, working, owning property, and holding investments in one country means several parts of the investor’s financial life depend on the same national economy.

Country diversification can reduce this overlap.

The investor does not need to reject the home market.

The goal is to recognize when familiarity has been mistaken for safety.

Familiarity is not the same as low risk

A domestic company may feel safer because its brand is visible.

The investor shops there, sees its advertisements, or knows people who work in the industry.

But familiar companies can still face:

• Excessive valuation

• Weak management

• Economic concentration

• Regulatory change

• Structural decline

• Competitive pressure

Foreign companies may feel riskier partly because the investor knows less about them.

That knowledge gap is real.

It should encourage careful research, not automatic exclusion.

Country diversification can protect against valuation concentration

One national market may become unusually expensive.

Another may remain more reasonably valued.

If the portfolio is restricted to the most popular market, the investor may be forced to choose among expensive opportunities.

Global diversification expands the opportunity set.

A strategy can compare similar businesses across countries and potentially find better prices.

However, lower valuation abroad may reflect:

• Slower growth

• Weaker governance

• Political risk

• Currency risk

• Poorer investor protection

• Different industry composition

Cheapness should be investigated, not assumed to be an automatic advantage.

Sector diversification can protect against economic shocks

Different sectors react differently to:

• Inflation

• Recession

• Interest rates

• Commodity prices

• Technology changes

• Consumer behaviour

A portfolio containing financials, industrials, consumer businesses, technology, healthcare, and resources may be less dependent on one economic mechanism.

But the benefit depends on what actually drives the companies.

A technology company and a biotechnology company may belong to different sectors while both depending on cheap capital and high investor risk tolerance.

Economic sensitivities matter more than formal labels.

Country diversification can protect against national shocks

A national shock might include:

• A domestic banking crisis

• A housing collapse

• A severe currency decline

• Political instability

• A tax change

• A trade conflict

• A natural disaster

• A regulatory intervention

A portfolio limited to one country may experience several companies weakening at once.

Foreign holdings may also decline because markets are connected.

But businesses outside the affected country may have different direct exposures.

Country diversification creates alternative economic pathways.

Global crises reduce diversification temporarily

During major financial stress, correlations often rise.

Investors sell assets across countries and sectors.

Credit becomes scarce.

Consumer and business confidence weaken.

A portfolio that appeared diversified may decline almost everywhere.

This does not prove diversification failed.

Diversification is not a promise that nothing will fall.

It aims to reduce dependence on one cause and improve the chance of eventual recovery.

Different countries may experience the crisis differently and recover at different speeds.

A diversified portfolio can still become a monoculture

Imagine owning:

• A Canadian oil producer

• An American oil producer

• A British-listed oil producer

• An Australian energy company

• A German industrial supplier dependent on oil investment

The countries differ.

The portfolio still depends heavily on energy.

Now imagine owning banks, real estate companies, retailers, and builders across one country during a housing boom.

The sectors differ.

The portfolio still depends on domestic property.

This is why both country and sector diversification can fail when they ignore shared roots:

Many names, sectors, or flags can still represent one economic bet.

The portfolio needs explicit country limits

A scoring system may rank several companies from one country near the top.

This could happen because:

• That market is unusually cheap

• Data coverage is stronger

• One sector is performing well

• The model fits that country particularly well

• Currency movement affected the signals

Buying only the highest-ranked companies may create national concentration.

A portfolio allocator can impose limits such as:

• Maximum number of companies per country

• Maximum capital allocation per country

• Minimum number of represented countries

• Country-specific liquidity requirements

• Limits based on revenue rather than listing location

The exact rules should be tested.

The principle is to stop the ranking system from accidentally turning one national pattern into the entire portfolio.

The portfolio also needs sector limits

Country limits alone could still produce:

• Five banks from five countries

• Four miners from four countries

• Several utilities across different markets

Sector constraints can prevent this.

Possible rules include:

• Maximum holdings per business model

• Maximum total weight in one sector

• Limits on commodity-linked exposure

• Limits on interest-rate-sensitive businesses

• Limits on companies sharing the same economic driver

The useful constraint may not always match an official sector classification.

The system may need its own economic categories.

Selection and allocation solve different problems

A company can rank highly because its financial evidence is strong.

That does not mean the portfolio needs another company from the same country or sector.

The allocator should be allowed to choose a slightly lower-ranked company when it improves the portfolio’s overall structure.

This follows the separation between finding attractive companies and combining them:

Selection asks which companies deserve consideration.

Allocation asks which combination deserves capital.

The highest-ranked country may not deserve the whole portfolio

Suppose Australian companies dominate the ranking during a commodity boom.

Their cash flow is strong.

Valuations appear low.

The top ten positions may contain seven Australian miners.

The ranking system may be working correctly at the company level.

The portfolio would still be making one large bet on:

• Australia

• Commodities

• China-linked demand

• The Australian dollar

• Resource-sector conditions

The allocator should recognize the concentration even when every company individually qualifies.

Two companies per country can be a useful starting rule

For a small portfolio of roughly eight to ten holdings, owning approximately two companies from each of several countries can create a simple structure.

It can prevent one market from dominating.

It can also encourage the system to search broadly.

But this should not become a rigid promise.

There may not always be two qualified opportunities in every country.

Forcing weak companies into the portfolio for symmetry would be a mistake.

Country targets should guide diversification without overriding minimum quality.

Holding cash may be better than filling a geographic quota with an unsuitable business.

Two companies from one country should still be different

Owning two Canadian banks does not create much internal Canadian diversification.

Neither do two Australian miners or two American technology companies.

Within each country allocation, the companies should ideally differ in:

• Business model

• Customer base

• Economic sensitivity

• Revenue source

• Balance-sheet structure

This creates diversification within diversification.

The country bucket should not become a miniature monoculture.

Equal weighting does not solve geographic concentration

Ten equal positions appear balanced.

But if six belong to one country, 60% of the portfolio remains nationally concentrated.

Equal weighting controls company-level exposure.

It does not control:

• Country exposure

• Sector exposure

• Currency exposure

• Shared economic drivers

Several layers of constraints may be needed.

Each constraint protects against a different kind of concentration.

Country weights should reflect more than market size

A market-cap-weighted global index gives the largest national markets the largest allocations.

That is a reasonable passive approach.

A small active portfolio may use different rules.

It may prefer:

• Equal country representation

• Opportunity-based country weights with limits

• Risk-based country caps

• A home-country anchor with foreign diversification

Each method has trade-offs.

A smaller country may offer attractive companies but limited liquidity.

A large country may contain more opportunities but also higher valuations.

The portfolio policy should match the strategy’s objective rather than imitate a benchmark accidentally.

Diversification should not force false precision

The system does not need to know the perfect allocation to Canada, Australia, Germany, Britain, and the United States.

A simple set of boundaries may be sufficient.

For example:

• No country controls the portfolio

• No sector controls the portfolio

• No single issuer controls survival

• No hidden economic theme dominates several positions

• Foreign returns are converted consistently

• Only qualified companies are included

This creates a broad safety structure without pretending the future can be optimized exactly.

Country diversification can improve learning

Testing strategies across several countries helps reveal whether an investment idea is general or market-specific.

Suppose a value signal works only in one country.

That may reflect:

• One historical period

• One sector mix

• One accounting convention

• One data-quality pattern

• A genuine local anomaly

The signal deserves more skepticism than one that behaves reasonably across several independent markets.

Country diversification therefore supports both portfolio resilience and research validation.

Sector diversification can test business-model generality

A strategy may appear successful because it works especially well for one kind of company.

That does not mean the same rules should be applied everywhere.

Testing across business models can reveal whether the strategy needs specialized animals.

A valuation measure useful for general operating companies may be inappropriate for banks, insurers, miners, or real estate businesses.

Diversification in the research universe should not be confused with universal scoring.

Different sectors may need different rules before their results can be combined.

A global system needs local knowledge

Country diversification is not achieved simply by downloading more tickers.

Each market may require understanding of:

• Exchange symbols

• Currency units

• Filing schedules

• Delisting history

• Liquidity

• Corporate actions

• Accounting conventions

• Broker access

A company can look attractive because the system misunderstood its local data.

Global diversification without data discipline can produce global errors.

The opportunity set becomes larger.

So does the number of ways the pipeline can fail.

Broker access sets a practical boundary

A company may qualify financially but remain unavailable through the investor’s broker.

A portfolio cannot diversify through securities it cannot purchase.

The system should record:

• Buyable

• Non-buyable

• Restricted

• Illiquid

• Unknown access

Country diversification must be investable, not theoretical.

An inaccessible foreign stock belongs in research, not in the live portfolio.

Country diversification has administrative costs

Foreign investing can introduce:

• Currency conversion

• Withholding taxes

• Different market hours

• Settlement differences

• Additional fees

• More complicated research

• Lower data availability

These costs should not be ignored.

A small portfolio does not need exposure to every country.

The benefit of each additional market should exceed the complexity it creates.

Several well-understood countries may provide more useful diversification than dozens of poorly understood ones.

Sector diversification has complexity costs too

Every business model requires specialized knowledge.

Understanding banks, miners, software, insurers, utilities, and biotechnology companies at the same depth is difficult.

A portfolio can become diversified on paper while the investor lacks the ability to evaluate several holdings properly.

This is another trade-off.

Diversification reduces concentration risk.

Excessive breadth can reduce analytical quality.

The system’s specialized animals are one way to address this by evaluating each company through appropriate rules.

The best portfolio may not maximize either form

Maximum country diversification could require owning weak companies from markets with few opportunities.

Maximum sector diversification could require owning industries the strategy cannot evaluate reliably.

The goal is not to maximize the number of flags or sector labels.

It is to reduce the largest avoidable dependencies while maintaining investment quality.

A portfolio can be sensibly diversified without representing every category.

Country and sector are only two layers

Other important dimensions include:

• Currency

• Company size

• Customer concentration

• Commodity sensitivity

• Interest-rate sensitivity

• Valuation style

• Business maturity

• Liquidity

• Data quality

A portfolio can be diversified by country and sector while remaining concentrated in expensive growth stocks.

It can be globally diversified while holding only highly leveraged companies.

No single label captures the full system.

Country and sector are useful starting points because they reveal different classes of dependence.

The diversification matrix

A practical way to view the portfolio is as a matrix.

Countries run along one side.

Business types run along the other.

Each company occupies a cell.

A concentrated portfolio may contain many holdings clustered in one row or column.

A more balanced portfolio spreads holdings across several cells.

The matrix does not need to be perfectly filled.

It provides a visual warning when several holdings repeat the same exposure.

It can also reveal empty areas where a future candidate might improve the portfolio.

Diversification should improve the portfolio, not decorate it

A company should not be purchased solely because it adds a new country or sector.

It must still meet the strategy’s financial standards.

Diversification is a constraint placed around good selection.

It is not a substitute for good selection.

Weak companies from different countries remain weak companies.

A resilient forest still needs healthy trees.

What I want my allocator to ask

When evaluating a candidate, the allocator should ask:

• Which country does it genuinely depend on?

• Which sector or business model drives its economics?

• What currency exposures does it create?

• Does the portfolio already contain similar risks?

• Would this company add a new source of return?

• Would it make one country or sector too dominant?

• Is the lower-ranked alternative more useful to the total system?

• Is holding cash better than forcing diversification?

These are portfolio questions.

They cannot be answered fully by the company’s standalone score.

A simple example

Suppose the system identifies five highly ranked candidates:

1. Canadian bank

2. Canadian insurer

3. Australian bank

4. German industrial company

5. American software company

The portfolio already owns a Canadian bank and an Australian bank.

The highest-ranked new company may be the Canadian bank.

But adding it would increase financial-sector and Canadian exposure.

The German industrial or American software company may rank slightly lower while contributing more diversification.

The best standalone company is not always the best next component.

Diversification should be measured after every decision

Portfolio risk changes as prices move.

A country that began at 20% may grow to 35%.

One sector may outperform and dominate the portfolio.

Currency movement can alter home-currency weights even without trading.

The allocator should therefore measure exposure periodically.

Diversification is not established once and forgotten.

It is a property of the current portfolio.

Rebalancing should not become constant trading

Exposure drift does not require immediate correction every day.

Frequent rebalancing can create costs and taxes.

The system can use boundaries.

For example:

• Allow moderate weight drift

• Review at scheduled intervals

• Rebalance when limits are breached

• Avoid trading when the improvement is minor

• Remove companies that no longer qualify

The goal is controlled diversification, not perfect geometric symmetry.

The system should explain its country decisions

If a candidate is excluded because the portfolio already has too much exposure to its country, that reason should be recorded.

Similarly, the system should identify whether the limit refers to:

• Listing country

• Headquarters

• Revenue exposure

• Asset exposure

• Trading currency

This makes the decision auditable.

A country rule that cannot explain what it measures may create false confidence.

The same applies to sectors

The system should know whether sector exposure is based on:

• Provider classification

• Internal business-model routing

• Revenue source

• Economic sensitivity

• Commodity dependence

Official classifications are useful but imperfect.

A company can sit in one sector while sharing risks with another.

The internal animal system may provide a more economically meaningful classification for portfolio construction.

Country diversification is not a prediction about nations

Owning several countries does not require forecasting which government, currency, or economy will outperform.

It reduces the need to make that forecast correctly.

The investor accepts that one country may struggle unexpectedly.

The portfolio retains exposure to other environments.

This is similar to owning several companies instead of one.

Diversification replaces the demand for perfect prediction with a structure designed to survive uncertainty.

Sector diversification is not a prediction about industries

The system does not need to know which sector will lead the market next year.

It can maintain exposure to several economic functions while selecting companies that meet its rules.

Some sectors will underperform.

Others may compensate.

The objective is not equal performance.

It is avoiding complete dependence on one industry outcome.

The two forms work together

Country diversification protects against national concentration.

Sector diversification protects against business-model concentration.

Together, they reduce the chance that one event damages every position for the same reason.

They are complementary.

A portfolio spread across several sectors in several countries has more potential pathways to survive.

That does not make it safe.

It makes its risks less dependent on one story.

The final distinction

Sector diversification asks:

Do my companies make money in different ways?

Country diversification asks:

Do my companies operate under different national conditions?

A portfolio needs both questions.

Several sectors in one country can still share currency, regulation, interest rates, and domestic economic risk.

Several countries in one sector can still share the same business cycle, commodity, or industry failure.

The number of sectors does not reveal the number of countries.

The number of countries does not reveal the number of economic engines.

Diversification becomes real only when the portfolio understands what each company depends on—and how those dependencies overlap.

A collection of flags is not automatically global diversification.

A collection of sector labels is not automatically economic diversification.

The goal is a portfolio whose future does not depend on one country remaining strong or one industry remaining dominant.

Country diversification changes the environment.

Sector diversification changes the organism.

A resilient portfolio needs more than one of each.

Subscribe now

Share

Share Fungal Intelligence for Stocks

Leave a comment

It is a wonderful company with strong management, rising cash flow, little debt, a durable competitive advantage, and a stock price far below its true value.

You understand the business completely.

The risks appear manageable.

The upside looks enormous.

Why dilute that opportunity by buying anything else?

Because the perfect idea may exist only inside the investor’s mind.

A company can appear nearly flawless while still containing risks that have been misunderstood, underestimated, or never discovered.

That is why I would rather own a small portfolio of good ideas than place everything into one supposedly perfect idea.

I do not need every stock to be extraordinary.

I need the portfolio to survive when one of my conclusions is wrong.

Conviction does not remove uncertainty

Research can increase confidence.

It cannot eliminate uncertainty.

An investor may study:

• Financial statements

• Management history

• Industry competition

• Customer behaviour

• Valuation

• Debt

• Cash flow

• Economic conditions

The thesis may be thoughtful and well supported.

The company can still fail.

A competitor may introduce a better product.

A regulator may change the rules.

Management may make a destructive acquisition.

A key customer may leave.

Fraud may be discovered.

A recession may arrive.

The investment may simply have been purchased at the wrong time.

Research narrows uncertainty.

It does not turn the future into a known variable.

The best-understood company can still surprise you

Investors often believe concentrated portfolios are safe when the businesses are understood deeply.

That is partly true.

Understanding a company is better than owning something blindly.

But depth of knowledge can create another danger.

The more time invested into studying one company, the more psychologically difficult it becomes to recognize contradictory evidence.

The thesis becomes part of the investor’s identity.

Every new fact is interpreted through the existing belief.

Weak results become temporary.

Rising debt becomes strategic.

Dilution becomes an investment in growth.

A falling price becomes proof that the opportunity is even better.

The investor may understand the company better than most people and still misunderstand the investment.

A perfect thesis can contain one fatal assumption

An investment thesis often rests on several assumptions.

For example:

• Revenue will continue growing

• Margins will remain stable

• Debt can be refinanced

• Customers will remain loyal

• Management will allocate capital intelligently

• Competition will remain manageable

• The valuation will eventually rise

Most of these assumptions may be correct.

One can still destroy the outcome.

A company does not need to fail completely for the investment to disappoint.

It may simply grow more slowly than expected.

Margins may decline slightly.

The market may assign a lower valuation.

A concentrated position turns one mistaken assumption into a portfolio-level event.

Ten good ideas do not need to be perfect

A good investment idea should have:

• A reasonable business

• Understandable economics

• Financial evidence

• A valuation that provides some protection

• Risks that can be identified

• A plausible path to acceptable returns

It does not need to be the single greatest company available.

It does not need to dominate every financial category.

It does not need a dramatic story.

A portfolio containing ten such ideas can tolerate imperfection.

Some companies may underperform.

Some may remain flat.

A few may do well.

The result comes from the group rather than one heroic prediction.

Diversification is protection against ignorance

Diversification is sometimes described as protection for investors who do not know what they are doing.

That framing misses the point.

Even skilled investors do not know everything.

Diversification acknowledges that limitation.

It says:

I have evidence, but I may still be wrong.

This is not a rejection of conviction.

It is a boundary around conviction.

The investor can believe strongly in an idea without allowing that belief to determine the survival of the entire portfolio.

One position can dominate more than the return

A very large position affects the investor psychologically.

Every price movement becomes important.

Every news release demands attention.

A weak quarter feels personal.

The investor begins searching constantly for confirmation or reassurance.

This can distort decision-making.

A concentrated position may cause the investor to:

• Ignore new opportunities

• Defend weak management

• Trade emotionally

• Monitor the stock excessively

• Change rules after losses

• Take more risk to recover

The position occupies not only capital but attention.

A balanced portfolio distributes both financial and emotional pressure.

The portfolio should not require one company to be right

A resilient portfolio does not depend on every company succeeding.

It especially should not depend on one company succeeding.

This is part of thinking about the portfolio as a connected system rather than a list of independent tickers:

The quality of a portfolio is determined partly by how much damage one incorrect idea can cause.

A company may deserve inclusion without deserving control.

Concentration makes forecasting more important

Suppose one stock represents 50% of the portfolio.

The investor must be highly accurate about:

• The company

• The industry

• The valuation

• The timing

• The broader economy

• The market’s future expectations

A mistake in any one of these areas can dominate the result.

When ten companies each represent approximately 10%, the required precision falls.

The investor still needs a useful selection process.

But one error is less likely to become permanent financial damage.

Diversification reduces the amount of forecasting skill required for survival.

The arithmetic of loss is unforgiving

Large losses require even larger gains to recover.

A 10% loss requires an 11.1% gain to return to the starting point.

A 25% loss requires a 33.3% gain.

A 50% loss requires a 100% gain.

A 75% loss requires a 300% gain.

This asymmetry matters.

Suppose one position represents the entire portfolio and falls by 70%.

The investor needs an extraordinary recovery merely to return to the beginning.

If the same company represents 10% of the portfolio and becomes nearly worthless, the damage remains serious but survivable.

Avoiding catastrophic loss can be more important than maximizing the gain from the best prediction.

A ten-stock portfolio can survive one complete failure

Imagine an equal-weighted portfolio containing ten companies.

Each begins at 10%.

One company fails completely.

The direct loss is approximately 10%, assuming the others do not change.

That is painful.

It is not fatal.

The remaining nine companies continue operating.

Future savings can still be invested.

The strategy can learn from the failure.

Now imagine the same company represented 80% of the portfolio.

The mistake becomes the defining financial event.

A robust system should expect that complete failures are possible, even when the screening process is careful.

Equal weighting limits false precision

Suppose the system ranks ten qualified companies.

The first receives a score of 86.

The tenth receives a score of 80.

Does the six-point difference justify giving the first company five times more capital?

Perhaps.

But the score may contain:

• Measurement error

• Arbitrary weights

• Filing-timing differences

• Temporary price movement

• Missing data

• Model uncertainty

The system may be good at identifying a useful group without being good at predicting the exact order of future returns.

Equal weighting reflects that possibility.

It allows the ranking system to select the ingredients without pretending it knows the exact amount of each ingredient required.

This is why selection and sizing should remain distinct decisions:

Finding an attractive company does not automatically determine how much capital it deserves.

Equal weighting lets unexpected winners matter

Investors often assume their highest-conviction idea will produce the best result.

Reality may choose differently.

The seventh-ranked company may become the strongest performer.

The apparently ordinary business may execute exceptionally well.

The favourite may disappoint.

Equal weighting gives every selected company enough capital to contribute meaningfully.

It prevents the investor’s confidence ranking from overwhelming the evidence that arrives later.

Diversification should not become random collection

Owning ten ideas does not help if all ten depend on the same underlying force.

Ten oil producers remain one commodity bet.

Ten regional banks exposed to the same property market remain one credit bet.

Ten speculative growth companies may all depend on low interest rates and investor enthusiasm.

The number of companies is not the same as the number of independent risks.

A useful ten-stock portfolio should contain businesses with different sources of strength and different ways of failing.

This is why avoiding portfolio monocultures matters:

The goal is not merely ten names.

It is ten reasonably independent ideas.

Good ideas should come from several environments

A resilient portfolio may include companies from different:

• Countries

• Industries

• Business models

• Currency exposures

• Economic sensitivities

• Sources of cash flow

One company may benefit from economic expansion.

Another may sell essential services.

One may benefit from higher commodity prices.

Another may benefit when input costs fall.

One may be domestically focused.

Another may earn revenue globally.

These differences do not guarantee that the portfolio will avoid losses.

During a crisis, many assets can decline together.

But different economic roots reduce the chance that one specific problem destroys everything.

Geographic diversification adds another layer

A portfolio concentrated in one country depends on that country’s:

• Currency

• Political system

• Regulation

• Interest rates

• Economic cycle

• Industry structure

• Market valuation

Even a strong national market can experience long periods of weakness.

Owning companies from several countries spreads exposure across different environments.

This introduces currency and data complications.

Those complications should be handled rather than used as a reason to ignore the rest of the world.

My long-term structure may contain roughly two companies from each of several countries.

The exact number can change.

The principle is to prevent one national outcome from controlling the portfolio.

Ten is not a magical number

There is nothing mathematically perfect about owning exactly ten companies.

Eight may be reasonable.

Twelve may be reasonable.

The correct number depends on:

• Strategy quality

• Available opportunities

• Position size

• Portfolio complexity

• Transaction costs

• The investor’s ability to understand holdings

I am attracted to roughly eight to ten positions because it creates a balance.

The portfolio is concentrated enough for each holding to matter.

It is diversified enough that one mistake should not destroy everything.

The number creates discipline without pretending to be universal.

Too few positions increase fragility

A portfolio of two or three companies can produce extraordinary gains.

It also places enormous weight on a small number of judgments.

Company-specific events become portfolio events.

A legal dispute, product recall, regulatory decision, or accounting problem can cause major damage.

Even when the companies belong to different industries, the portfolio remains vulnerable to individual surprises.

The investor may understand each business deeply.

The future can still introduce information that no amount of research previously revealed.

Too many positions can dilute understanding

Owning hundreds of stocks creates a different problem.

The portfolio begins to resemble the market itself.

Individual research matters less.

Weak ideas enter because each position appears too small to matter.

Monitoring becomes difficult.

The investor may no longer understand what the portfolio owns or why it owns it.

At some point, buying a low-cost index may be more honest than operating a complicated imitation of one.

The goal is not maximum diversification.

It is sufficient diversification.

Ten positions force meaningful selection

When the portfolio has only a limited number of spaces, each company must earn inclusion.

A new company cannot enter merely because it appears interesting.

It must be more useful than something already owned.

The question becomes:

Does this company improve the portfolio enough to deserve one of the limited positions?

That encourages discipline.

A candidate may be attractive but redundant.

Another may be slightly lower-ranked yet provide a new source of return.

A limited portfolio creates competition among ideas.

A stock can be good but unnecessary

Suppose the portfolio already contains two Canadian banks.

A third bank appears attractively valued and financially strong.

The stock may be a good investment.

It may not be the best portfolio addition.

Buying it increases exposure to:

• Canadian credit conditions

• Housing

• Interest rates

• Domestic regulation

• The Canadian dollar

A lower-ranked industrial or consumer company from another country might improve the overall system more.

The portfolio does not exist to collect every good company.

It exists to combine enough good companies into a resilient whole.

Perfect ideas can lead to perfect-story investing

The search for one perfect company encourages storytelling.

The investor begins imagining:

• The company dominating its market

• Margins expanding indefinitely

• Management executing every plan

• Competitors remaining weak

• Valuation eventually becoming generous

The thesis becomes a complete future narrative.

Every part depends on the others.

Ten good ideas require less narrative perfection.

Each company needs only a reasonable path to success.

The portfolio does not require any one story to unfold exactly as imagined.

The best historical strategy may not produce the best future stock

A backtest can identify a strategy that ranked certain types of companies successfully in the past.

It cannot reveal with certainty which current candidate will become the greatest winner.

The top-ranked stock may benefit from historical noise.

The score may reflect conditions that are about to change.

A diversified portfolio accepts that uncertainty.

This follows the same reason I do not immediately trust the highest-performing historical strategy:

A champion selected from historical data may be skilled, lucky, or both.

The same is true of the top-ranked company.

Diversification protects the strategy from ranking error

A scoring model does not need to identify the exact future winner to be useful.

It may only need to create a group that performs better than the rejected population.

Suppose the top ten stocks collectively behave well, but their internal order is mostly unpredictable.

A concentrated portfolio using only number one may fail.

An equal-weighted portfolio using the group can still capture the broader signal.

This is similar to ensemble methods in machine learning.

One prediction can be unstable.

A group of reasonably independent predictions can produce a more reliable result.

Ten ideas create more learning opportunities

A one-stock portfolio produces one major observation.

Did the company perform well or poorly?

A ten-stock portfolio produces a richer set of evidence.

The system can ask:

• Which financial characteristics helped?

• Which business models struggled?

• Did country diversification matter?

• Did the strongest score outperform?

• Did one industry dominate?

• Did equal weighting help?

• Were losses caused by data, strategy, or uncertainty?

More independent decisions create a stronger research sample.

The goal is not to trade frequently.

It is to learn from more than one outcome.

One result should not determine whether the strategy is trusted

Suppose the concentrated company performs exceptionally well.

The strategy may look brilliant.

But one outcome does not reveal whether the process was repeatable.

The company may have benefited from luck.

The investor may mistake an extraordinary result for proof.

A portfolio of several ideas gives the process more opportunities to demonstrate whether its selections are consistently reasonable.

One spectacular winner still matters.

It does not need to carry the entire argument.

A concentrated winner can encourage dangerous escalation

An investor who makes a large gain from one concentrated bet may increase concentration next time.

The success reinforces the belief that deep conviction deserves maximum capital.

The next idea may not work.

This is how a lucky result can become a dangerous lesson.

A disciplined portfolio structure prevents one success from rewriting the risk rules.

Position limits should survive both wins and losses.

Owning several ideas reduces timing dependence

A company can be attractive and still decline after purchase.

The business may take longer to improve.

The market may remain pessimistic.

A recession may delay results.

A concentrated investor may need to time the entry almost perfectly.

A portfolio of several companies spreads timing risk.

Not every holding will begin working at the same moment.

Some may rise while others remain weak.

The portfolio can compound without requiring one exact entry point to be correct.

The portfolio can rebalance among imperfect outcomes

Equal weighting creates a natural discipline.

When one company rises substantially, its portfolio weight increases.

When another falls, its weight decreases.

At a scheduled rebalance, the system may trim the winner and add to selected companies that remain attractive.

This can systematically sell some relative strength and buy some relative weakness.

Rebalancing is not automatically profitable.

It can cut exposure to a great compounder or add to a deteriorating company.

That is why continued eligibility checks matter.

But the structure prevents one successful stock from gradually becoming the entire portfolio without an explicit decision.

Winners should be allowed to grow within boundaries

Equal weighting does not require constant adjustment.

A portfolio can begin with equal positions and allow weights to drift within limits.

This reduces trading costs and lets successful companies contribute.

The system may rebalance only when:

• A scheduled review occurs

• A position becomes excessively large

• The company no longer qualifies

• A better portfolio combination becomes available

The exact policy should be tested.

The principle remains:

No company should become dominant merely because nobody noticed its weight increasing.

Position limits are not predictions

A 10% position limit does not mean the company has only a 10% chance of succeeding.

It means the portfolio refuses to let one opinion control more than a defined share of the capital.

This is risk architecture.

The limit exists because future outcomes are uncertain, not because every company is equally attractive.

A bridge has load limits even when engineers believe the structure is strong.

The limit protects against conditions that were not modelled perfectly.

Ten good ideas can still produce meaningful upside

Diversification is sometimes treated as surrendering the possibility of exceptional returns.

But a ten-stock portfolio can still benefit greatly from a major winner.

Suppose one 10% position rises fivefold.

Its contribution to the starting portfolio is roughly 40 percentage points, before considering rebalancing and the other holdings.

The investor did not need to put everything into the company to benefit meaningfully.

A great investment can transform a portfolio without being allowed to destroy it.

The portfolio does not need every company to win

Imagine ten equal positions:

• Two perform exceptionally well

• Three produce moderate gains

• Two remain roughly flat

• Two decline

• One fails completely

The overall portfolio can still succeed.

This is the advantage of combining imperfect but positively selected ideas.

The process does not require perfection from every component.

It requires the winners and survivors to outweigh the mistakes.

Concentration can still be appropriate for some investors

Some investors possess unusual expertise, access, temperament, and willingness to accept large losses.

A concentrated approach may suit them.

Famous investors have built fortunes through a small number of major decisions.

That does not make concentration universally correct.

Their success may depend on:

• Decades of experience

• Control over the company

• Access to management

• Exceptional analytical skill

• Permanent capital

• A high tolerance for volatility

• Opportunities unavailable to ordinary investors

Copying the concentration without possessing the supporting advantages can reproduce the risk without reproducing the edge.

My system is designed around humility

My project is not built on the belief that I can identify one perfect company with certainty.

It is built around a more modest claim:

A structured process may be able to identify a small group of companies with reasonable financial strength, valuation, and survival characteristics.

That claim still needs testing.

Until it earns strong evidence, the allocation system should remain humble.

My starting structure is likely to include:

• Roughly eight to ten holdings

• Approximately equal initial weights

• No duplicate issuer exposure

• Several countries

• Limits on shared economic risks

• No leverage

• No one company capable of determining survival

Complex position sizing can come later, if it proves useful.

The allocator should earn the right to concentrate

A future system may estimate that one company has a better risk-adjusted opportunity than another.

It may consider:

• Downside probability

• Data quality

• Historical strategy reliability

• Liquidity

• Market regime

• Correlation

• Valuation

Perhaps this evidence eventually supports unequal sizing.

But the burden of proof should be high.

The allocator must demonstrate that its confidence estimates are useful outside the data where they were developed.

Until then, equal weighting is a defence against pretending the model knows more than it does.

Shadow portfolios can test concentration safely

Different allocation rules can be tested in shadow portfolios.

One version may use equal weights.

Another may use moderate score-based weights.

Another may use risk-based caps.

Their rules should remain frozen.

Then the system can compare:

• Return

• Drawdown

• Concentration

• Turnover

• Stability

• Behaviour during stress

• Dependence on one position

The goal should not be to select whichever version wins during one short period.

The goal should be to learn whether complexity creates durable improvement.

The perfect company may already be perfectly priced

Even when the business is exceptional, the stock may not be.

A company admired by everyone may trade at a valuation requiring years of near-perfect execution.

The business can perform well while the investment disappoints.

A portfolio of several reasonably valued companies reduces dependence on one valuation thesis.

Some may exceed expectations.

Others may disappoint.

The group does not require the market to reprice one company exactly as expected.

Diversification creates room to admit mistakes

When one company controls most of the portfolio, admitting the thesis is wrong becomes extremely painful.

Selling locks in a major loss.

Holding risks further damage.

The investor may delay because the consequences are too large.

A smaller position makes rational correction easier.

The company can be removed without destroying the portfolio.

The system can learn and continue.

Good risk design makes honesty less expensive.

Ten ideas protect the project itself

My investment system is still being built and tested.

Early strategies will make mistakes.

Data problems will be discovered.

Rules will need revision.

A concentrated portfolio would convert every early weakness into a large financial consequence.

A small equal-weighted portfolio allows the system to prove itself gradually.

The purpose of the first real-money deployment should not be to maximize wealth immediately.

It should be to verify that the process works outside research.

The portfolio should survive my own development

The strategy I use years from now may be better than the first version.

That future improvement is valuable only if capital survives long enough to benefit from it.

A concentrated early mistake can prevent later learning from mattering.

Survival keeps the experiment alive.

It preserves:

• Capital

• Confidence

• Flexibility

• Future opportunities

• The ability to improve

A portfolio designed for learning should not require the first model to be perfect.

Good enough can compound

Investors are drawn toward extraordinary opportunities.

But long-term wealth does not require every holding to become extraordinary.

A collection of reasonably purchased, financially functioning companies can compound through:

• Earnings growth

• Cash generation

• Dividends

• Buybacks

• Valuation normalization

• Time

The result may be less dramatic than one perfect bet.

It may also be much more survivable.

Compounding rewards durability.

A process that remains alive can continue collecting returns and learning from mistakes.

The goal is not to avoid conviction

I still want the system to express opinions.

It should reject weak companies.

It should rank the survivors.

It should select the opportunities with the strongest evidence.

A portfolio containing every available stock would abandon the purpose of research.

The goal is disciplined conviction.

The strategy says:

These companies appear better than the alternatives.

The allocator adds:

None of those conclusions is certain enough to control everything.

Selection creates focus.

Diversification creates survival.

The best portfolio may contain no perfect idea

A portfolio can succeed without owning the single best-performing stock in the market.

It can succeed through a group of companies that:

• Generate cash

• Maintain manageable obligations

• Avoid severe dilution

• Trade at sensible valuations

• Operate across different environments

• Survive long enough to compound

None needs to be perfect.

The system itself creates strength by combining them carefully.

The final choice

One perfect idea offers the possibility of an extraordinary outcome.

It also asks one company, one management team, one industry, and one thesis to carry the entire future.

Ten good ideas accept a less dramatic truth.

I may be wrong about some of them.

The data may be incomplete.

The future may produce surprises.

The highest-ranked company may not become the best performer.

But the portfolio does not need perfection.

It needs enough independent sources of value that mistakes remain survivable.

I would rather own ten companies that each deserve a place than one company that must justify everything.

The perfect idea asks me to predict the future.

Ten good ideas allow me to participate in it without requiring that I understand every part of it in advance.

Subscribe now

Share

Share Fungal Intelligence for Stocks

Leave a comment

The company may perform well.

Its local share price may increase.

But if the company’s currency falls far enough against the investor’s home currency, the final return can shrink—or disappear entirely.

The reverse can also happen.

A mediocre local return can become attractive because the foreign currency strengthened.

This is why a global backtest cannot simply collect stock prices from several countries and place them into one portfolio.

Every return must eventually be expressed in a common currency.

Otherwise, the portfolio may report profits that no investor actually earned.

A stock has more than one return

Suppose a Canadian investor buys a German stock.

The stock rises by 12% in euros.

That is the local-market return.

But the investor began with Canadian dollars and eventually wants the investment measured in Canadian dollars.

During the holding period, the euro may strengthen or weaken against the Canadian dollar.

The investor’s result therefore depends on two movements:

1. The stock’s return in euros

2. The euro’s movement against the Canadian dollar

The stock and the currency work together.

A global investment is not merely ownership of a foreign business.

It also creates currency exposure.

A simple example

Imagine a stock begins at €100 and ends at €110.

The local return is 10%.

At the beginning, suppose one euro is worth $1.50 Canadian.

The investment begins with a Canadian-dollar value of:

€100 × $1.50 = $150

By the end, suppose the euro has weakened to $1.30 Canadian.

The final Canadian-dollar value becomes:

€110 × $1.30 = $143

The company’s stock rose by 10% in its home market.

The Canadian investor lost about 4.7%.

Nothing is wrong with the company’s local price history.

The missing piece is the currency conversion.

Currency can also improve the result

Now imagine the same stock rises from €100 to €110, but the euro strengthens from $1.50 Canadian to $1.65.

The final value becomes:

€110 × $1.65 = $181.50

The investor’s Canadian-dollar return is 21%.

The stock contributed 10%.

Currency movement added the rest.

This does not mean the investor intentionally made a currency forecast.

The exposure appeared automatically when the foreign asset was purchased.

Local returns are still useful

A stock’s native-currency return tells us something important.

It reflects how the company was valued in its local market.

It allows comparisons with domestic competitors.

It helps separate company performance from currency movement.

But local returns do not represent the complete experience of a foreign investor.

A proper system may therefore preserve both:

• The native-currency stock return

• The portfolio-currency return

The first helps analyze the business and its local market.

The second tells us what the investor actually earned.

One portfolio needs one measuring unit

A global portfolio may contain companies trading in:

• U.S. dollars

• Canadian dollars

• British pounds

• Australian dollars

• Euros

• Other currencies

These values cannot be added directly.

A $1,000 position, a £1,000 position, and a €1,000 position do not represent the same amount of capital.

Before measuring portfolio weights or returns, the values must be converted into a shared unit.

This is similar to measuring a building.

One worker reports centimetres.

Another reports inches.

Another reports metres.

The measurements may all be correct individually.

Adding them without conversion produces nonsense.

A portfolio requires a common financial language.

Trading currency and reporting currency may differ

Currency problems become more complicated because a company can involve several currencies at once.

A business may:

• Trade in one currency

• Report financial statements in another

• Earn revenue across several currencies

• Be evaluated by an investor using a third currency

For example, a company may trade in London while earning most of its revenue in U.S. dollars.

Another may trade in Canada while reporting certain operating results in U.S. dollars.

The ticker’s exchange does not always tell us the complete economic exposure.

The system must distinguish among:

• Trading currency

• Financial-statement currency

• Operational currency exposure

• Portfolio reporting currency

These are related but different concepts.

Financial ratios also require currency consistency

Currency conversion is not only a return problem.

It affects valuation.

Suppose a company’s market capitalization is calculated in Canadian dollars while its free cash flow remains in U.S. dollars.

Dividing one by the other produces a number.

That number has no valid economic interpretation.

Before calculating ratios such as price to free cash flow, the numerator and denominator must use compatible units.

This applies to:

• Market value and earnings

• Enterprise value and operating profit

• Price and book value

• Debt and cash flow

• Portfolio value and position size

A clean ratio requires more than the correct formula.

Its ingredients must speak the same currency.

This is one way apparently attractive opportunities can be manufactured by a broken pipeline:

A currency mismatch can make a normal company look extraordinarily cheap or absurdly expensive.

Pounds and pence are not the same unit

Some markets create an additional complication.

A stock listed in London may be quoted in pence rather than pounds.

One pound contains one hundred pence.

A displayed stock price of 250 may mean 250 pence, or £2.50.

If a database interprets the number as £250, the stock becomes one hundred times more expensive than it really is.

If the opposite error occurs, the company may appear one hundred times cheaper.

This is not an exchange-rate problem in the usual sense.

It is a unit problem inside the currency.

A robust database must track both:

• Currency

• Quotation unit

“GBP” and “GBX” cannot be treated as interchangeable without conversion.

Currency errors often create extreme results

A currency mismatch can produce valuation ratios that look almost impossible.

That is exactly why the error can attract attention.

A company may appear to have:

• A price-to-earnings ratio below 1

• A free-cash-flow yield above 100%

• A market capitalization smaller than one quarter’s profit

• A sudden historical collapse or explosion in value

These results can look like extraordinary opportunities.

They should first be treated as validation failures.

Extreme numbers are often evidence that the system should inspect:

• Currency

• Units

• Share count

• Corporate actions

• Reporting period

• Company identity

The strongest apparent bargain may be the record most urgently requiring repair.

The exchange rate must match the date

Using the correct currency pair is not enough.

The system also needs the correct exchange rate for the relevant date.

Suppose a historical portfolio buys a German stock on March 31, 2018.

Using today’s euro exchange rate to convert that purchase would rewrite the historical transaction.

The correct rate should correspond to the investment date—or to a documented nearby date when markets were open.

The same principle applies when:

• Calculating entry value

• Calculating exit value

• Measuring daily portfolio value

• Converting financial statements

• Calculating position sizes

• Rebalancing the portfolio

A point-in-time backtest needs point-in-time currency information.

Missing FX dates create hidden assumptions

Stock markets and currency markets do not always share identical calendars.

A stock exchange may be closed while the currency still trades.

A currency data source may omit a holiday.

A financial statement may use a weekend period-end date.

The system may therefore need a rule for missing exchange-rate dates.

Possible approaches include:

• Use the most recent prior available rate

• Use the next available rate

• Use a monthly average

• Reject the calculation until an exact rate exists

Each approach has consequences.

The important thing is not pretending that no decision was made.

The policy should be documented and applied consistently.

A silent fill can create point-in-time errors that are difficult to detect later.

Monthly averages and daily rates answer different questions

A company’s annual financial statements may reasonably be translated using an average exchange rate for flows such as revenue or expenses.

A balance-sheet item may be translated using the rate on the reporting date.

A stock transaction should usually use the rate available on the transaction date.

These are different conversion problems.

Revenue accumulated throughout a year.

Cash on the balance sheet existed at one point in time.

A stock was purchased on a specific date.

Using one universal exchange rate for every purpose may simplify the database while distorting the economics.

The conversion rule should match the type of value being converted.

Currency conversion must not create look-ahead bias

Suppose an exchange-rate series is revised later.

Or a monthly average is calculated using the entire month, including days after the simulated investment decision.

The backtest may quietly receive future currency information.

This can matter when the strategy uses converted financial metrics for ranking.

If a company’s valuation is calculated using an exchange rate unavailable on the decision date, the historical screen is no longer honest.

The same point-in-time discipline applied to financial filings must also apply to FX data.

Otherwise, the backtest may know more than the investor could have known.

This is one reason impressive historical results require careful skepticism:

A correct formula applied with future currency information still produces an invalid test.

Currency affects position sizing

Suppose the portfolio intends to allocate $10,000 Canadian to each of ten holdings.

One stock trades in U.S. dollars.

Another trades in euros.

Another trades in Australian dollars.

The system must convert the intended Canadian-dollar allocation into each trading currency before determining how many shares to purchase.

If the exchange rate is wrong, the positions will not be equal.

One may receive $8,000 of exposure.

Another may receive $12,000.

The portfolio appears equal-weighted in the model but is unequal in reality.

Accurate currency conversion is therefore part of allocation, not merely performance reporting.

Rebalancing creates repeated FX decisions

A long-term global portfolio does not convert currency only once.

Each rebalance may involve:

• Selling foreign shares

• Converting the proceeds

• Purchasing securities in another currency

• Measuring new target weights

• Recording transaction costs

If a strategy rotates among countries, its currency exposure also changes.

A backtest that ignores these conversions may understate costs and mismeasure allocation.

The effect may be small for a low-turnover strategy.

It can become substantial when rebalancing is frequent.

Currency conversion has costs

The market exchange rate shown in a database may not be the exact rate available to a retail investor.

Real conversions can include:

• Bid-ask spreads

• Broker markups

• Conversion commissions

• Settlement differences

• Minimum fees

These frictions reduce returns.

A backtest does not need to model every brokerage detail perfectly to be useful.

But it should not assume unlimited free conversion when the portfolio regularly trades across currencies.

The more frequently the strategy converts capital, the more important these costs become.

Holding foreign currency can sometimes reduce conversion frequency

An investor may maintain cash balances in several currencies.

Proceeds from selling a U.S. stock can remain in U.S. dollars and later fund another U.S. purchase.

This can reduce repeated conversion costs.

But it creates another portfolio position:

Foreign cash.

That cash changes value relative to the investor’s home currency.

The system must decide whether currency balances are:

• Converted immediately

• Held until needed

• Treated as portfolio assets

• Included in performance

There is no universal correct answer.

The rule should reflect how the strategy would actually operate.

Hedged and unhedged returns are different strategies

Some investors accept currency movement.

Others hedge it using financial instruments designed to reduce exchange-rate exposure.

These approaches should not be mixed.

An unhedged foreign-stock return includes currency movement.

A hedged return attempts to isolate more of the local asset return, though hedging has costs and may be imperfect.

The choice depends on:

• Investment horizon

• Currency volatility

• Hedging costs

• Portfolio objectives

• Available instruments

A backtest should clearly state whether foreign positions are hedged.

Otherwise, two strategies may appear comparable while carrying different sources of risk.

Currency exposure can create diversification

Currency movement is not always an unwanted problem.

It can provide diversification.

A Canadian investor whose portfolio is entirely denominated in Canadian dollars depends heavily on one currency.

Owning assets connected to U.S. dollars, euros, or other currencies may reduce that concentration.

During some periods, foreign currencies may strengthen when the investor’s home market weakens.

This can support portfolio value.

During other periods, it can work against the investor.

Currency exposure is neither automatically good nor bad.

It is a source of risk and return that should be visible.

The company’s economic exposure may not match its listing

A Canadian-listed company may earn most of its revenue in the United States.

A British-listed miner may sell commodities priced globally in U.S. dollars.

A German manufacturer may export heavily to Asia.

The trading currency tells us how the shares are quoted.

It does not tell us where the company earns its money.

This distinction matters because currency can affect the business before it affects the investor.

A weaker home currency may increase the translated value of foreign revenue.

It may also increase the cost of imported materials.

The company may use hedges of its own.

A complete analysis separates:

• The investor’s currency exposure

• The company’s operating currency exposure

The two can interact in complicated ways.

Converted statements can hide native economics

For comparison, the database may convert every company’s financial statements into one currency.

That makes cross-country ratios easier to calculate.

But the native values should remain preserved.

Suppose a company’s revenue grows by 5% in its home currency.

After conversion, it appears to fall because the currency weakened.

The business did not necessarily shrink locally.

The investor’s translated view changed.

Both facts matter.

Native values help measure operating progress.

Converted values help compare capital across the global portfolio.

Deleting the native record would lose important context.

This is another reason raw financial history should remain durable:

Conversion policies can change.

The original evidence should remain available for rebuilding.

One conversion direction must be chosen consistently

Currency pairs can be quoted in opposite directions.

One source may report:

USD per EUR

Another may report:

EUR per USD

These values are reciprocals.

Confusing them can completely reverse a conversion.

If one euro equals 1.10 U.S. dollars, then one U.S. dollar equals about 0.91 euros.

Multiplying when the system should divide can create large errors.

The database should define every FX pair clearly.

Names such as EUR_USD can still be ambiguous unless the convention is documented.

A robust pipeline should know:

• Base currency

• Quote currency

• Conversion direction

• Formula applied

Synthetic currency pairs require care

A data provider may not supply every currency pair needed by the portfolio.

The system may need to calculate one through a common currency.

For example:

• Convert Australian dollars to U.S. dollars

• Convert U.S. dollars to Canadian dollars

• Combine the two to obtain Australian dollars to Canadian dollars

This is a synthetic cross-rate.

It can be perfectly valid.

But both component rates must correspond to compatible dates and conventions.

A mistake in either leg affects the final result.

The derived pair should retain provenance showing how it was constructed.

Stable conversion policy matters more than perfect precision

Historical FX data from different sources may vary slightly.

One may use daily closing rates.

Another may use central-bank reference rates.

Another may use market averages.

The exact values may not match perfectly.

For many long-term backtests, consistency is more important than chasing false precision.

The system should choose a reasonable source and apply it uniformly.

Switching providers whenever one produces a more attractive result would contaminate the research.

A conversion policy should be based on accuracy and availability—not on whether it improves performance.

The database and strategy have different FX responsibilities

The database should preserve:

• Native values

• Currency identifiers

• FX rates

• Dates

• Units

• Source information

• Conversion results

• Missing-rate states

The strategy decides:

• Portfolio reporting currency

• Whether positions are hedged

• Whether foreign cash is held

• Which conversion costs are applied

• How currency affects allocation

This separation follows the broader distinction between financial memory and investment policy:

The database provides the conversion evidence.

The strategy determines how foreign exposure is handled.

FX errors can contaminate every layer

A single currency problem can spread widely.

It may affect:

• Market capitalization

• Valuation ratios

• Screening eligibility

• Company rankings

• Position sizes

• Portfolio weights

• Historical returns

• Risk statistics

• Outcome labels

• Machine-learning targets

By the time the final backtest looks wrong, the original FX mistake may be several layers away.

This is why currency validation should happen early in the pipeline.

Bad conversion should not be allowed to become a sophisticated investment score.

Useful validation checks

A global system can test for suspicious results.

For example:

• Did the exchange rate change by an impossible amount?

• Is a British stock being treated as pounds when it is quoted in pence?

• Are the statement and market-value currencies aligned?

• Is a required date missing?

• Did converted market capitalization change dramatically without a price or share-count change?

• Are reciprocal currency pairs consistent?

• Does another reliable source show a similar rate?

These checks do not guarantee perfection.

They catch many errors before those errors become investment decisions.

The home currency must be explicit

“Global return” is incomplete without naming the investor’s measuring currency.

A Canadian investor, American investor, and European investor can earn different home-currency returns from the same stock over the same period.

The underlying company and local share-price return are identical.

The conversion path differs.

A backtest should therefore state its portfolio currency clearly.

For my system, I may standardize research values into a common reference currency while also preserving native information.

The important part is that the choice is explicit and reproducible.

Comparing countries requires converted outcomes

Suppose the system wants to compare six-month returns among companies from Canada, Australia, Germany, Britain, and the United States.

Using only native returns can be useful for studying local stock selection.

But if those companies are competing for positions inside one real portfolio, converted returns matter.

A 15% Australian-dollar gain and a 12% euro gain cannot be compared completely without considering how those currencies moved relative to the portfolio currency.

Otherwise, the system may learn that one country produced stronger returns when much of the difference came from FX.

That may still be useful information.

It should not be mislabeled as company-selection skill.

Currency can distort outcome labels

Future models may be trained using labels such as:

• Six-month return

• Six-month market-relative alpha

• Maximum drawdown

• Disaster outcome

• Twelve-month continuation

If those outcomes mix native and converted currencies inconsistently, the model learns from incompatible targets.

One country’s companies may appear systematically better or worse because of the currency treatment.

The problem is especially dangerous because the labels look like ordinary numbers.

The model may find strong patterns.

It may be learning a pipeline inconsistency.

The output currency of every outcome should be defined clearly.

Benchmark returns require the same treatment

Market-relative performance compares a stock with a benchmark.

The stock and benchmark must be measured consistently.

Suppose a German stock’s return is converted into Canadian dollars, while the German benchmark remains in euros.

The resulting alpha mixes company performance with currency movement.

The comparison is invalid.

Possible valid approaches include:

• Compare stock and benchmark in their shared native currency

• Convert both into the portfolio currency

• Separate local alpha from FX contribution

Each answers a different question.

The system should decide which question it wants to ask.

Decomposing the return improves understanding

A foreign investment’s return can be separated conceptually into:

• Local stock return

• Currency return

• Interaction between the two

This helps explain performance.

Suppose a position gained 18% in Canadian dollars.

Perhaps the stock rose 10% locally and currency contributed most of the rest.

Another position may have fallen 5% even though the company rose locally, because its currency weakened sharply.

Without decomposition, the investor may credit or blame the company for movement caused elsewhere.

A transparent system should explain both sources.

Currency movements can dominate short periods

Over long periods, business performance may become the primary driver of value.

Over shorter periods, currency movement can dominate.

A company may report no major change, yet its converted stock value moves significantly because exchange rates changed.

This matters when the strategy uses short holding periods or frequent rebalancing.

A six-month strategy may experience much more FX noise than a twenty-year investor expects.

The system should not assume currency becomes irrelevant simply because the selected companies are fundamentally strong.

Global diversification is still valuable

These complications do not mean investors should avoid foreign markets.

Global investing can provide access to:

• Different industries

• Different economic cycles

• Different valuations

• Different currencies

• Companies unavailable domestically

• Reduced dependence on one country

The lesson is not that currency risk makes global investing impossible.

It is that the benefit must be measured honestly.

Diversification cannot be evaluated accurately when each country’s returns use a different ruler.

The goal is an investable backtest

A useful backtest should describe something an investor could realistically have done.

For a global portfolio, that means knowing:

• What each stock cost in its trading currency

• What the exchange rate was

• How much home-currency capital was allocated

• What conversion costs applied

• What the position was worth later

• Whether proceeds remained foreign or were converted

• What information was available on each date

Ignoring currency creates a simpler simulation.

It may also create a simulation no investor could reproduce.

The final principle

Currency conversion does not create value.

It reveals whose value is being measured.

A company operates in one economic world.

Its stock trades in a market currency.

The investor experiences the result in a portfolio currency.

Those layers cannot be collapsed carelessly.

A global backtest should preserve native prices, native financial statements, trading currencies, historical FX rates, and the exact conversion policies used.

It should measure every portfolio position in a common unit without erasing the local evidence underneath it.

Otherwise, stock-selection skill, currency movement, and data errors become mixed together.

A foreign stock can rise while the investor loses money.

It can fall locally while currency softens the damage.

A company can appear cheap merely because its financial statements and market value use different units.

The arithmetic may look correct while the investment never truly existed.

Global investing expands the opportunity set.

Currency conversion makes that opportunity set measurable.

Without it, the portfolio is not global.

It is a collection of incompatible numbers.

Subscribe now

Share

Share Fungal Intelligence for Stocks

Leave a comment

An investment strategy can produce a clear buy signal while depending on only a small part of that database.

These two systems work together.

They are not the same thing.

The database preserves evidence.

The strategy applies beliefs, rules, and objectives to that evidence.

Confusing those responsibilities can create a system that appears intelligent but is difficult to verify, modify, or trust.

That is why I treat my financial database and my investment strategy as separate parts of the project.

A database describes

A financial database attempts to describe the market.

It may contain:

• Company identities

• Financial statements

• Filing dates

• Reporting periods

• Share prices

• Trading currencies

• Exchange rates

• Share counts

• Corporate actions

• Industry classifications

• Missing-data records

• Provider metadata

None of these facts tells the investor what to do by itself.

A company reported $500 million in revenue.

Its share count increased by 4%.

Its debt declined.

Its stock traded at a particular price.

These are observations.

They become investment information only after a strategy places them into context.

A strategy decides

An investment strategy asks questions that the database cannot answer alone.

For example:

• Is the company profitable enough?

• Is its debt manageable?

• Is the valuation attractive?

• Is growth healthy?

• Is dilution excessive?

• Does it fit the portfolio?

• Should it be rejected?

• How much capital should it receive?

These questions depend on objectives.

A conservative value strategy and an aggressive growth strategy can examine the same company and reach opposite conclusions.

Neither conclusion is stored naturally inside the company’s financial statements.

The difference comes from the rules being applied.

Facts do not contain their own verdict

Suppose a company has debt-to-equity of 1.8.

Is that high?

The database cannot answer without additional context.

For a stable infrastructure business, the debt may be manageable.

For a fragile cyclical company, it may be dangerous.

For a bank, the ratio may not even be the correct way to think about financial structure.

The number is real.

The verdict depends on the business model and the investment strategy.

That is why facts and conclusions should remain distinguishable.

The same database can support opposing strategies

Imagine two investors studying the same market.

The first searches for high-quality companies with stable cash flow and moderate valuations.

The second searches for distressed companies with a chance of recovery.

The quality investor may reject a business because:

• Earnings are negative

• Debt is rising

• Revenue is falling

• Cash flow is unstable

The distressed investor may examine that exact company because those weaknesses caused the price to collapse.

Both strategies can use the same database.

They simply interpret the evidence differently.

A database should not quietly delete the company because one strategy does not want it.

The evidence may still matter to another strategy—or to evaluating whether the first strategy’s rejection rule was useful.

A screener belongs to the strategy layer

A stock screener often looks like a database tool because it searches financial fields.

But a screener is already expressing an investment opinion.

The moment it says:

• Market capitalization must exceed $300 million

• Free cash flow must be positive

• Debt-to-equity must remain below 2.5

• Price-to-free-cash-flow must remain below 25

it has moved beyond recording facts.

It is applying a strategy.

The company does not become objectively unsuitable because it failed those rules.

It becomes unsuitable for that particular investment process.

This is why a screener’s role should be understood clearly:

The database stores the company.

The screener decides whether the company may proceed.

The database should not care which strategy wins

A reliable database should remain neutral between competing ideas.

Suppose Strategy A performs better when it emphasizes valuation.

Strategy B performs better when it emphasizes quality.

Strategy C performs better when it combines both.

The underlying historical evidence should remain unchanged across all three tests.

If each strategy uses a differently cleaned or differently filtered version of history, the comparison becomes unfair.

The database should create a common world.

The strategies should compete inside that world.

Strategy rules should remain replaceable

Investment theories change.

A ratio that appears powerful in one test may fail in another.

A threshold may turn out to be too strict.

A company classification may need improvement.

The portfolio may require stronger country limits.

These are normal research developments.

If the rules are embedded directly into the data, changing the strategy may require rewriting the database.

That risks changing the historical world whenever the investment theory changes.

A better design keeps the evidence stable and makes the strategy replaceable.

This is the broader architecture behind:

The database should outlive individual strategies.

A database is broader than a strategy

A strategy uses only the fields needed for its current decisions.

The database should usually preserve more.

A general operating strategy may currently care about:

• Revenue

• Net income

• Free cash flow

• Debt

• Book value

• Share count

• Valuation

But future research may need:

• Inventory

• Receivables

• Lease obligations

• Interest expense

• Research spending

• Segment results

• Acquisition history

• Customer concentration

If the system stores only what one strategy currently requires, future questions become impossible to test.

The database should preserve possibilities.

The strategy should narrow them into decisions.

A strategy is more than a formula

It is easy to imagine an investment strategy as a scoring equation.

Perhaps value receives 25%, quality receives 25%, growth receives 20%, and safety receives 30%.

But a complete strategy contains much more.

It includes:

• Which companies are eligible

• Which business models are excluded

• Which data is required

• Which values trigger rejection

• How companies are ranked

• How many positions are held

• How often the portfolio rebalances

• How positions are sized

• When a stock is sold

• How trading costs are handled

• What happens when no candidate qualifies

The database cannot make these choices automatically.

They reflect the investor’s goals and tolerance for uncertainty.

The database can be correct while the strategy is wrong

Suppose the financial data is accurate.

Prices, currencies, filing dates, and share counts have all been handled correctly.

The strategy still performs poorly.

Perhaps it:

• Overvalues growth

• Ignores deterioration

• Uses weak valuation measures

• Concentrates in one industry

• Trades too frequently

• Overfits historical relationships

This is a strategy failure, not a database failure.

The distinction matters because the repair is different.

The data does not need to be recollected.

The investment rules need to be reconsidered.

The strategy can be reasonable while the data is wrong

The opposite can also occur.

The strategy may have sensible rules, but it receives:

• A stale stock price

• The wrong company’s financial statement

• An outdated share count

• A currency mismatch

• A duplicated filing

• A future filing inside a historical test

The strategy produces a bad decision because the evidence was corrupted.

Changing the investment rules will not repair the problem.

The pipeline must be fixed.

This is how poor inputs can manufacture attractive-looking stocks:

A system must know whether it misunderstood reality or received the wrong reality.

The distinction makes failures useful

When an investment system fails, I want to know where it failed.

Was the problem:

1. Source data?
   The provider returned an incorrect or incomplete record.

2. Transformation?
   Currency, units, dates, or corporate actions were handled incorrectly.

3. Classification?
   The company was judged using the wrong business model.

4. Screening?
   The mouth admitted a company that should have been rejected.

5. Scoring?
   The strategy emphasized the wrong financial characteristics.

6. Allocation?
   Reasonable companies were combined into a fragile portfolio.

7. Uncertainty?
   The process was defensible, but the future still went badly.

Each failure teaches a different lesson.

Mixing the database and strategy together makes these lessons harder to recover.

A database should preserve rejected companies

When a strategy rejects a company, the company should remain in the database.

Otherwise, the research history becomes biased toward the strategy’s preferences.

Rejected companies are valuable because they allow the system to measure:

• How many later recovered

• How many failed

• Whether thresholds were too strict

• Whether missing data caused excessive rejection

• Whether the strategy avoided major losses

• Whether certain business types were treated unfairly

A database should preserve the full population.

The strategy creates a temporary eligible population from it.

The database should also preserve the strategy’s decisions

Keeping the layers separate does not mean investment decisions should disappear.

The system should record:

• Which strategy version ran

• Which companies were eligible

• Which companies were rejected

• Why each rejection occurred

• Which scores were produced

• Which portfolio was constructed

• Which data snapshot supported the decision

This creates an audit trail.

The decision belongs to the strategy layer, but it should be permanently associated with the evidence that produced it.

Raw evidence allows the strategy to be rebuilt

Suppose I later discover that my definition of free cash flow was too simplistic.

If raw cash-flow records remain available, I can create a better definition and rerun the strategy.

Suppose a country requires a different filing-availability policy.

The historical snapshots can be rebuilt.

Suppose an entire business category needs specialized metrics.

The companies can be reclassified and reprocessed.

This is possible because the evidence was preserved before the strategy compressed it into scores.

That is why raw history should survive every current model:

Strategies are experiments.

The raw warehouse is the laboratory record.

A database does not create an edge by itself

A large financial database may be valuable infrastructure.

But possessing more data does not automatically produce better investment decisions.

The investor still needs to decide:

• Which information matters

• Which relationships are durable

• Which companies are comparable

• Which risks deserve rejection

• Which valuations provide enough margin of safety

A database can make research possible.

It cannot guarantee that the research question is intelligent.

More information can produce more sophisticated mistakes when the strategy lacks discipline.

A strategy does not create truth

A strategy can express a strong opinion.

It can rank every company from best to worst.

That confidence does not make the conclusion true.

A precise score may rest on:

• Uncertain estimates

• Arbitrary weights

• Weak historical relationships

• Incomplete data

• An overfit backtest

The strategy should therefore remain accountable to the evidence.

It should be possible to trace every conclusion backward.

The more decisive the output, the more important that chain becomes.

Machine learning still belongs to the strategy layer

A future machine-learning model may identify relationships too complicated for fixed rules.

It might estimate future returns, risk, or the probability of deterioration.

But the model is still an interpretation system.

It learns from selected features and selected outcomes.

Its predictions depend on choices involving:

• Training periods

• Feature definitions

• Labels

• Missing-data handling

• Model architecture

• Validation rules

The model does not replace the database.

It depends on it.

Powerful models increase the need for reliable, traceable evidence because mistakes become harder to see inside complex predictions.

The portfolio is not stored in the company data

A company can look attractive individually and still be a poor portfolio addition.

The database may show that several oil producers are financially strong and inexpensive.

The selection strategy may rank all of them highly.

The portfolio layer must recognize that owning several of them creates one large commodity exposure.

This is another example of a decision that exists outside the raw company facts.

The database describes each tree.

The portfolio strategy decides how the forest should be assembled.

Strategy performance should not rewrite the data

Suppose a strategy performs badly during a certain period.

There may be a temptation to:

• Remove unusual companies

• Replace inconvenient missing values

• Adjust classifications

• Change the historical universe

• Prefer a different provider because it produces better results

Some corrections may be legitimate.

But data changes should be made because they improve accuracy, not because they improve strategy performance.

The database should not be optimized to make the strategy look intelligent.

The strategy should be tested against the most honest database available.

One supports many; the other chooses one path

The database should be capable of supporting many strategies.

A value strategy may use it one way.

A quality strategy may use it another.

A bank-specific model may select different nutrients from a miner-specific model.

The strategy chooses one path through the evidence.

That path may succeed or fail.

The database remains available for the next question.

A useful architectural sequence

The project can be understood as a series of distinct layers:

1. Raw warehouse
   Preserve filings, provider responses, prices, currencies, identities, and metadata.

2. Cleaning and normalization
   Resolve units, dates, corporate actions, mappings, and obvious contradictions.

3. Point-in-time snapshots
   Reconstruct what information was available on each historical decision date.

4. Business classification
   Determine which financial rules belong to each company.

5. Screening
   Reject candidates that fail non-negotiable requirements.

6. Feature compilation and scoring
   Calculate relevant nutrients and rank the survivors.

7. Portfolio allocation
   Combine companies while controlling position size and shared risk.

8. Outcome tracking
   Record what happened after the decision.

9. Validation
   Test whether the strategy survives unseen conditions.

The first three layers primarily construct the historical world.

The later layers decide how to act inside it.

The distinction creates trust

When data and strategy are separated, the system can answer two different questions.

What did the system know?

This can be traced through:

• Source records

• Filing dates

• Prices

• Currencies

• Missing values

• Transformations

Why did the system act?

This can be traced through:

• Strategy version

• Screening rules

• Scores

• Thresholds

• Portfolio constraints

• Position sizes

Trust requires both answers.

A decision without evidence is unsupported.

Evidence without a decision process cannot explain the portfolio.

The database is memory; the strategy is policy

A useful way to understand the distinction is:

The database is memory.

It preserves what entered the system.

The strategy is policy.

It determines how the system responds.

Memory should be broad, durable, and honest.

Policy should be explicit, testable, and replaceable.

A system with policy but no reliable memory repeats mistakes.

A system with memory but no policy collects information without acting.

Both are necessary.

Neither should impersonate the other.

The final difference

A database asks:

What happened, when did it happen, and what evidence do we have?

An investment strategy asks:

Given that evidence, what should we reject, select, own, and risk?

The database attempts to reconstruct reality.

The strategy expresses a disciplined opinion about that reality.

The database can be accurate while the strategy fails.

The strategy can be sensible while the data pipeline fails.

Keeping them separate allows the project to discover which one needs repair.

The database remembers every company, including the failures and rejected candidates.

The strategy chooses which companies matter for one particular objective.

The database should survive changes in investment theory.

The strategy should earn trust through testing.

One preserves the world.

The other decides how to move through it.

That is the difference between a database and an investment strategy.

Subscribe now

Share

Share Fungal Intelligence for Stocks

Leave a comment

It has consistent names, standardized currencies, resolved duplicates, aligned periods, and calculated investment metrics.

That makes it tempting to keep only the cleaned result.

Once revenue growth, free cash flow, debt ratios, and valuation scores have been calculated, why preserve the awkward source records underneath them?

Because the cleaned result contains assumptions.

Those assumptions may later prove incomplete, inconsistent, or wrong.

If the original evidence has been deleted, the system cannot return to the point before the mistake was made.

That is why I want to keep raw financial data permanently.

The raw database is not merely storage.

It is the project’s memory.

Clean data is an interpretation

A company might report operating cash flow of $120 million and capital expenditures of $35 million.

The system calculates free cash flow of $85 million.

That looks simple.

But even this basic calculation contains decisions:

• Which cash-flow period was selected?

• Was the capital-expenditure value already negative?

• Were both figures reported in the same currency?

• Was the statement annual, quarterly, or trailing?

• Was the filing later restated?

• Was the company mapped to the correct security?

• Was the figure reported directly or estimated?

The final number is useful.

It is not raw truth.

It is the result of a process applied to the evidence.

Keeping the source data allows that process to be examined and repeated.

The database should remember more than the strategy needs today

My current strategy may use a limited set of metrics.

It may care about:

• Free cash flow

• Net income

• Revenue growth

• Debt

• Book value

• Share dilution

• Valuation

• Deterioration

• Trading liquidity

That does not mean these are the only fields that will ever matter.

A future version of the system may need:

• Inventory

• Receivables

• Customer concentration

• Segment results

• Capitalized costs

• Lease obligations

• Pension assumptions

• Interest coverage

• Acquisition spending

• Research expenses

• Deferred revenue

If I collect only what the current strategy uses, the database becomes trapped inside the current theory.

When the theory changes, the evidence required to test the new idea may already be gone.

This is one reason I am building a database instead of relying only on a finished screener:

A screener answers a predefined question.

A raw database preserves the ability to ask new questions later.

Raw data is like source code

In programming, a compiled application is useful because it can run.

But the executable is not a substitute for the source code.

The source code explains how the program was built.

It can be inspected, repaired, modified, and compiled again.

Clean financial features are similar to the executable.

Raw filings, provider responses, prices, mappings, and metadata are the source.

If I keep only the finished features, the system may continue running, but I lose the ability to understand exactly how it was constructed.

The calculations may still work.

The knowledge behind them becomes fragile.

Strategies change

Investment research is not a one-time process.

A strategy may begin by focusing heavily on valuation.

Later, testing may show that cheap companies with deteriorating cash flow perform poorly.

The system adds a deterioration check.

Later still, it may discover that companies issuing large amounts of stock frequently destroy shareholder value.

The system adds dilution.

Each improvement requires historical evidence.

If the raw data survives, the new feature can be compiled across the entire history.

The updated strategy can then be tested fairly.

If only the old scores remain, the system cannot reconstruct what would have happened under the improved rules.

The strategy changes.

The historical world should remain available.

Metric definitions change too

Even when the general idea remains useful, its definition may improve.

Free cash flow is a good example.

One version may calculate:

Operating cash flow minus capital expenditures

Another may attempt to distinguish maintenance spending from expansion spending.

Another may exclude unusually large working-capital movements.

A company’s result can change depending on the definition.

The original cash-flow statement should remain untouched.

Each strategy version can calculate the feature it needs from the same underlying evidence.

This makes comparisons cleaner.

The database stores the source.

The strategy stores the definition.

Raw data protects the separation between facts and opinions

An investment score is an opinion produced from evidence.

A filing value is closer to the evidence itself.

The two should not be confused.

Suppose a system stores only:

Quality score: 72

That number tells me almost nothing by itself.

Why was the score 72?

Which fields were used?

Which values were missing?

Which strategy version created it?

Was the company compared with banks, miners, or general operating businesses?

A score can be rebuilt when the evidence and rules remain available.

The evidence cannot be recovered from the score.

This is why my architecture separates the database from the decision-making layers:

The raw warehouse preserves what entered the system.

The strategy decides what to do with it.

Providers can change their history

Financial-data providers sometimes correct errors.

They may update old statements.

They may change field names.

They may replace a missing value.

They may revise how a metric is calculated.

A query made today may not return the same record that was received a year ago.

If I preserve each source response, I can see what the system actually knew at the time.

That matters for both debugging and historical testing.

Without the original response, it may become impossible to determine whether a changed result came from:

• A new company filing

• A provider correction

• A revised mapping

• A new calculation rule

• A software bug

• A different source entirely

Raw preservation creates a stable reference point.

Restatements require more than one version of history

Companies sometimes restate earlier financial results.

A previously reported value may be corrected months or years later.

For current analysis, the restated number may be the best available estimate of reality.

For point-in-time backtesting, the original number may be more important.

An investor making a decision before the restatement did not know the corrected value.

A durable warehouse may therefore need to preserve both:

• The statement originally available

• The later corrected statement

These are two different forms of truth.

One answers:

What is now believed to have happened?

The other answers:

What could an investor have known then?

Deleting the original version destroys the second question.

Filing dates must survive alongside financial periods

A statement may describe the year ending December 31.

That does not mean the market received it on December 31.

It might have been published in February, March, or later.

The period-end date describes the business activity.

The filing date describes when the information entered the world.

Both matter.

If the cleaned database keeps only the period-end date, a historical strategy may gain access to results before they were published.

That creates look-ahead bias.

Preserving raw dates and source metadata helps keep the simulation honest.

The financial value alone is not enough.

The timing of knowledge is part of the evidence.

Raw prices matter too

A historical price series may eventually be adjusted for:

• Stock splits

• Dividends

• Currency conversion

• Exchange differences

• Missing trading days

• Corporate actions

The adjusted series is useful for return calculations.

The original trading prices should still be preserved.

Suppose a stock split was handled incorrectly.

If only the adjusted history remains, it may be difficult to determine what happened.

With the original values and the corporate-action record, the series can be rebuilt.

Raw prices also help answer execution questions.

What price was actually quoted?

How much volume traded?

Was the price stale?

Was the security genuinely available?

An adjusted return series is not a complete trading record.

Currency conversions should remain reversible

Global financial data creates several layers:

• The company’s reporting currency

• The stock’s trading currency

• The investor’s portfolio currency

• The exchange rate used on each date

A cleaned feature may show that a company generated $200 million in U.S.-dollar-equivalent cash flow.

That is convenient for comparison.

But the original value may have been reported in euros, pounds, Canadian dollars, or Australian dollars.

Keeping the native figure and the exchange-rate record allows the conversion to be checked.

It also allows the system to change its currency policy later.

A Canadian investor and a U.S. investor may care about different converted outcomes.

The original currency remains the common source.

Raw data helps expose fake opportunities

Bad data can make a company appear dramatically mispriced.

A stale share count can understate market capitalization.

A pence-to-pound error can distort valuation by a factor of one hundred.

A duplicated filing can double earnings.

A negative capital-expenditure sign can inflate free cash flow.

Once the error has been transformed into a clean ratio, the fake opportunity may look completely legitimate.

The raw record provides the evidence needed to trace the mistake.

This matters because extreme investment results should not be trusted automatically:

The most exciting output may be the strongest reason to inspect the inputs.

A correction should not erase the mistake

Suppose the system incorrectly mapped one company’s financial statements to another company’s stock price.

The error is found and fixed.

It may seem reasonable to overwrite the bad record and continue.

But keeping an audit trail is valuable.

The system should know:

• What was wrong

• When it was discovered

• Which records were affected

• Which features were compiled from them

• Which historical selections changed

• Which reports or backtests used the bad result

• How the issue was repaired

The goal is not to preserve errors as valid data.

It is to preserve the history of how the system learned.

Without that history, the same mistake may return under a different form.

Failed strategies need the old evidence

Suppose a strategy performs poorly.

To understand why, I need the exact data it used.

If the underlying database has since been cleaned, corrected, or replaced, rerunning the strategy may produce a different result.

The failed version can no longer be reproduced.

That creates a temptation to explain the failure using today’s improved information.

But the strategy made its decisions using yesterday’s evidence.

Raw snapshots allow the failure to be studied honestly.

Was the strategy wrong?

Was the data wrong?

Did the market change?

Those are different lessons.

They require the original decision environment.

Reproducibility requires frozen inputs

A backtest should not change every time it is rerun unless the researcher intentionally changes something.

If the rules remain frozen but the results move, the data may have changed underneath them.

This makes scientific comparison difficult.

Preserving raw versions allows the system to create reproducible datasets.

A strategy can be tested against:

• Raw-data version 1

• Cleaning policy version 2

• Feature definition version 4

• Strategy rules version 3

This may sound technical, but the principle is simple.

The result should have an identifiable recipe.

Otherwise, performance can drift without a clear cause.

A permanent warehouse supports a lean working database

Keeping raw data forever does not mean every active process must load everything.

The system can use two layers.

The permanent warehouse

This stores:

• Original provider responses

• Raw filings

• Historical versions

• Source metadata

• Native currencies

• Raw prices

• Corporate-action records

• Missing or contradictory values

• Provenance

It is comprehensive and durable.

The lean research database

This contains only the features needed for the current strategies and backtests.

It may include:

• Current nutrients

• Point-in-time snapshots

• Selected prices

• Outcome labels

• Eligibility flags

• Strategy-ready features

The lean database can be rebuilt.

The warehouse should not need to be.

This is like keeping a freezer full of raw ingredients while preparing only the meals currently on the menu.

The menu can change.

The ingredients remain available.

Rebuilding is safer than endlessly patching

A derived database can accumulate old assumptions.

Fields may be calculated under different rules.

Some rows may use an earlier currency policy.

Others may use updated provider mappings.

Over time, the database becomes a mixture of historical logic.

If the raw warehouse remains intact, the derived database can be rebuilt from a clean specification.

Every record can pass through the same current transformation rules.

This is often safer than repairing millions of derived values one by one.

The raw store makes the system disposable in the right place.

Derived features are replaceable.

Source evidence is not.

Storage is cheaper than recollection

Raw financial data can consume substantial disk space.

Keeping old versions may appear inefficient.

But the alternative can be much more expensive.

Recollecting the data later may involve:

• New provider fees

• API limits

• Missing historical records

• Changed provider policies

• Delisted companies that are no longer available

• Different results from updated sources

• Weeks or months of engineering work

Disk space is often cheaper than rebuilding lost history.

The more difficult the dataset was to collect, clean, and document, the more valuable preservation becomes.

Delisted companies make preservation especially important

Active-company databases are easier to maintain because the securities still exist.

Failed and delisted companies are more fragile.

Their pages may disappear.

Provider coverage may weaken.

Ticker symbols may be reused.

Corporate information may become harder to obtain.

Yet these companies are essential for honest historical research.

They represent the failures that real investors could have selected.

If their raw records are not preserved, the market’s history gradually becomes cleaner than reality.

The losers disappear.

The survivors remain.

A permanent warehouse protects against this slow creation of survivorship bias.

Missing data may become available later

A filing that is incomplete today may be repaired later.

A provider may add history.

A ticker mapping may be resolved.

An exchange may clarify a corporate action.

Keeping the original missing state still matters.

The system may need to know that the value was unavailable at the historical decision date.

Later availability should not be backdated into the past.

This creates two useful records:

• What is known now

• What was known then

The absence itself is part of the historical environment:

A later correction should improve current understanding without rewriting what earlier investors could have seen.

Raw data makes estimation visible

Sometimes missing values may need to be estimated.

An estimated filing-availability date may be necessary for a country where exact publication dates are unavailable.

A financial feature may be reconstructed from related fields.

These estimates can be useful.

They should never become indistinguishable from reported facts.

The warehouse should preserve:

• The original missing value

• The estimation policy

• The estimated result

• The confidence level

• The date the estimate was produced

This allows the strategy to treat exact and estimated evidence differently.

It also allows the estimate to be replaced if better information appears.

Machine learning increases the need for raw preservation

A future machine-learning model may consume hundreds of derived features.

If the model behaves strangely, debugging the final prediction alone will not be enough.

The feature may be wrong because:

• The source data was incorrect

• A transformation changed

• A missing value was handled differently

• The company was classified incorrectly

• A filing was attached to the wrong date

• Currency conversion failed

Machine learning adds distance between the original evidence and the final decision.

That makes the underlying chain more important, not less.

A powerful model trained on flawed history becomes a more efficient way to repeat the flaw.

Raw preservation protects against accidental leakage

Outcome data must remain separate from decision inputs.

A future return, bankruptcy label, or later restatement should not enter the historical feature set before it existed.

When raw sources, transformed features, and outcomes are kept in distinct layers, leakage is easier to detect.

The system can ask:

• Was this field available on the decision date?

• Was it derived from a later filing?

• Did a cleaning rule use knowledge of the outcome?

• Was a company excluded because it later failed?

Without preserved source dates and versions, these questions become much harder to answer.

The system should never depend on memory alone

A researcher may remember why a field was transformed a certain way.

Months later, the reason becomes less clear.

Years later, it may be forgotten entirely.

Documentation helps.

Raw records provide stronger evidence.

They show exactly what entered the pipeline.

The system should not depend on me remembering:

• Which provider was used

• Why a value was rejected

• Which currency unit appeared

• Whether a filing was restated

• Which mapping was uncertain

• Why one company disappeared from the screen

A durable project needs memory outside the creator’s head.

Permanent does not mean untouchable

Keeping raw data forever does not mean treating every incoming value as correct.

Raw records can be marked as:

• Valid

• Invalid

• Superseded

• Duplicate

• Contradictory

• Estimated

• Unresolved

The key is that corrections should add context rather than erase history.

The system can say:

This was the original provider response. It was later found to be wrong for this reason. This corrected value replaced it in derived calculations.

That is more informative than silently changing the record.

Data deletion can create invisible strategy changes

Suppose a screening rule requires five years of revenue history.

A company originally failed because two years were missing.

Later, the missing history is deleted from the record entirely, leaving only the three visible years.

The strategy may no longer know that the company failed due to missing information.

It may interpret the shorter history as the complete history.

Small data deletions can therefore alter selection logic without any change to the formal rules.

Preserving raw states prevents absence from being rewritten as completeness.

A warehouse creates accountability

When every score can be traced back to its source, the system cannot easily hide behind complexity.

A company was selected because certain values, rules, and mappings produced the decision.

If the result was wrong, the chain can be inspected.

This makes failures more useful.

The project can distinguish between:

• Data failure

• Transformation failure

• Classification failure

• Strategy failure

• Portfolio-construction failure

• Ordinary uncertainty

Without the chain, every failure becomes a vague disappointment.

With it, failure can become information.

The warehouse should outlive individual strategies

Most investment strategies will eventually change, fail, or become obsolete.

A durable financial history can support many generations of research.

One strategy may focus on value.

Another may study business quality.

Another may examine deterioration.

A later system may use classification models, regression, optimization, or shadow portfolios.

The underlying market history remains useful across all of them.

The warehouse is therefore broader than the current project stage.

It is infrastructure for questions that have not yet been invented.

The raw data is an asset

Companies often treat data as an operational by-product.

For this project, the historical warehouse is one of the most valuable assets being built.

It contains:

• Time

• Provider access

• Cleaning work

• Mapping decisions

• Historical availability

• Failed-company records

• Country-specific knowledge

• Lessons from errors

• The ability to reproduce earlier experiments

Deleting the raw evidence would throw away much of that accumulated work.

The derived score may be visible.

The real value lies underneath it.

The final principle

I do not keep raw financial data because every field will definitely be useful.

I keep it because I do not yet know which fields, versions, dates, or mistakes will become important later.

Strategies change.

Definitions improve.

Providers revise history.

Bugs are discovered.

New questions appear.

A permanent warehouse allows the system to return to the evidence and begin again without pretending that the earlier assumptions were facts.

The cleaned database helps the strategy operate.

The raw database helps the project remain honest.

One is designed for speed.

The other is designed for memory.

Financial history is difficult to reconstruct after it has been discarded.

Storage can be reorganized.

Derived features can be rebuilt.

Strategies can be replaced.

The original evidence deserves to survive them all.

That is why I keep raw financial data forever.

Subscribe now

Share

Share Fungal Intelligence for Stocks

Leave a comment

Its earnings may be attached to the wrong share price.

Its financial statements may use one currency while its market value uses another.

A stock split may have adjusted the price but not the share count.

A missing value may have been replaced with zero.

An outdated filing may be compared with today’s price.

When this happens, the investment opportunity exists only inside the database.

The numbers look precise.

The conclusion is false.

That is why financial data quality is not merely a technical concern. It is part of investment risk.

A ratio is only as reliable as its ingredients

Financial ratios often feel objective because they produce clean numbers.

A company may have:

• A price-to-earnings ratio of 6

• A price-to-book ratio of 0.5

• A free-cash-flow yield of 18%

• Revenue growth of 24%

• Debt-to-equity of 0.4

These figures can immediately attract attention.

But every ratio is built from underlying inputs.

A price-to-earnings ratio depends on:

• The correct stock price

• The correct earnings period

• The correct share count

• Matching currencies

• The correct company identity

• Proper treatment of unusual items

If one ingredient is wrong, the finished ratio can become meaningless.

A sophisticated formula cannot rescue contaminated inputs.

The fake bargain

Suppose a company appears to produce $100 million in annual earnings.

Its market capitalization appears to be $500 million.

The system calculates a price-to-earnings ratio of 5.

That looks extremely cheap.

But imagine that the market capitalization was calculated using an old share count from before the company issued a large number of new shares.

The true market capitalization might be $900 million.

The company was never trading at five times earnings.

The bargain was created by stale data.

An investor could spend hours researching the company, building a thesis, and becoming emotionally attached to an opportunity that disappeared as soon as the inputs were corrected.

Wrong ticker mappings can combine two different companies

Ticker symbols are not permanent global identities.

The same symbol can exist on multiple exchanges.

A ticker can change after a merger.

A delisted company’s symbol may later be reused.

A provider may use one suffix while another uses something different.

If the system maps a company’s financial statements to the wrong security, it can combine:

• One company’s earnings

• Another company’s price

• A third company’s currency

• An outdated share count

The resulting valuation may look spectacular.

It is also fictional.

This is one reason a database should preserve company identities, exchange information, provider symbols, and mapping confidence—not merely a short ticker symbol.

Currency mistakes can create enormous distortions

Global investing introduces another source of false opportunity.

Suppose a company reports revenue and earnings in Australian dollars, but its market value is accidentally treated as U.S. dollars.

Or a stock trading in British pence is interpreted as though it were trading in pounds.

A share priced at 250 pence should represent £2.50.

If the system interprets it as £250, the company appears one hundred times more expensive than it really is.

The reverse mistake can create an apparently unbelievable bargain.

Currency problems can affect:

• Market capitalization

• Earnings

• Cash flow

• Debt

• Book value

• Historical returns

• Portfolio performance

A ratio is valid only when its numerator and denominator speak the same financial language.

Foreign exchange changes the investor’s actual return

Even when company-level ratios are calculated correctly, portfolio returns can still be distorted.

A German stock may rise in euros while the euro falls against the investor’s home currency.

The local business performed well.

The investor’s converted return may be weaker.

A backtest that mixes native-currency prices with dollar-denominated portfolio values may report returns that no investor could have earned.

The system should distinguish between:

• Native prices for local trading behaviour

• Converted prices for portfolio truth

• The exchange rate available on each date

• The currency in which the financial statements were reported

Without that separation, international diversification can become mathematically inconsistent.

Stock splits can manufacture growth or collapse

Stock splits should not change the economic value of a company.

If one share becomes ten shares, the price per share should fall proportionally.

But databases do not always adjust every field consistently.

Suppose historical prices are split-adjusted, but historical share counts are not.

The system may conclude that:

• Market capitalization collapsed

• Earnings per share changed dramatically

• Share dilution occurred

• Valuation ratios improved overnight

None of those conclusions may be real.

The company simply changed the number of pieces into which ownership was divided.

Corporate actions must be treated consistently across price, share count, and per-share calculations.

Dilution disappears when share-count history is weak

A company can grow while repeatedly issuing new shares.

Revenue rises.

Total earnings rise.

The business becomes larger.

But each existing shareholder may own a smaller percentage of the company.

If the database contains only the latest share count, the system may miss years of dilution.

A company that increased total earnings by 50% while doubling its share count did not improve earnings per share.

The business grew.

The shareholder’s claim on the business weakened.

Bad share-count data can transform dilution into apparent growth.

That can make an expensive financing habit look like successful expansion.

Missing values can become imaginary strength

A blank field should mean that the system does not currently know the value.

But some pipelines replace missing values with zero.

This is dangerous.

Missing debt becomes zero debt.

Missing dilution becomes zero dilution.

Missing capital expenditures can make free cash flow appear stronger.

Missing losses can disappear from a growth calculation.

The system may then reward a company for information it never received.

Unknown is not a favourable result.

It is an unresolved state.

The investment process should recognize that boundary:

A company should not pass a test merely because the system was unable to perform it.

Duplicate filings can double the company’s apparent performance

Financial providers may store annual and quarterly records together.

A company may report the same period more than once after restating its results.

Two providers may deliver overlapping versions of the same filing.

If duplicates are not detected, the system may add values that should never have been combined.

Revenue can appear to double.

Cash flow can be counted twice.

Growth rates can jump unexpectedly.

The database may still contain legitimate-looking rows.

The error appears only when the records are interpreted as a time series.

Deduplication therefore requires more than removing identical rows.

The system must understand reporting periods, filing dates, statement types, and restatements.

Annual, quarterly, and trailing figures can be mixed incorrectly

A company’s latest annual earnings should not be compared directly with another company’s latest quarter as though they covered the same amount of time.

Yet this can happen when a database stores figures without clearly preserving their period.

One company may show twelve months of revenue.

Another may show three months.

A third may show a provider-calculated trailing figure.

The ranking system compares them as though they were equivalent.

The shorter-period company may look unusually small, unusually cheap, or unusually fast-growing.

Every financial value needs a clear period:

• Fiscal year

• Fiscal quarter

• Trailing twelve months

• Year-to-date

• Point-in-time balance

Without period discipline, comparisons become arithmetic without meaning.

Filing dates and period-end dates are different

A company may report results for the year ending December 31.

Investors might not receive those results until February or March.

A backtest that makes the December figures available on December 31 gives the strategy information from the future.

The numbers are historically correct.

Their availability date is wrong.

This creates a fake opportunity because the simulated investor can buy before the market had access to the evidence supporting the purchase.

Point-in-time research must distinguish between:

• The period the results describe

• The date the filing became public

• The date the system could reasonably have processed it

• The date the strategy acted

Without that chain, a backtest can become unintentionally clairvoyant.

Restated history can give the strategy knowledge investors lacked

Companies sometimes revise earlier financial statements.

Errors are corrected.

Accounting policies change.

Later filings may present a cleaner version of the past.

A modern database might replace the original historical value with the corrected one.

That is useful for understanding the company today.

It can be misleading for simulating an earlier investor.

The earlier investor saw the original filing, not the later correction.

A point-in-time database may therefore need to preserve both:

• What is now believed to be correct

• What was actually known at the historical decision date

Otherwise, the strategy learns from corrections that had not yet happened.

A beautiful backtest can be built on information unavailable in real life:

A historical test is only meaningful when its information was historically available.

Provider-defined metrics may hide inconsistent formulas

A data provider may offer a ready-made free-cash-flow field.

That seems convenient.

But how was it calculated?

One provider may subtract capital expenditures from operating cash flow.

Another may include acquisitions.

Another may use a trailing period.

Another may leave the field blank when one component is unavailable.

The same label can represent different formulas.

This can create inconsistent comparisons across companies or countries.

The safest approach is often to preserve the raw components and calculate the desired metric through a documented rule.

That does not eliminate all uncertainty.

It makes the assumption visible and reproducible.

Negative signs can reverse the conclusion

Cash-flow statements frequently represent outflows as negative numbers.

A capital-expenditure field may already be negative.

If the system calculates:

Operating cash flow minus capital expenditures

while capital expenditures are stored as a negative number, it effectively adds the expenditure.

Free cash flow becomes overstated.

A company spending heavily may appear to generate enormous excess cash.

This is a small technical mistake with a major investment consequence.

Sign conventions need to be understood rather than guessed.

Units can create thousandfold errors

Financial data may be reported in:

• Dollars

• Thousands of dollars

• Millions of dollars

• Pence

• Cents

• Local currency

• Converted currency

If a provider reports revenue as 500 because the value represents $500 million, another system may interpret it as $500.

If debt is stored in thousands while assets are stored in whole units, the leverage ratio becomes meaningless.

Unit errors often produce extreme ratios.

Those extreme ratios can rise to the top of a stock screen because they look exceptional.

The system may mistake a scaling error for a rare opportunity.

Extreme results should trigger validation, not immediate excitement.

Stale prices can create yesterday’s opportunity at today’s date

Small or foreign securities may trade infrequently.

A provider may show the last available price even when that price is several days old.

The financial database uses current information.

The price belongs to an earlier market.

If the stock moved sharply afterward, the valuation is no longer available.

A strategy can appear to buy at a price that existed only before the relevant information became public.

Price freshness matters.

The system should know:

• The trade date

• The market calendar

• Whether the market was open

• Whether the price was stale

• Whether enough volume traded

• Whether the investor could realistically transact

A number printed in a price field is not automatically an executable price.

Adjusted and unadjusted prices answer different questions

Unadjusted prices show what one share traded for at the time.

Adjusted prices account for events such as splits and dividends.

Both can be useful.

Problems appear when they are mixed.

A strategy may calculate entry prices using adjusted data but compare them with unadjusted exits.

Dividends may be counted twice.

A split may appear to create a large loss.

Historical returns become distorted.

The system should use a consistent price policy and document what the selected series represents.

Delisted companies can disappear from the market’s history

A modern data source may contain only companies that still trade today.

Failed, acquired, bankrupt, or delisted businesses may be missing.

A strategy tested on this population receives an easier market.

Many of the worst outcomes have already been removed.

The remaining companies appear healthier.

Financial ratios may seem more predictive because the population was filtered by survival before the strategy began.

This is not merely a missing-data problem.

It changes the investment universe.

A reliable historical system must remember the companies that failed, not only the companies that survived.

Incorrect classifications create false comparisons

A company can have correct financial data and still be evaluated incorrectly.

A bank may be treated as an ordinary operating company.

A real estate trust may be judged using standard net income.

A miner may be rewarded for unusually high profits at the peak of a commodity cycle.

The numerical inputs are real.

The interpretation is wrong.

This is why raw data and investment decisions should remain separate:

The database preserves the evidence.

The strategy decides which ruler belongs to each business.

Extreme values are often data problems before they are opportunities

When a company looks dramatically cheaper than everything around it, there are several possible explanations.

The market may be wrong.

The company may be extremely risky.

The business may be cyclical.

Or the data may be broken.

The correct first reaction is not excitement.

It is validation.

Questions should include:

• Does another provider show the same value?

• Does the original filing support it?

• Are the units correct?

• Are the currencies aligned?

• Is the price current?

• Is the share count accurate?

• Did a split or merger occur?

• Are the earnings annual or quarterly?

• Was a one-time gain included?

• Does the ratio remain attractive after correction?

A genuine bargain should survive inspection.

A fake one usually disappears when the pipeline is repaired.

Cheap screens are especially vulnerable

Value strategies search for unusual relationships between price and financial strength.

That makes them sensitive to data errors.

A growth screen may be distorted by one incorrect growth rate.

A value screen can place the most broken records at the very top because errors often create extreme cheapness.

Examples include:

• Missing debt interpreted as zero

• Market capitalization understated by stale share counts

• Earnings overstated by duplicated periods

• Foreign prices misread by a factor of 100

• One-time gains treated as recurring profit

• Negative capital expenditure handled with the wrong sign

A screen designed to find outliers will also find outlier errors.

The stronger the apparent bargain, the stronger the need for verification.

The screener should reject suspicious data

A stock screener should not merely apply financial thresholds.

It should also protect itself from questionable inputs.

A company might be quarantined when:

• Important fields disagree across sources

• Valuation ratios are economically impossible

• Prices are stale

• Currency is uncertain

• Share-count history is missing

• Filing periods overlap

• Corporate actions remain unresolved

• The company identity is ambiguous

This does not prove the company is unattractive.

It proves the system cannot yet evaluate it honestly.

The screener’s purpose is to reduce avoidable mistakes, not force every company into a ranking.

Rejecting unreliable evidence can be just as important as rejecting weak economics.

Data-quality scores should influence confidence

Suppose two companies receive the same investment score.

The first has:

• Ten years of consistent filings

• Verified prices

• Stable currency mappings

• Clear share-count history

• No unresolved contradictions

The second has:

• Missing periods

• Estimated values

• Uncertain ticker mappings

• Stale prices

• Several provider disagreements

The final numerical score may be identical.

The confidence behind it should not be.

A data-quality layer can help distinguish between:

• Strong conclusion from strong evidence

• Tentative conclusion from weak evidence

The second company may deserve investigation rather than immediate rejection.

It should not be treated as equally certain.

Provenance creates a chain of evidence

Every important value should carry context.

The system should know:

• Which provider supplied it

• Which filing it came from

• Which company identity it belongs to

• Which period it describes

• When it became available

• Which currency and unit it uses

• Whether it was reported or calculated

• Which transformations were applied

• Whether validation checks passed

This information is called provenance.

It allows a final investment score to be traced back to its ingredients.

Without provenance, correcting an error becomes difficult.

The system may know that a ratio is wrong without knowing where the problem entered.

Reproducibility is an investment feature

A trustworthy result should be reproducible.

Given the same data, rules, and date, the system should reach the same conclusion.

If it cannot, the historical evidence is unstable.

Reproducibility allows questions such as:

• Why did this company pass yesterday?

• Which filing changed the score?

• Did the company change, or did the pipeline change?

• Which strategies used the incorrect value?

• Did the correction alter historical selections?

• Does the original investment thesis still exist?

This is not merely good software engineering.

It helps prevent capital from being allocated based on conclusions that cannot be explained.

Validation should happen before ranking

The pipeline should not calculate every possible ratio and hope that the final score hides the bad inputs.

Validation should occur earlier.

A reasonable sequence is:

1. Confirm the company identity.

2. Confirm the exchange and trading currency.

3. Verify filing dates and reporting periods.

4. Normalize units.

5. Resolve splits and corporate actions.

6. Check share-count consistency.

7. Detect duplicates and contradictions.

8. Preserve missing values honestly.

9. Calculate strategy-specific features.

10. Rank only the companies with usable evidence.

This keeps bad inputs from becoming sophisticated-looking outputs.

Manual inspection still has a role

Automation is necessary when studying thousands of companies.

But unusual opportunities deserve manual verification.

A human can open the filing, inspect the price history, compare providers, and examine corporate actions.

The system’s job is not to eliminate judgment.

It is to direct judgment toward the cases where it matters most.

The most attractive candidates should face more scrutiny, not less.

A company that appears ordinary may pass through standard checks.

A company that appears almost impossibly cheap should be treated like a software result that fails a sanity test.

Bad data can also hide real opportunities

Data errors do not only create fake bargains.

They can conceal genuine ones.

A company may appear excessively expensive because:

• Its price was interpreted in the wrong units

• Earnings were missing

• A filing was mapped incorrectly

• A share split was mishandled

• Cash flow was attached to the wrong period

The screener rejects the company.

The opportunity disappears from the research queue.

This is why data quality should not be viewed only as protection from false positives.

It also protects against false negatives.

A better database does not guarantee better returns.

It creates a fairer opportunity set.

The investment system must distinguish three failures

When a company performs poorly, several things may have happened:

1. The data was wrong.
   The opportunity never existed as measured.

2. The interpretation was wrong.
   The data was correct, but the strategy misunderstood it.

3. The future changed.
   The evidence and reasoning were reasonable, but an uncertain outcome went badly.

These failures require different responses.

A data error requires pipeline repair.

An interpretation error requires strategy improvement.

An uncertain outcome may require no rule change at all.

Without clean separation, every loss can be blamed on the strategy—or excused as bad luck—without understanding the true cause.

The goal is not perfect data

No financial database will be perfect.

Companies report differently.

Providers make mistakes.

Currencies change.

Corporate actions create complications.

Historical records remain incomplete.

The correct objective is not flawless knowledge.

It is controlled uncertainty.

The system should know where evidence is strong, where it is weak, and where it is missing.

It should preserve the original records, document transformations, quarantine suspicious cases, and avoid pretending that every number deserves equal trust.

A fake opportunity is worse than no opportunity

An empty screen can feel frustrating.

A database full of apparently cheap stocks feels productive.

But activity is not the goal.

A fake investment opportunity consumes:

• Research time

• Attention

• Emotional energy

• Backtesting capacity

• Confidence

• Potentially real capital

It can influence future rules because the researcher tries to explain a pattern that came from an error.

It can even make a broken strategy look successful if the same bad data affects the backtest.

Sometimes the safest and most accurate conclusion is:

This company cannot yet be evaluated reliably.

That is better than manufacturing conviction from corrupted evidence.

The final lesson

Markets contain genuine mispricings.

Businesses do become misunderstood.

Strong companies are sometimes sold too cheaply.

But the most extraordinary bargain in a database may not be a market failure.

It may be a data failure.

Before asking why other investors missed the opportunity, the system should ask whether the opportunity survives basic verification.

Are the company, price, currency, period, share count, and financial statements aligned?

Were the values actually available at the time?

Are missing fields still missing, or were they turned into favourable assumptions?

Can the result be reproduced from the original evidence?

A reliable investment process does not begin by trusting the most exciting number.

It begins by trying to prove that the number is real.

Bad financial data creates fake investment opportunities.

Good data does not tell us what to buy.

It gives us a market that actually existed—and a fair chance to make an honest decision inside it.

Subscribe now

Share

Share Fungal Intelligence for Stocks

Leave a comment

They reveal how much revenue the business generated, whether it earned a profit, how much cash moved through the company, what it owns, and what it owes.

But financial statements are not the company itself.

They are a structured record of selected financial events.

They tell us what was measured under accounting rules.

They do not automatically tell us why it happened, whether it will continue, or what may happen next.

A business can look strong in its statements while becoming weaker underneath.

Another can look temporarily weak while building something valuable.

The numbers matter.

So do the limits of what those numbers can explain.

Financial statements describe the past

Every financial statement looks backward.

Even the newest quarterly report describes activity that has already happened.

Revenue was earned.

Expenses were recorded.

Cash was collected or spent.

Assets and liabilities existed on a particular reporting date.

Investors, however, receive returns from the future.

The financial statements provide evidence about the company’s history, but the investment result depends on what happens after the report.

A company with excellent historical results may face:

• New competition

• Changing customer preferences

• Technological disruption

• Higher interest rates

• Regulatory pressure

• Management turnover

• Economic recession

• A product failure

• A major legal problem

None of these needs to appear clearly in last year’s income statement.

Historical strength is useful evidence.

It is not a guarantee of future durability.

Revenue does not explain customer loyalty

A company may report growing revenue.

The financial statements show the amount.

They usually do not fully explain the relationship between the company and its customers.

Revenue could be growing because:

• More customers are buying

• Existing customers are spending more

• Prices were raised

• A competitor temporarily disappeared

• The company acquired another business

• Customers were offered aggressive discounts

• Sales were pulled forward from a later period

• The company entered a temporary boom

These causes have different implications.

Revenue created by loyal, recurring customers may be more durable than revenue created by heavy promotions.

Revenue gained through price increases may reflect strong pricing power—or customers may be preparing to leave.

The income statement reports the sale.

It does not completely reveal the strength of the relationship behind the sale.

Profit does not explain competitive advantage

A company can be highly profitable without possessing a durable moat.

It may benefit from favourable conditions that competitors will eventually attack.

High margins can attract new entrants.

A temporary shortage can increase pricing power.

A patent can approach expiration.

A popular product can be copied.

A retailer can occupy attractive locations until customer behaviour moves online.

The financial statements show the profit generated during the period.

They do not directly measure:

• Brand strength

• Customer switching costs

• Network effects

• Regulatory protection

• Distribution advantages

• Organizational culture

• Supplier relationships

• The difficulty of copying the product

These factors often determine whether the profit can continue.

A margin is a result.

A competitive advantage is one possible cause.

The numbers alone may not distinguish a durable cause from a temporary one.

Management quality does not fit neatly into a line item

Financial statements contain evidence about management.

Capital allocation decisions appear through acquisitions, debt, dividends, buybacks, and investment.

Operating discipline appears through costs, margins, and working capital.

But management quality cannot be reduced to one number.

A leadership team may produce strong short-term earnings by:

• Cutting necessary research

• Delaying maintenance

• Underinvesting in employees

• Reducing product quality

• Taking excessive financial risk

• Promoting sales that will later be returned

• Using optimistic accounting assumptions

The current statements may improve.

The future business may become weaker.

Good management often makes decisions that reduce current profit to protect long-term value.

Poor management can make the current quarter look excellent while creating problems that arrive later.

The financial statements contain clues.

They do not reveal every motive, trade-off, or internal decision.

Culture is mostly invisible

Culture influences how a company behaves when nobody is watching.

It affects whether employees report problems, whether managers challenge bad ideas, whether customer complaints are taken seriously, and whether leaders reward honesty or obedience.

A healthy culture can improve:

• Innovation

• Employee retention

• Product quality

• Risk control

• Customer service

• Internal communication

• Long-term decision-making

A toxic culture can remain hidden while financial performance still looks strong.

Problems may not become visible until employees leave, customers lose trust, regulators intervene, or a major failure occurs.

By the time the financial statements clearly reflect the damage, the cultural decline may have been developing for years.

Accounting categories do not capture every valuable asset

Some of a company’s most important assets may never appear clearly on the balance sheet.

These can include:

• Employee knowledge

• Internally developed software

• Brand reputation

• Customer relationships

• Proprietary processes

• Distribution networks

• Organizational learning

• Data

• Community trust

Accounting rules treat purchased assets differently from internally developed ones.

A company that buys another business may record goodwill and intangible assets.

A company that builds similar value internally may expense much of the cost as it occurs.

This can make two economically similar businesses look different on paper.

The balance sheet remains useful.

It is not a complete inventory of everything that makes the company valuable.

This is another reason the same ratios should not be used blindly across different business models:

The numbers need to be interpreted through the economic structure that produced them.

The statements may not reveal product quality

A company can report growing sales while product quality declines.

Customers may not notice immediately.

They may remain because changing suppliers is difficult.

Warranty claims may arrive later.

A damaged reputation may take years to affect revenue.

Alternatively, a company may invest heavily in improving its product before the benefit appears in the statements.

Financial reports can show research spending, returns, and warranty costs.

They cannot fully measure how customers experience the product.

Understanding product quality may require:

• Customer reviews

• Retention data

• Return rates

• Industry testing

• Distributor feedback

• Employee knowledge

• Product comparisons

• Direct use of the product

The financial statements provide one viewpoint.

The customer experiences another.

The balance sheet may hide the importance of people

Some businesses depend heavily on a small number of employees.

A research company may rely on a few scientists.

A software company may rely on key engineers.

An investment firm may rely on a small group of decision-makers.

A creative company may depend on particular designers or artists.

These people rarely appear as assets on the balance sheet.

Their salaries appear as expenses.

If they leave, the economic value lost may be much greater than the accounting value suggests.

A company can own buildings, equipment, and cash while losing the people who know how to use those resources effectively.

The balance sheet records what the company legally owns.

It does not completely capture what the company knows.

Reported risk is not the same as total risk

Companies disclose many risks in their filings.

These sections can be useful.

But they often contain broad language covering many possibilities.

The most important future threat may not yet be recognized.

Risk can come from:

• A new technology

• A changing regulation

• A concentrated customer base

• A fragile supplier

• A geopolitical conflict

• Cybersecurity weakness

• Environmental exposure

• A shift in public opinion

• An unexpected competitor

Some risks cannot be estimated precisely.

Others become obvious only after the company begins failing.

Financial statements measure known obligations and recognized losses.

They cannot assign a reliable number to every possible future disruption.

Customer concentration can look safe until it is not

A company may generate strong revenue and profit while depending on one major customer.

As long as the relationship continues, the financial statements look healthy.

If the customer leaves, demands lower prices, or experiences its own financial problems, the supplier may deteriorate quickly.

Concentration can exist in:

• Customers

• Suppliers

• Products

• Countries

• Distribution channels

• Financing sources

• Key employees

Some concentration is disclosed.

The strategic importance of the dependency may still require interpretation.

A company can be financially strong today while structurally dependent on one relationship.

Industry structure lives outside the statements

A company’s results cannot be understood without examining the environment in which it competes.

Two businesses may have identical margins and growth rates.

One operates in an industry with high barriers to entry and rational competitors.

The other operates in an industry where competitors constantly lower prices and copy one another.

The current numbers look similar.

The future economics may be very different.

Important external questions include:

• How easy is it to enter the industry?

• How powerful are customers?

• How powerful are suppliers?

• Are products becoming commodities?

• Is demand growing?

• Is the industry regulated?

• Is technology changing the economics?

• Are competitors behaving rationally?

The financial statements show how the company performed within the industry.

They do not fully explain the forces shaping the industry itself.

Cash flow cannot explain every use of cash

Cash-flow statements show where money moved.

They do not automatically tell us whether those uses were intelligent.

A company may spend billions on an acquisition.

The cash-flow statement records the payment.

It does not determine whether management overpaid.

A company may repurchase shares.

The statement records the cash used.

It does not decide whether the shares were undervalued or absurdly expensive.

A company may increase capital expenditures.

That could represent:

• Essential maintenance

• Wasteful expansion

• A valuable new factory

• A defensive response to competition

• A project that will never earn an adequate return

The cash flow is factual.

The quality of the capital-allocation decision requires judgment.

This is why my system separates evidence from interpretation:

The database records what happened.

The strategy must decide what that action may mean.

Debt numbers do not explain the full obligation

The balance sheet reports recognized debt and liabilities.

But a company can have additional economic obligations that require attention.

These may include:

• Long-term leases

• Pension commitments

• Purchase agreements

• Legal exposure

• Environmental cleanup

• Customer guarantees

• Supplier financing

• Contractual commitments

• Future capital requirements

Some appear in the notes rather than the main statements.

Others require estimates.

A company with moderate reported debt may still be financially constrained.

The headline number can be correct while the overall picture remains incomplete.

The notes can matter more than the headline numbers

The primary statements are only part of a financial report.

Important information often appears in the notes.

The notes may explain:

• Accounting-policy changes

• Debt maturities

• Customer concentration

• Related-party transactions

• Legal disputes

• Pension assumptions

• Acquisition details

• Revenue recognition

• Share-based compensation

• Segment performance

Two companies can report the same earnings while using different assumptions.

One may recognize revenue more aggressively.

Another may use more conservative estimates.

The headline figure does not reveal those differences by itself.

Reading only the summary numbers can create false confidence.

A clean number can come from an uncertain estimate

Many accounting values are not directly observed.

They are estimated.

Examples include:

• Loan-loss reserves

• Insurance liabilities

• Asset impairments

• Useful lives of equipment

• Warranty obligations

• Pension costs

• Bad-debt allowances

• Acquisition values

Estimates are necessary.

But small changes in assumptions can materially change reported profit or assets.

A number appearing to the nearest dollar may rest on uncertain predictions about the future.

Precision in presentation does not guarantee certainty in reality.

The financial statements cannot tell us the stock’s future return

A strong company is not automatically a strong investment.

Price matters.

A company can possess:

• High margins

• Excellent management

• Loyal customers

• Strong growth

• A durable moat

• A healthy balance sheet

and still produce disappointing returns if investors paid too much.

The stock price already contains expectations.

Future returns depend not only on business performance but also on how that performance compares with what the market expected.

Financial statements help estimate business value.

They cannot tell us exactly what price other investors will assign to the company later.

A weak company can sometimes be a successful stock

The opposite is also possible.

A troubled company can produce strong investment returns if expectations were extremely low and the business improves slightly.

This does not make financial quality irrelevant.

It shows that business quality and investment return are related but different concepts.

The financial statements help us understand the company.

Valuation helps us understand what investors are paying for that company.

Neither gives a certain answer.

Missing information creates another boundary

Sometimes the financial statements themselves are incomplete, inconsistent, delayed, or difficult to compare.

A company may have:

• Limited reporting history

• Missing cash-flow data

• Changed accounting standards

• Restated results

• Inconsistent segment reporting

• Currency complications

• Unclear share-count history

The correct response is not to invent certainty.

A missing number may indicate a provider problem, a reporting difference, or a genuine weakness in transparency.

That uncertainty is part of the evidence:

Knowing what we do not know is an important part of analysis.

Fraud can survive inside apparently strong statements

Financial statements are prepared by the company and reviewed under established processes.

Most companies report honestly.

But accounting systems can be manipulated.

Management may:

• Recognize revenue too early

• Hide obligations

• Misclassify expenses

• Inflate asset values

• Use related-party transactions

• Shift costs between periods

• Present nonstandard measures selectively

A screen based entirely on reported numbers may reward a company whose numbers are unreliable.

Detecting this can require examining inconsistencies, incentives, audit concerns, cash conversion, governance, and behaviour over time.

The statements remain essential.

Trust should not be automatic.

Screening can remove obvious weakness, not uncertainty

A stock screener can reject companies with severe debt, persistent losses, extreme dilution, or collapsing cash flow.

That is useful.

But passing a screen does not mean the company has been fully understood.

The screen cannot completely evaluate:

• Product relevance

• Management honesty

• Competitive dynamics

• Customer loyalty

• Cultural health

• Technological risk

• Future regulation

• Strategic adaptability

Passing means the company survived the first layer.

It does not mean the company is safe.

That is why a screener should behave as a defensive filter rather than an oracle:

The surviving companies still require deeper judgment.

Qualitative analysis is not an excuse for storytelling

Recognizing the limits of financial statements does not mean abandoning numbers and believing any attractive narrative.

Qualitative analysis can become dangerous when it is used to excuse weak evidence.

Investors can claim that:

• The brand is priceless

• Management is visionary

• The market opportunity is enormous

• Current losses do not matter

• Competition cannot catch up

• Traditional valuation no longer applies

Sometimes these claims contain truth.

They can also become ways to avoid discipline.

The solution is not to choose between numbers and stories.

It is to force them to challenge each other.

If management claims strong customer loyalty, retention and pricing should eventually provide evidence.

If a company claims a powerful moat, margins and returns on capital should show some benefit.

If heavy investment is creating value, future cash generation should improve.

Qualitative claims should lead to testable expectations.

Numbers and context should work together

The financial statements provide structure.

They allow companies to be compared and changes to be measured.

Context explains what those changes may mean.

A complete analysis might combine:

• Financial history

• Business-model classification

• Industry structure

• Customer behaviour

• Product quality

• Management incentives

• Capital allocation

• Competitive position

• Valuation

• Data quality

No single source gives the entire answer.

The objective is not perfect knowledge.

It is a decision built from several independent pieces of evidence.

The statements are a map, not the territory

A map is valuable because it simplifies reality.

It shows roads, boundaries, rivers, and distances.

But a map cannot show every tree, pothole, storm, or person.

Financial statements work similarly.

They compress a complex organization into a standardized form.

That compression makes analysis possible.

It also removes information.

A map should not be rejected because it is incomplete.

It should be used while remembering what it leaves out.

What the statements do well

Financial statements remain indispensable.

They help answer important questions:

• Is the company profitable?

• Does it generate cash?

• Is debt increasing?

• Are margins changing?

• Is the share count growing?

• How much capital does the business require?

• Are assets and liabilities changing?

• Is the company becoming financially stronger or weaker?

These questions provide a strong foundation.

The mistake is treating the foundation as the entire building.

What must be added

After studying the numbers, the next questions should include:

• Why are the results changing?

• How durable are the causes?

• What does the company depend on?

• What could make customers leave?

• What could competitors copy?

• Is management allocating capital intelligently?

• Which risks are absent from the main statements?

• What expectations are already reflected in the price?

• What evidence would prove the investment thesis wrong?

These questions turn accounting information into business analysis.

The limit is part of the lesson

Financial statements cannot tell us everything.

That is not a flaw that can be completely repaired with a larger spreadsheet.

Businesses are living systems made of people, relationships, incentives, technology, resources, and competition.

Not every important feature can be reduced to a clean financial field.

The goal of an investment system should not be to pretend uncertainty has disappeared.

It should use financial data where financial data is strong, preserve uncertainty where it is not, and prevent confident conclusions from outrunning the evidence.

Financial statements tell us what the company has reported.

They help us understand what the business has been.

They provide clues about what it may become.

But they cannot see the future, judge management’s character, experience the product, measure every competitive threat, or determine the correct price with certainty.

They are the beginning of understanding.

They are not the end.

Subscribe now

Share

Share Fungal Intelligence for Stocks

Leave a comment

The database should record what happened.

The strategy should decide what that information might mean.

That distinction sounds simple, but many investment systems blur the two layers together.

They collect only the information required by the current strategy. They calculate ratios using hidden assumptions. They discard companies that do not fit the model. They overwrite missing values. They store conclusions as though they were raw facts.

This may make the system easier to build at first.

It also makes the system harder to trust, test, repair, and improve.

That is why my stock system separates data from decisions.

The database should preserve evidence

A database should behave like a financial memory.

It should store evidence such as:

• Revenue

• Net income

• Operating cash flow

• Capital expenditures

• Debt

• Assets

• Share count

• Filing dates

• Reporting periods

• Trading prices

• Currencies

• Company classifications

• Data sources

• Missing or invalid values

These facts should be preserved as accurately as possible.

The database should not decide that a company is attractive.

It should not decide that debt is too high.

It should not decide that growth is strong enough.

Those are strategy decisions.

The database’s job is to remember what was reported, when it became available, where it came from, and what transformations were applied.

A strategy is an interpretation layer

Once the evidence is stored, a strategy can interpret it.

For example, the strategy might calculate:

• Free-cash-flow yield

• Revenue growth

• Debt-to-equity

• Return on invested capital

• Share-count growth

• Profit deterioration

• Valuation percentiles

• Financial-quality scores

It can then apply rules.

Perhaps a company must generate positive cash flow.

Perhaps debt cannot exceed a certain level.

Perhaps valuation must remain below a practical limit.

Perhaps share dilution cannot be excessive.

These are not universal financial truths.

They are decisions made for a particular investment approach.

Another strategy may interpret the same underlying company differently.

That flexibility is possible only when the evidence is kept separate from the opinion.

Facts and conclusions are not interchangeable

Suppose a provider reports that a company has $500 million in debt.

That is a data point.

Whether the debt is dangerous depends on context.

A company producing predictable cash flow may support it comfortably.

A cyclical company with falling earnings may be at serious risk.

Similarly, a price-to-earnings ratio of 10 is a calculation.

Calling the company cheap is a conclusion.

The low ratio might represent an attractive valuation.

It might also reflect collapsing profits, legal trouble, or a temporary commodity peak.

The same number can support different interpretations.

A trustworthy system should preserve the distinction.

Why this matters for different business models

A bank, miner, software company, utility, and retailer do not use capital in the same way.

They do not produce the same financial statements for the same economic reasons.

A debt ratio that looks dangerous for one company may be ordinary for another.

A price-to-book ratio may be meaningful for a bank but much less useful for a software company whose most important assets are intangible.

The database can store the financial records for all these businesses.

The strategy layer decides which comparisons are valid.

That is why classification must happen before applying investment rules:

The same evidence can feed several specialized models without rewriting the historical record.

Raw data should survive strategy changes

Investment strategies evolve.

A metric that seems useful today may later prove weak.

A rule may be too strict.

A new business category may need to be added.

A better definition of free cash flow may become available.

If the database contains only the strategy’s finished scores, changing the strategy may require rebuilding everything.

Worse, the original information may already have been discarded.

Preserving the raw financial history allows the system to recalculate features later.

The strategy can change without rewriting the past.

This is one reason I am building a fundamental database rather than depending entirely on a finished stock screener:

A provider remains part of the supply chain.

But the strategy should not be trapped inside the provider’s definitions.

The database should not be optimized for one conclusion

Imagine building a database specifically for a value strategy.

You store earnings, book value, and price.

You ignore customer retention, revenue quality, dilution, cash conversion, and balance-sheet deterioration because the current strategy does not use them.

Later, you discover that the cheapest companies frequently have weakening fundamentals.

You want to test whether deterioration can help distinguish genuine bargains from value traps.

But the required history was never stored.

The original database was shaped around the answer you expected to find.

That is dangerous.

A good database should preserve enough evidence to test competing ideas.

It should not be designed to prove one preferred strategy correct.

Transformation should be visible

Raw financial data often needs to be cleaned or transformed.

Currencies may need conversion.

Stock prices may need adjustment for splits.

Minor currency units may need correction.

Duplicate filings may need resolution.

Trailing values may need calculation.

Free cash flow may be derived from operating cash flow and capital expenditures.

These transformations are necessary.

They should also be traceable.

The system should know:

• The original value

• The source

• The reporting currency

• The conversion rate

• The formula used

• The date the transformation occurred

• Whether the value was reported, calculated, or estimated

Without this information, a calculated feature can appear more certain than the evidence underneath it.

Missing data belongs to the evidence layer

Suppose a company has no usable share-count history.

The database should not quietly assume that dilution was zero.

Suppose debt information is missing.

The system should not automatically treat the company as debt-free.

Unknown is not the same as zero.

A missing value may reflect:

• An absent filing

• A provider problem

• An incorrect ticker mapping

• A reporting-format difference

• A new company with limited history

• A genuine disclosure weakness

The strategy can decide how to respond.

It may reject the company.

It may reduce confidence.

It may request manual investigation.

It may place the company into a temporary quarantine.

But the database should preserve the uncertainty rather than inventing an answer.

This is why missing values should remain visible:

The absence of evidence is part of the financial record.

Scores should be reproducible

Suppose the system gives a company a quality score of 82.

That score should not appear from nowhere.

It should be possible to trace it backward.

Which financial values were used?

Which reporting date?

Which formulas?

Which percentile comparisons?

Which weights?

Which missing-data rules?

Which strategy version?

If the same inputs and rules are used again, the system should produce the same result.

This is essential for debugging.

If a company suddenly moves from the top of the ranking to the bottom, I need to know whether:

• The financial condition changed

• A new filing arrived

• The company was reclassified

• A provider corrected an error

• The scoring rules changed

• The database pipeline broke

Without separation between the layers, these causes become difficult to distinguish.

Historical truth must remain stable

Suppose I test a strategy from 2015 to 2020.

Later, I improve the scoring model.

The historical financial evidence should not change merely because the strategy changed.

The same filings, prices, and availability dates should remain in place.

Only the interpretation layer should differ.

This creates fair comparisons.

Strategy A and Strategy B can be tested using the same historical world.

If each strategy quietly uses a different version of history, their results cannot be compared honestly.

The database should provide a stable environment.

The strategies should compete inside it.

One database can support many animals

My long-term system is based on the idea that different types of companies require different financial diets.

General operating companies may focus on valuation, quality, growth, safety, dilution, deterioration, timing, and data quality.

Banks may need measures involving deposits, loan quality, capital adequacy, and credit losses.

Miners may require reserve quality, production costs, commodity sensitivity, and capital intensity.

The database stores the available evidence for each company.

Each specialized model selects the nutrients relevant to its business type.

This prevents one universal scoring formula from forcing every company into the same comparison.

The evidence remains shared.

The interpretation becomes specialized.

The screener should not alter the source data

A stock screener may reject companies with negative free cash flow.

That does not mean those companies should be deleted from the database.

They may be useful for:

• Testing another strategy

• Studying business failures

• Measuring recovery rates

• Building outcome labels

• Evaluating sector cycles

• Improving future screening rules

The screener is allowed to say no.

The database should still remember the company.

Deleting rejected companies would make the historical record cleaner than the real market.

It would also introduce survivorship and selection bias.

The research system needs access to both successes and failures.

A rejected company can become useful evidence

Suppose the mouth rejects a company because of severe dilution.

Two years later, the company performs extremely well.

That does not automatically mean the filter was wrong.

Perhaps avoiding dilution sacrifices some upside in exchange for lower risk.

Perhaps that company was an unusual exception.

To evaluate the rule, I need to retain the rejected company and its later outcome.

The database records what happened.

The strategy can then be judged on the trade-offs created by its rules.

Without the rejected population, I cannot measure false negatives.

I see only what the strategy selected.

That would make the strategy appear more accurate than it really was.

Outcome data should also remain separate

A future return is not a feature available at the time of selection.

It is an outcome.

The system may eventually label each historical candidate with:

• Three-month return

• Six-month return

• Twelve-month return

• Market-relative alpha

• Maximum drawdown

• Disaster or failure labels

• Continuation or survival labels

These outcomes can be used to evaluate strategies or train later models.

But they must never leak into the decision inputs for the same historical date.

Separating features from outcomes prevents the model from learning the future.

This is similar to machine learning.

Training data and labels must be aligned correctly.

A small leak can make a model appear extraordinarily intelligent while destroying its real predictive value.

The strategy should be versioned

Investment rules will change.

That is normal.

But every meaningful change should create a new strategy version.

The system should record:

• The filters used

• The feature definitions

• The scoring weights

• The classification rules

• The portfolio constraints

• The effective date

• The reason for the change

This prevents the strategy from slowly changing while keeping the same name.

Otherwise, a historical performance record may combine several different systems.

A versioned strategy allows honest comparison.

Version 1 may fail because debt limits were too permissive.

Version 2 may improve the rule.

Both histories remain visible.

Failure becomes part of the learning process instead of something quietly erased.

The portfolio allocator is another decision layer

Stock selection and portfolio allocation should also remain separate.

The scoring system may identify several attractive companies.

The portfolio layer decides which combination to own and how much capital each receives.

It considers questions such as:

• Country exposure

• Industry concentration

• Shared economic risks

• Liquidity

• Position size

• Issuer overlap

• Portfolio balance

A company may receive an excellent score and still be excluded because the portfolio already contains similar exposure.

The database does not make that decision.

The company-ranking system does not need to make that decision either.

Each layer has its own job.

This makes the final result easier to understand and test.

Clean boundaries improve debugging

In programming, systems are easier to maintain when components have clear responsibilities.

A database stores information.

A feature compiler transforms that information.

A screener rejects unsuitable candidates.

A scoring model ranks survivors.

A portfolio allocator combines selected holdings.

A shadow portfolio tests the frozen process forward.

If something goes wrong, the error can be located.

Was the filing missing?

Was the currency conversion wrong?

Was the ratio calculated incorrectly?

Was the screening threshold inappropriate?

Was the portfolio too concentrated?

When all these responsibilities are mixed together, one mistake can spread through the entire system without being visible.

Separation creates an audit trail.

Clean boundaries also reduce overfitting

Overfitting does not occur only in scoring weights.

It can enter through the data pipeline.

A researcher may:

• Remove difficult companies

• Fill missing values in a favourable way

• Choose the best provider after viewing outcomes

• Redefine metrics until a backtest improves

• Exclude periods that produce weak results

• Change company classifications after seeing performance

These choices can make the historical test look stronger.

By separating the evidence layer from the strategy layer, changes become more visible.

A data correction should be justified by accuracy.

A strategy change should be justified by investment logic.

The two should not be blended together merely because the backtest improved.

The database should be boring

A good database does not need an opinion.

It does not need to be exciting.

It should be:

• Durable

• Traceable

• Consistent

• Reproducible

• Honest about uncertainty

• Flexible enough to support future research

The intelligence belongs in how the evidence is interpreted.

Trying to make the database itself “smart” can cause it to quietly hide information that does not fit the current model.

The database should preserve reality even when reality is inconvenient.

Decisions should remain reversible

Suppose I later discover that one valuation metric was poorly defined.

If the raw evidence remains intact, I can repair the calculation and rerun the strategy.

Suppose I discover that a business was classified incorrectly.

I can move it into the proper comparison group and recalculate its scores.

Suppose a provider value was wrong.

I can correct the record and identify every decision affected by it.

This reversibility is valuable.

A system that stores only final scores cannot easily recover from mistaken assumptions.

Once the evidence is compressed into a conclusion, important context may be lost.

The strategy should earn authority

A clean database does not guarantee a successful investment strategy.

It simply prevents the strategy from receiving artificial help from a distorted foundation.

The strategy must still prove that its rules are useful.

It should face:

• Historical testing

• Sealed validation

• Different countries

• Different market regimes

• Realistic costs

• Frozen shadow portfolios

• Limited real-money deployment

The database gives the strategy a fair test.

It does not guarantee a favourable verdict.

That is exactly what I want.

What the architecture looks like

The system can be understood as a sequence:

1. Raw evidence
   Preserve filings, prices, currencies, company identities, and metadata.

2. Clean transformations
   Normalize units, align dates, calculate reproducible financial features, and preserve provenance.

3. Business classification
   Determine which economic model is appropriate for each company.

4. The mouth
   Reject companies that fail the strategy’s non-negotiable requirements.

5. Scoring
   Compare the survivors using relevant financial nutrients.

6. Portfolio allocation
   Build a resilient collection while controlling concentration and position size.

7. Outcome tracking
   Record what happened after each historical or live decision.

8. Validation
   Test whether the frozen process survives outside the environment in which it was built.

Each layer receives information from the previous one.

Each layer has a different responsibility.

The database records the world; the strategy expresses a belief

Financial data is never completely objective.

Accounting choices, provider definitions, and transformations all introduce judgment.

But the goal should still be to preserve the evidence before adding more interpretation.

The database attempts to answer:

What did the company report, and what could an investor have known at that time?

The strategy asks:

Given that evidence, does this company fit the rules and objectives of this investment approach?

Those questions should not be collapsed into one.

The first describes the world.

The second expresses a belief about the world.

Why the separation matters

Separating data from decisions gives the system several advantages:

• Strategies can change without rewriting history

• Multiple business models can use the same evidence differently

• Missing values remain visible

• Calculations can be reproduced

• Errors can be traced

• Failed companies remain part of the test

• Competing strategies can use the same historical foundation

• Outcome labels can be kept away from decision inputs

• Portfolio mistakes can be distinguished from stock-selection mistakes

This does not make the project simpler.

It makes the complexity organized.

The final principle

An investment system should not hide its opinions inside its facts.

It should know which parts came from the company, which parts came from the data provider, which parts were calculated, and which parts were decisions made by the strategy.

That separation creates accountability.

When the system succeeds, I can investigate why.

When it fails, I can locate the weakness.

When the strategy changes, the historical evidence remains available.

The database should remember.

The strategy should interpret.

The screener should reject.

The scoring model should compare.

The portfolio allocator should combine.

And none of those layers should pretend to be the others.

That is why my stock system separates data from decisions.

Subscribe now

Share

Share Fungal Intelligence for Stocks

Leave a comment

Enter a few financial conditions, press a button, and receive a list of investment ideas.

That makes the screener feel like a machine designed to find winners.

I think its first job should be different.

A stock screener should eliminate businesses that do not deserve deeper attention.

It should reject far more companies than it selects.

The goal is not to predict the future perfectly.

The goal is to prevent obviously unsuitable companies from consuming research time—or entering the portfolio by accident.

A screener is a filter, not an oracle

A screener can compare thousands of companies much faster than a human analyst.

But speed does not create understanding.

A company may appear cheap because its earnings are temporarily inflated.

Another may appear profitable because an important expense was excluded.

A fast-growing business may be funding that growth through heavy dilution.

A stock with a high dividend yield may be approaching a dividend cut.

The screener sees numbers.

It does not automatically understand the story that produced them.

That is why I do not want it to declare:

This is a good investment.

I want it to say:

This company has survived the first layer of rejection and deserves further examination.

That is a much more realistic responsibility.

Rejection is valuable

Investors often focus on missed opportunities.

What if the screener rejects a stock that later rises dramatically?

That will happen.

Any meaningful filter will exclude some future winners.

But a system should not be judged only by how many successful stocks it discovers.

It should also be judged by how many avoidable disasters it prevents.

Rejecting a company with:

• Unsustainable debt

• Persistent losses

• Severe dilution

• Weak liquidity

• Collapsing revenue

• Negative cash generation

• Unreliable data

• An unsuitable business model

may protect the portfolio from a loss that would require years to recover.

Avoiding one catastrophic mistake can matter more than finding several moderate winners.

The first question is not “How high can it go?”

When investors become excited about a company, they often begin with potential.

How large could the market become?

How quickly could revenue grow?

What happens if margins expand?

What if the company becomes the industry leader?

These questions can be useful later.

The first screening question should be more defensive:

What could make this company unacceptable before I spend time imagining the upside?

This changes the order of analysis.

Before forecasting success, the system looks for reasons the company may not survive, may not create shareholder value, or may not be measurable with enough confidence.

The upside remains uncertain.

Some major risks are already visible.

My screener has a mouth

In the biological metaphor behind my project, each company is like a tree producing financial fruit.

Before those nutrients enter the deeper scoring system, they pass through a mouth.

The mouth does not decide which company will produce the highest return.

It decides what the system is willing to consume.

Rotten, dangerous, unsuitable, or unrecognizable inputs should be rejected before they reach the more complicated layers.

This is similar to a security boundary in programming.

A system should validate inputs before allowing them into the core process.

Once bad data or unsuitable candidates move deeper into the pipeline, they become harder to detect and more expensive to remove.

Classification must happen before rejection

A universal screener can reject good companies for the wrong reasons.

A bank may appear highly leveraged because deposits are liabilities.

A utility may carry more debt because it owns expensive infrastructure.

A software company may look costly relative to book value because much of its value comes from intangible assets.

A miner’s earnings may swing dramatically with commodity prices.

The screening rules must match the type of company being evaluated.

This is why classification comes before filtering:

A company should be rejected because it fails an appropriate test—not because it was measured using rules designed for a different economic species.

Different species need different mouths

A general operating company may need positive free cash flow, manageable debt, stable profitability, and reasonable valuation.

A bank may need adequate capital, reliable deposits, controlled loan losses, and disciplined lending.

A miner may need competitive production costs, sufficient reserves, manageable obligations, and protection against commodity downturns.

A biotechnology company may require enough cash to reach important research milestones.

The concept of rejection stays the same.

The reason for rejection changes with the business model.

One universal mouth would either become too permissive or reject entire categories unfairly.

Hard filters protect the research process

Some screening conditions should behave like hard gates.

If the company fails, it does not proceed.

For a general operating business, examples might include:

• Market capitalization below a practical minimum

• Trading liquidity too low for realistic execution

• Negative free cash flow

• Negative net income

• Excessive debt

• Severe revenue deterioration

• Rapid share dilution

• A valuation beyond the strategy’s limits

These conditions do not prove that a company is bad.

They indicate that the company does not fit this particular strategy.

That distinction matters.

A tiny company with negative cash flow may become enormously successful.

It may also belong in a venture-capital-style strategy rather than a conservative public-equity screen.

Rejection is not a universal judgment.

It is a boundary.

A boundary keeps the strategy honest

Without hard boundaries, almost any company can be justified.

High debt can be called an investment in growth.

Negative cash flow can be called temporary.

Dilution can be called strategic financing.

A high valuation can be defended by a large future market.

Falling revenue can be described as a transition year.

Sometimes these explanations are valid.

But a strategy that creates an exception for every attractive story no longer has rules.

The screener should make it difficult to move unsuitable companies into deeper analysis simply because I like the narrative.

Rules are most valuable when they disagree with me.

Liquidity should be part of selection

A company may look attractive on paper but be difficult to trade.

Low-liquidity stocks can have:

• Wide bid-ask spreads

• Large price movements from small orders

• Stale prices

• Limited buyer interest

• Difficult exits during stress

• Backtests based on prices that were not realistically available

A screen that ignores liquidity can identify opportunities that exist only inside a database.

The final portfolio needs securities that can actually be purchased and sold.

This is especially important when examining small companies or foreign exchanges.

An investment system should not confuse theoretical availability with practical investability.

Broker access matters too

A stock can trade publicly and still be unavailable through the investor’s broker.

The security may be listed on an unsupported exchange.

It may require special permissions.

It may trade through an instrument the investor cannot hold.

A final investability check should therefore happen before selection.

The database can continue storing and studying the company.

The portfolio should not select something it cannot buy.

A screening system that recommends inaccessible securities is not finished.

Positive earnings are not enough

A profitable company can still be weak.

Accounting earnings may be supported by:

• Aggressive revenue recognition

• One-time gains

• Low-quality receivables

• Asset sales

• Temporary commodity prices

• Reduced investment

• Non-cash adjustments

Cash flow provides another perspective.

If reported earnings repeatedly fail to produce cash, the system should become suspicious.

This does not mean cash flow is automatically perfect.

Working-capital changes can distort short periods.

Capital-intensive businesses require interpretation.

But requiring evidence that profits eventually become cash can eliminate many fragile candidates.

Free cash flow can also mislead

Even a positive free-cash-flow number needs context.

A company may improve free cash flow by delaying necessary maintenance.

Another may temporarily invest heavily in a valuable expansion.

A rigid filter can confuse underinvestment with efficiency.

This is one reason the screener should not make the final decision.

Its job is to identify obvious problems and establish a minimum standard.

Companies that pass still require analysis.

The filter reduces the research population.

It does not replace research.

Debt should be tested against the business

Debt becomes dangerous when the company cannot support it.

The raw debt number matters less than the relationship among:

• Interest expense

• Cash flow

• Earnings stability

• Asset quality

• Debt maturity

• Refinancing needs

• Economic cyclicality

A business with stable recurring revenue may support obligations that would be reckless for a cyclical company.

Still, the screener needs a practical limit.

If a company already depends on favourable conditions just to service its debt, it may not belong in a strategy designed for resilience.

The future is uncertain enough without beginning from a fragile capital structure.

Dilution is an invisible cost

A company can grow while each shareholder owns less of it.

Management may issue new shares to fund acquisitions, cover operating losses, compensate employees, or reduce debt.

Issuing shares is not always harmful.

It can create value when the capital is invested productively.

But persistent dilution can absorb much of the benefit from business growth.

Suppose revenue grows by 10% while the share count grows by 12%.

The company became larger.

Existing shareholders may not have become wealthier.

A screener focused only on company-level growth can miss this.

The system should care about growth per share, not growth at any cost.

Deterioration deserves hard attention

Many investors are attracted to stocks after prices decline.

The lower valuation creates the appearance of opportunity.

But falling prices can reflect deteriorating economics.

A defensive screener should look for sharp declines in:

• Revenue

• Free cash flow

• Net income

• Margins

• Balance-sheet strength

• Customer activity

• Shareholder ownership

A single weak period does not always justify rejection.

Businesses experience temporary setbacks.

But extreme deterioration should not be ignored merely because the stock has become cheaper.

A low valuation is useful only if the underlying business remains strong enough to recover.

Growth should be a defense, not merely a reward

In my general operating framework, I do not need every company to be growing rapidly.

I do want evidence that the business is not collapsing.

Revenue, free cash flow, and net income provide different views.

One may decline temporarily while the others remain stable.

A reasonable filter can require that at least some of these measures remain healthy.

The purpose is not to chase the fastest-growing company.

It is to avoid businesses whose financial foundation is deteriorating across several dimensions at once.

Growth becomes a survival check before it becomes a scoring advantage.

Valuation belongs in the mouth

A wonderful business can still be a poor investment at an unreasonable price.

The screener should therefore reject companies whose valuations require too much optimism.

For example, a general operating strategy might impose limits on:

• Price to free cash flow

• Price to book value

• Enterprise value relative to cash generation

• Earnings yield

• Other model-appropriate valuation measures

The exact ratios should depend on the company type.

The principle is broader:

The system should not allow excitement about business quality to erase price discipline.

Future returns depend partly on what is paid.

Cheapness should not overpower weakness

Value screens can produce lists filled with distressed companies.

Low valuations often appear after:

• Earnings collapse

• Heavy dilution

• Excessive borrowing

• Industry decline

• Governance problems

• Customer losses

• Legal or regulatory trouble

Some of these companies recover.

Others are value traps.

A rejection-first screener should require evidence of basic health before allowing valuation to become attractive.

Cheapness earns attention only after the company demonstrates that it still has something worth buying.

Missing data should not receive a free pass

A company cannot honestly pass a test that the system was unable to perform.

Suppose debt data is missing.

The screener should not assume debt is low.

Suppose share-count history is incomplete.

The system should not assume there was no dilution.

Missing evidence should create a separate status:

• Pass

• Fail

• Unknown

• Requires investigation

Unknown is not the same as failure.

But it is also not permission to proceed as though the company passed.

This is why missingness must remain visible:

The system should understand the boundary between evidence and assumption.

A smaller research list is a feature

A screen that begins with thousands of stocks may return only a small number of candidates.

That can feel disappointing.

It may also mean the system is working.

The point of screening is to concentrate limited attention.

Research time is scarce.

Every hour spent studying a company with obvious financial problems is an hour not spent understanding a stronger candidate.

A strict screener improves the quality of the research queue.

It does not need to keep the analyst constantly busy with new ideas.

Selectivity reduces narrative temptation

The more companies an investor examines, the more persuasive stories they encounter.

Every business has a possible future.

A struggling retailer can reinvent itself.

A miner can benefit from rising commodity prices.

A biotechnology company can make a breakthrough.

A heavily indebted company can refinance.

A declining manufacturer can restructure.

Possibility is unlimited.

Capital is not.

A rejection-first screener prevents every possible turnaround from becoming an investment candidate.

It narrows the field to companies that already meet a minimum standard.

The screener should be explainable

Every rejection should have a reason.

The system should record whether a company was excluded because of:

• Business classification

• Market size

• Liquidity

• Profitability

• Cash flow

• Debt

• Dilution

• Deterioration

• Valuation

• Missing data

• Broker access

This makes the screening process auditable.

It also allows the strategy to improve.

If thousands of companies are rejected because one data field is missing, that may reveal a database problem rather than a market truth.

A filter without diagnostics can hide its own failures.

A database and a screener have different jobs

A database should preserve the financial evidence as faithfully as possible.

A screener applies strategy-specific rules to that evidence.

Mixing those layers creates problems.

If the database stores only the fields currently needed by one strategy, future research becomes limited.

If the screener silently modifies data to make more companies pass, historical truth becomes distorted.

This separation is a central reason I am building the underlying infrastructure myself:

The database remembers.

The screener rejects.

The scoring system compares.

The portfolio allocator decides what to own.

Each layer should have a clear responsibility.

Passing is not an endorsement

When a company survives the mouth, it has not been declared safe.

It has simply avoided the known reasons for immediate rejection.

The company can still fail because of:

• Competitive pressure

• Weak management

• Fraud

• Technological disruption

• Regulation

• Economic recession

• Unexpected liabilities

• A poor future that historical numbers could not predict

A screening system cannot eliminate uncertainty.

It can only improve the starting population.

The surviving companies move to deeper analysis, where their strengths, risks, and valuation can be compared properly.

Scoring should happen after survival

Once unsuitable companies are removed, the system can rank the survivors.

It may compare:

• Quality

• Valuation

• Growth

• Safety

• Cash generation

• Deterioration

• Dilution

• Timing

• Data quality

This order matters.

If scoring occurs before rejection, an excellent result in one category may compensate for a fatal weakness in another.

A company with extreme debt could remain highly ranked because its growth score is strong.

A deeply unprofitable business could survive because momentum is excellent.

Some weaknesses should not be averaged away.

The mouth defines what is unacceptable.

The scoring system compares companies that have already demonstrated basic fitness.

The threshold should be frozen

Screening rules should not change every time an attractive company narrowly fails.

That is how discipline erodes.

Suppose a company has a price-to-free-cash-flow ratio of 27 while the strategy’s limit is 25.

It may still be an excellent investment.

But changing the rule to 28 solely to admit that company turns the threshold into a suggestion.

A rule can be revised when there is broad evidence that it is poorly designed.

The revised version should then be documented and tested.

It should not be changed as a favour to one story.

False negatives are unavoidable

A strict screener will reject companies that later perform extremely well.

That does not automatically make the rules wrong.

Every strategy chooses which kinds of uncertainty it is willing to accept.

A conservative system may miss speculative turnarounds, early-stage growth businesses, and highly leveraged recoveries.

That is the cost of focusing on companies with stronger current evidence.

The correct comparison is not between the system and perfect hindsight.

It is between the system and the realistic alternatives available before the future was known.

A missed winner is visible.

An avoided disaster is often forgotten.

Both belong in the evaluation.

The screen should match the portfolio’s purpose

A retirement portfolio, a venture portfolio, and a deep-value turnaround portfolio should not use the same mouth.

My system is being designed around:

• Financially functioning companies

• Reasonable valuations

• Manageable risk

• Global diversification

• A small equal-weighted portfolio

• Long-term survival

• No leverage

• No real money until the process earns trust

The screening rules should reflect that objective.

A company can be interesting and still be unsuitable.

The system does not need to consume every kind of opportunity.

Rejection prevents complexity from hiding mistakes

Advanced models can create the temptation to score everything.

If machine learning produces a probability for every stock, it can feel wasteful to exclude any company.

But a sophisticated prediction does not make an unacceptable candidate appropriate.

A model may identify a highly leveraged, illiquid, deteriorating company as having strong upside potential.

That may even be statistically true.

The mouth can still say no.

Constraints exist because models are imperfect.

The deeper the system becomes, the more valuable clear boundaries may be.

The best screen is not the largest screen

A screener that returns hundreds of companies may feel productive.

It may simply be permissive.

The quality of a screen should not be measured by the number of results.

It should be measured by whether the surviving population matches the strategy’s actual requirements.

A small list can create better decisions if every candidate deserves serious attention.

An enormous list transfers the screening work back to the human.

That defeats the purpose.

What the mouth is really protecting

The mouth does more than protect capital.

It protects the integrity of the entire research system.

It prevents:

• Weak businesses from contaminating comparisons

• Unreliable data from becoming precise scores

• Exciting stories from bypassing rules

• Illiquid securities from creating fictional backtests

• One attractive metric from hiding a fatal weakness

• Advanced models from optimizing unsuitable candidates

The mouth creates a clean boundary between the investable research universe and everything else.

Selection should be earned

A stock should not enter deeper analysis merely because it exists in the database.

It should earn attention by surviving a sequence of appropriate tests.

It should earn a score by providing enough trustworthy evidence.

It should earn a portfolio position by improving the system as a whole.

And it should earn real capital only after the strategy itself survives validation.

This creates a ladder of increasing trust.

The process begins with rejection.

The job of the screener

A screener does not need to identify the next extraordinary winner.

It needs to reduce a chaotic market into a population that can be studied honestly.

It should protect the investor from obvious fragility, unsuitable businesses, untradeable securities, misleading data, and prices that require too much hope.

Most companies should not pass.

That is not pessimism.

It is selectivity.

The market contains thousands of possibilities, but a portfolio needs only a few carefully chosen positions.

A good screener does not ask how many stocks it can include.

It asks how much evidence a company should provide before it is allowed any further into the system.

Its first success is not finding the winner.

Its first success is saying no.

Subscribe now

Share

Share Fungal Intelligence for Stocks

Leave a comment

Price-to-earnings.

Debt-to-equity.

Free-cash-flow yield.

Return on assets.

Revenue growth.

These measurements can be useful.

But they can also become misleading when the companies being compared make money in completely different ways.

A bank, a software company, a miner, a retailer, and a utility may all appear in the same stock database.

That does not mean the same financial ruler should be used for each one.

A company can look weak simply because it is being measured incorrectly.

This idea begins with a broader principle I use when designing my investment system:

The same number can mean different things

Suppose two companies have large amounts of debt.

The first is a regulated utility.

The second is a highly cyclical retailer.

The debt ratio may look similar.

The risk is not.

A utility may generate relatively stable revenue from essential services. Its infrastructure is expensive, so long-term financing can be a normal part of the business model.

A retailer may depend on discretionary consumer spending, changing tastes, and thin profit margins. The same amount of debt may become dangerous much faster during a recession.

The number is identical.

The surrounding system changes its meaning.

Financial ratios do not interpret themselves.

Banks are built differently

Debt is a simple example, but banks make the problem even clearer.

For most operating companies, debt is money the business owes.

For banks, deposits are also liabilities.

Yet deposits are not merely a burden. They are a central source of funding that allows the bank to make loans and earn interest.

A traditional debt-to-equity comparison can therefore become confusing when banks are mixed with ordinary businesses.

Banks also use assets differently.

A manufacturer owns factories, equipment, inventory, and cash.

A bank’s assets are largely financial claims, including loans and securities.

The quality of those assets depends on whether borrowers repay.

A bank should therefore be evaluated using questions such as:

• Are its loans likely to be repaid?

• Does it have enough capital to absorb losses?

• Are deposits stable?

• Is management taking excessive credit risk?

• How sensitive is the bank to interest rates?

• Are loan losses rising?

A bank with a low price-to-book ratio may be attractive.

It may also be signalling that the market distrusts the value of its assets.

The ratio alone cannot decide which interpretation is correct.

Software businesses can look expensive for structural reasons

A software company may appear expensive using traditional asset-based measurements.

It might own few factories, little inventory, and limited physical infrastructure.

Much of its value may come from:

• Code

• Brand

• Customer relationships

• Network effects

• Intellectual property

• Employee knowledge

• Recurring subscriptions

Many of these assets do not appear cleanly on the balance sheet.

A low price-to-book ratio may therefore be a poor way to judge the business.

Software companies can also have unusual cost structures.

Creating the first version of a product may be expensive.

Serving an additional customer may cost very little.

That can allow margins to expand as the company grows.

A manufacturer cannot usually double production without buying more materials, hiring more workers, or expanding capacity.

A software business may sometimes grow much faster without increasing costs at the same rate.

This does not mean software companies deserve any valuation.

It means the business model changes which measurements deserve attention.

Retailers live through inventory

Inventory is essential for a retailer.

Without products on shelves or available online, there is nothing to sell.

But inventory also creates risk.

Products can become outdated.

Demand can weaken.

Discounting can destroy margins.

Cash can become trapped in unsold goods.

A rising inventory balance might mean management is preparing for growth.

It might also mean customers are no longer buying.

The number needs context.

An inventory increase should be compared with:

• Revenue growth

• Sales trends

• Product mix

• Gross margins

• Seasonal patterns

• Management commentary

A retailer with growing inventory and falling sales may be in trouble.

A fast-growing retailer opening new locations may show similar inventory growth for a completely different reason.

Miners depend on what is underground

A mining company may report strong earnings during a period of high commodity prices.

That does not necessarily mean the business has become permanently stronger.

Its economics may depend on:

• The market price of the commodity

• The quality of the resource

• Extraction costs

• Remaining reserve life

• Energy prices

• Local politics

• Environmental obligations

• The amount of capital needed to maintain production

A low price-to-earnings ratio can be especially deceptive.

Commodity producers often appear cheapest near the top of a cycle because current earnings are unusually high.

They may appear expensive near the bottom because profits have temporarily collapsed.

The ratio can encourage investors to buy when earnings are least sustainable and avoid the company when the cycle is already weak.

For a miner, normalized commodity assumptions and production costs may matter more than one year of reported profit.

Utilities trade growth for stability

Utilities are often slower-growing than technology or consumer companies.

That does not automatically make them inferior businesses.

They frequently operate in regulated markets, provide essential services, and generate relatively predictable revenue.

Their strengths may include:

• Stable demand

• Long-lived assets

• Regulated returns

• High barriers to entry

• Reliable cash generation

Their weaknesses may include:

• Heavy debt

• Large capital requirements

• Regulatory limits

• Interest-rate sensitivity

• Slow growth

Comparing a utility’s revenue growth directly with a software company’s growth would miss the point.

The software company may offer greater upside.

The utility may offer greater stability.

They perform different roles.

A race car and a cargo ship are both forms of transportation.

Comparing only their speed ignores what each one was built to do.

Insurers sell promises about the future

An insurer receives premiums today in exchange for agreeing to cover possible future losses.

That creates a business model built around uncertainty.

A company may appear profitable because it collected more premiums than it paid in claims.

But the quality of those profits depends on whether management estimated future losses accurately.

Important questions include:

• Is the company pricing risk correctly?

• Are claims becoming more expensive?

• Are reserves sufficient?

• Is the investment portfolio safe?

• Is management growing too quickly?

• Are old policies producing unexpected losses?

Revenue growth alone may not be positive.

An insurer can grow rapidly by underpricing risk.

The business looks stronger until the claims arrive.

Growth without underwriting discipline can destroy value.

Real estate businesses are not ordinary operating companies

Real estate investment trusts often distribute a large portion of their income to investors.

Traditional net income can be distorted by depreciation charges on buildings that may maintain or increase their market value.

This is why investors often examine measures such as funds from operations instead.

A normal operating-company screen may reject a healthy real estate business because the wrong earnings measure was used.

Real estate companies also depend heavily on:

• Property types

• Occupancy

• Tenant quality

• Lease duration

• Interest rates

• Refinancing needs

• Local supply and demand

An office landlord, apartment owner, warehouse operator, and data-centre owner may all be classified as real estate.

Their risks can still be very different.

Even within one business category, further classification may be necessary.

Growth does not mean the same thing everywhere

Revenue growth is often treated as universally positive.

But growth can come from different sources.

A company may grow by:

• Selling more products

• Raising prices

• Buying another company

• Opening new locations

• Entering new countries

• Taking more financial risk

• Accepting lower-quality customers

• Issuing large amounts of stock

These forms of growth are not equally valuable.

A bank can grow loans by weakening lending standards.

A retailer can grow revenue by opening unprofitable stores.

A miner can increase production by spending enormous amounts of capital.

A software company can increase customers while offering discounts that make each customer unprofitable.

The correct question is not simply:

Is the company growing?

It is:

What resources are being consumed to create that growth, and is the growth economically valuable?

Profit margins depend on the model

A supermarket may operate successfully with very low profit margins.

It sells large volumes of essential goods and turns inventory quickly.

A software company with the same margin might be considered weak because its product should theoretically have much better economics.

A luxury brand may maintain high margins because customers value its identity and scarcity.

A manufacturer may operate with moderate margins while requiring significant investment in factories and equipment.

Margins cannot be judged without understanding:

• Capital intensity

• Pricing power

• Customer behaviour

• Competition

• Inventory requirements

• Operating leverage

• Industry norms

A low margin is not automatically bad.

A high margin is not automatically safe.

High margins can attract competitors or depend on temporary shortages.

Capital intensity changes the meaning of cash flow

Two companies may report the same operating cash flow.

One must spend nearly all of it maintaining factories and equipment.

The other requires little physical investment.

Their economic outcomes are very different.

This is why free cash flow matters.

But even free cash flow needs interpretation.

A business may temporarily reduce capital spending and produce attractive cash flow while neglecting essential maintenance.

Another may invest heavily because it has unusually valuable opportunities.

Low capital spending can signal efficiency.

It can also signal underinvestment.

High capital spending can destroy value.

It can also build a powerful future advantage.

The business model determines what level of reinvestment is necessary.

One universal score can reward the wrong companies

Imagine a stock-ranking system that rewards:

• Low debt

• High free cash flow

• High return on assets

• Fast growth

• Low price-to-book value

This might work reasonably well for some ordinary operating companies.

But it could unfairly punish:

• Banks, because liabilities are central to the model

• Utilities, because infrastructure requires debt

• Software companies, because intangible assets are missing from book value

• Miners, because profits fluctuate with commodity cycles

• Real estate companies, because accounting depreciation distorts earnings

The ranking may look objective.

It is still based on subjective choices about which metrics matter.

A precise formula can produce meaningless comparisons when its categories are poorly designed.

This is also why I want control over the evidence underneath my final scores instead of relying entirely on the conclusions produced by someone else’s screener:

This is a classification problem

Before comparing two companies, the system should decide whether they belong in the same comparison group.

That means asking:

• How does the company make money?

• What resources does it depend on?

• What risks can destroy it?

• Which accounting measures describe it accurately?

• What does normal financial structure look like for this business?

• Which other companies are genuinely comparable?

Classification reduces the chance of comparing businesses that follow different economic rules.

It does not eliminate complexity.

A diversified industrial company may operate across several business models.

A technology company may also provide financing.

A retailer may own significant real estate.

Companies do not always fit perfectly into one category.

But an imperfect classification is still often better than pretending classification does not matter.

Comparable companies should share economic logic

Companies do not need to be identical to be compared.

They should share enough of the same economic structure that the comparison remains meaningful.

Two regional banks can be compared using deposit quality, credit losses, capital, and lending profitability.

Two software companies can be compared using retention, margins, recurring revenue, and growth efficiency.

Two miners can be compared using costs, reserves, production, and commodity exposure.

This creates a more useful question.

Instead of asking:

Which company has the best ratios in the entire market?

The system asks:

Which company appears strongest among businesses that survive in similar ways?

That is a much fairer contest.

Different models create different data gaps

Business models also affect what information is available and how consistently it is reported.

A young software company may have little operating history.

A miner may report reserve information that has no direct equivalent in another industry.

A foreign company may use a different reporting structure.

An insurer may revise its estimates as claims develop over time.

A blank field does not always mean that the business has a value of zero or failed a financial test.

Sometimes it means the system does not yet have enough evidence to make a valid comparison.

That uncertainty should remain visible:

There is still value in broad measurements

Some financial ideas remain useful across many industries.

A company that consistently destroys cash deserves attention.

Extreme dilution can harm investors regardless of the business model.

Dishonest reporting is dangerous everywhere.

Unmanageable obligations can destroy almost any company.

Paying an unreasonable price can weaken any investment.

The goal is not to abandon common principles.

It is to apply them intelligently.

The same high-level concern may require different measurements for different businesses.

Financial strength matters everywhere.

What financial strength looks like depends on the company.

Specialized rules make the system more honest

A specialized investment system may need separate models for:

• General operating companies

• Banks

• Insurers

• Real estate businesses

• Commodity producers

• Utilities

• Biotechnology companies

• Funds and other financial vehicles

Each category can have its own:

• Required data

• Financial filters

• Scoring metrics

• Risk warnings

• Valuation methods

• Portfolio constraints

This creates more work.

But the alternative is a simpler system that produces clean answers to the wrong question.

Complexity should not be added merely to look sophisticated.

It should be added where the underlying reality genuinely differs.

The wrong ruler creates false confidence

A company may rank poorly because it has been misunderstood.

Another may rank highly because the formula accidentally rewards its accounting structure.

Both errors create false confidence.

The investor believes the comparison is scientific because every company received a numerical score.

But numbers cannot repair a category mistake.

Measuring a fish by its ability to climb a tree does not produce an objective ranking of animals.

It produces a precise measurement of the wrong capability.

Understand the engine before reading the dashboard

Financial statements are like the dashboard of a machine.

They display speed, fuel use, temperature, and pressure.

But those readings mean different things depending on the machine.

A high engine temperature might be normal in one system and dangerous in another.

A large fuel tank might represent strength for one vehicle and unnecessary weight for another.

Before interpreting the dashboard, we need to understand the engine.

Companies are the same.

Ratios summarize what happened.

The business model explains why it happened and whether it is sustainable.

The real goal of comparison

The purpose of comparing companies is not to force every business onto one universal leaderboard.

It is to improve decisions.

A useful comparison should help answer:

• Which business is financially stronger?

• Which risks are being compensated?

• Which company uses resources more effectively?

• Which valuation assumptions are reasonable?

• Which weaknesses are temporary?

• Which strengths belong to the business model itself?

Those questions require context.

Without context, comparison becomes decoration.

Different businesses need different rulers

Markets contain many kinds of companies because the economy contains many kinds of work.

Banks allocate credit.

Utilities provide infrastructure.

Miners extract resources.

Retailers distribute goods.

Software companies organize information.

Insurers absorb risk.

Each business transforms resources into value through a different process.

That process determines which numbers matter most.

The problem is not that financial ratios are useless.

The problem is using them without first asking what kind of business produced them.

A good measurement begins with the right ruler.

A good comparison begins with the right category.

And a good investment system should understand the business model before deciding whether the company looks strong, weak, cheap, or expensive.

Subscribe now

Share

Share Fungal Intelligence for Stocks

Leave a comment

Trusting it with real money should be difficult.

A convincing backtest, an intelligent-looking ranking, or a few successful stock picks are not enough. A system can look brilliant while benefiting from bad data, hidden assumptions, overfitting, or luck.

That is why I want clear rules separating research from deployment.

The system should not receive capital simply because I am eager to use it.

It should earn that responsibility.

Rule 1: The data foundation must be reliable

Every decision begins with data.

Before I trust the system, I need reasonable confidence that it understands:

• Which company each record belongs to

• When each financial statement became available

• Which currency the figures use

• Whether prices are adjusted correctly

• Whether missing values are recognized

• Whether duplicated or contradictory records exist

• Whether a stock could actually have been purchased

Perfect data is impossible.

But known limitations are different from invisible errors.

The system must record uncertainty rather than quietly transforming incomplete information into precise-looking scores.

Rule 2: Historical tests must use historical information

A strategy tested in the past should only receive information that existed at that time.

It should not use later restatements, corrected classifications, future company membership, or financial data published after the simulated decision date.

Otherwise, the backtest describes an investor with access to the future.

That investor never existed.

Point-in-time discipline is therefore not an optional technical improvement. It determines whether the test represents a possible strategy or a fictional one.

Rule 3: Failed companies must remain visible

A historical market contains more than today’s survivors.

Companies fail, delist, merge, become uninvestable, or disappear from modern databases.

Removing those businesses makes the past look safer than it was.

The system should face the companies that investors actually encountered at the time, including the failures.

A strategy that succeeds only after dead companies are removed has not earned trust.

It has been protected from reality.

Rule 4: The rules must be frozen before validation

A strategy cannot be honestly tested while its rules continue changing in response to the results.

Once a version enters validation, its important settings should be recorded and frozen:

• Financial filters

• Metric definitions

• Scoring weights

• Rebalancing rules

• Position sizes

• Portfolio constraints

• Exit conditions

• Data sources

If I change the rules, I have created a new strategy.

That new version can begin a new test, but it should not inherit the credibility of the old one.

Rule 5: The best backtest does not automatically win

Testing many strategies will always produce a top performer.

That does not prove the winner found a durable advantage.

It may simply be the version that matched historical noise most closely.

I want to examine whether success appears across a family of similar strategies.

If one exact combination produces extraordinary results while nearby combinations collapse, that precision is suspicious.

A real relationship should usually tolerate small changes.

The market probably does not care whether one scoring weight is exactly 17% instead of 16%.

Rule 6: Performance must come from more than one lucky event

A strategy can appear successful because of one stock, one industry, one country, or one extraordinary period.

I want to know where the returns came from.

Did many holdings contribute?

Did the system work across several market environments?

Would the result remain respectable without its largest winner?

Did the strategy depend on one technology boom, commodity cycle, or period of falling interest rates?

Concentrated luck should not be mistaken for repeatable skill.

Rule 7: The strategy must survive unseen data

The system should face data that played no role in its creation.

This could include:

• Earlier sealed periods

• Later market periods

• Different countries

• Different economic regimes

• Companies outside the original development population

The purpose is not to demand identical performance everywhere.

Different environments should produce different outcomes.

The purpose is to learn whether the strategy’s basic logic survives outside the laboratory where it was designed.

Rule 8: Realistic costs and execution must be included

A theoretical opportunity is not necessarily an investable opportunity.

Before real money is involved, the system should account for:

• Bid-ask spreads

• Trading fees

• Currency conversion

• Taxes where relevant

• Illiquid stocks

• Delayed data

• Market holidays

• Unavailable securities

• Prices that move before execution

A strategy that trades constantly or depends on tiny advantages may disappear once real friction is included.

The system should be judged as something that could actually operate, not as a frictionless equation.

Rule 9: Every selected stock must be investable

A database may contain companies that my broker does not support.

Some securities may trade only on inaccessible exchanges. Others may have poor liquidity, unusual settlement rules, or restrictions that make them impractical.

The system should verify investability before a company enters the final portfolio.

It is not enough for the opportunity to exist inside a database.

The investor must be able to purchase, hold, and eventually sell it under realistic conditions.

Rule 10: Portfolio risk must be controlled separately

Selecting attractive companies does not automatically create a safe portfolio.

The final holdings may still share the same hidden dependency.

Several companies might rely on:

• One commodity

• One country

• Cheap credit

• Consumer confidence

• High valuation expectations

• The same customer group

• The same economic cycle

The portfolio therefore needs its own rules.

My initial structure is likely to remain simple:

• Roughly eight to ten companies

• Approximately equal position sizes

• No duplicate issuer exposure

• Exposure across several countries

• Limits on shared economic risks

• No single stock capable of determining survival

The goal is not to eliminate losses.

It is to prevent one error from destroying the experiment.

Rule 11: No leverage

Borrowed money magnifies both returns and mistakes.

An unproven system should not receive extra power before it has demonstrated basic reliability.

Leverage can turn an ordinary drawdown into forced selling. It can make timing more important and reduce the amount of time available for a strategy to recover.

The system should first prove that it can survive with ordinary capital.

Adding leverage before trust exists would be like increasing the speed of a machine before confirming that its brakes work.

Rule 12: Position sizing should remain humble

A ranking may identify which companies appear more attractive.

It does not necessarily tell me how much more capital the first-ranked company deserves than the fifth-ranked company.

Small scoring differences may come from noise, reporting timing, or arbitrary modelling choices.

Until the system demonstrates genuine sizing skill, approximately equal weighting is more honest.

Equal weighting says:

I may be able to identify a useful group without knowing precisely which member will perform best.

That is a reasonable starting assumption.

Rule 13: Drawdowns must be survivable

A strategy can produce strong long-term returns while suffering losses that would be psychologically or financially impossible to tolerate.

A 60% decline followed by an eventual recovery may look acceptable when compressed into a historical chart.

Living through it without knowing the ending is different.

Before deployment, I need to understand:

• The size of historical drawdowns

• How long recoveries took

• What caused the losses

• Whether several holdings failed together

• Whether the strategy remained investable

• Whether I could realistically continue following it

A system that cannot be followed during bad periods is not a usable system.

Rule 14: The strategy must enter a shadow portfolio

Before receiving money, a strategy should operate forward in time with frozen rules.

The shadow portfolio should record every decision as though it were real:

• Selected companies

• Entry dates

• Available prices

• Position sizes

• Rebalancing actions

• Data failures

• Trading limitations

• Later outcomes

The future should arrive without allowing the system to rewrite its past decisions.

This tests more than the strategy.

It tests the database, execution process, portfolio rules, and my own willingness to avoid interference.

Rule 15: Longevity matters more than a short burst of performance

A shadow portfolio that performs extremely well for three months may simply be lucky.

I am more interested in which strategy family continues functioning under frozen rules for the longest clean period.

Trust should come from survival across changing conditions.

Return still matters.

But the highest short-term return should not automatically defeat a strategy with broader, steadier evidence.

I want a system that can survive many seasons, not one that happened to thrive during a perfect summer.

Rule 16: Failures must remain documented

A weak strategy should not disappear from the record.

Its failure may reveal:

• An overfit rule

• A broken data assumption

• Hidden concentration

• Excessive turnover

• Poor risk control

• A regime dependency

• An execution problem

Keeping failed versions creates institutional memory.

Without that memory, I could accidentally rebuild the same mistake under a new name.

The project should learn from failure rather than hide it.

Rule 17: Real-money deployment must begin small

Passing every research stage would not justify immediately committing a large amount of capital.

Real money introduces pressures that simulations cannot fully reproduce.

My behaviour may change.

Execution may be worse than expected.

Taxes and spreads may matter more.

Unexpected technical problems may appear.

The first deployment should therefore be deliberately limited.

The goal would not be to maximize profit.

It would be to confirm that the complete system works outside research.

Capital should increase only after additional evidence accumulates.

Rule 18: The system must remain understandable

Complexity is not automatically intelligence.

If I cannot explain why the system selected a company, rejected another, or changed a position, I should be cautious about trusting it.

Machine learning and optimization may eventually contribute.

But their output should remain surrounded by constraints, diagnostics, and understandable evidence.

I do not need every relationship to be simple.

I do need enough transparency to detect when the system is behaving irrationally.

A black box should not gain authority merely because its backtest looks impressive.

Rule 19: The system must know when it does not know

Sometimes there will not be enough reliable information.

A company may have incomplete filings.

A market regime may be unfamiliar.

The available candidates may share too much risk.

Prices may be stale.

The system should be permitted to say:

No decision is justified right now.

That may mean rejecting a company, reducing confidence, delaying a rebalance, or holding cash.

A system forced to produce an answer will eventually manufacture certainty.

Honest uncertainty is safer than false precision.

Rule 20: Trust must be earned gradually

There should be no single moment when the system suddenly becomes “proven.”

Trust should accumulate through stages:

1. Reliable data

2. Reproducible calculations

3. Honest historical testing

4. Sealed validation

5. Frozen shadow portfolios

6. Limited real-money deployment

7. Continued survival without constant repair

Each stage answers a different question.

A successful backtest is only one piece of that chain.

What would disqualify the system?

The system should lose permission to manage money if:

• Results cannot be reproduced

• Data errors materially change selections

• Performance depends on hidden future information

• Rules are repeatedly changed after losses

• One stock or theme explains most returns

• Realistic costs eliminate the advantage

• Drawdowns exceed acceptable limits

• The shadow portfolio behaves differently from the research

• The process cannot be followed consistently

• I no longer understand why decisions are being made

Stopping is not failure.

Continuing after the evidence breaks would be failure.

The final rule

The system should never receive money because I need it to work.

That emotional pressure would make honest evaluation nearly impossible.

I would begin defending the strategy instead of testing it.

The project should face the same standard I would apply to someone else asking for my capital.

Show the evidence.

Show the failures.

Show the assumptions.

Show the live record.

Then begin carefully.

A strategy does not earn trust by producing the most exciting promise.

It earns trust by surviving every reasonable attempt to prove it wrong.

Until then, it belongs in research—not in control of real money.

Subscribe now

Share

Share Fungal Intelligence for Stocks

Leave a comment

A shadow portfolio can tell me how that strategy behaves after I stop changing it.

That difference matters.

Backtests are useful for developing ideas, comparing rules, and rejecting weak systems.

But they also create temptation.

When a result looks disappointing, I can adjust a threshold, change a weight, remove a metric, or alter the testing period.

Every adjustment may appear reasonable.

Together, they can gradually shape the strategy around the past.

A shadow portfolio removes that freedom.

The rules are frozen.

The future arrives.

The strategy must live with its decisions.

A backtest happens inside the laboratory

A backtest is a controlled experiment.

I choose the data.

I choose the rules.

I choose the time period.

I choose which results to examine.

This is useful because it allows rapid testing.

A weak idea can fail in minutes instead of consuming years of real-world observation.

But the laboratory has a major weakness:

The researcher already knows the history.

Even when I try to be objective, I know which years contained crashes, recoveries, bubbles, and unusual market conditions.

That knowledge can influence design.

I may create rules that appear general but quietly respond to events I already know happened.

The strategy can become adapted to the historical environment in which it was built.

A shadow portfolio enters an unknown environment

A shadow portfolio follows the strategy using current information and real market prices, but without risking actual money.

The system makes selections as though the portfolio were real.

It records:

• What would have been purchased

• When it would have been purchased

• The price available at the time

• The position size

• The reason for selection

• The rules active on that date

• The later outcome

The important part is that the future is not yet visible.

There is no opportunity to redesign the strategy around what happens next.

The test becomes genuinely forward-looking.

Frozen rules create honest evidence

Suppose a strategy selects several companies.

A few months later, those companies perform poorly.

It would be easy to explain the failure after the fact.

Perhaps one metric should have received less weight.

Perhaps the valuation limit was too generous.

Perhaps the market regime should have been classified differently.

Some of those explanations may be correct.

But immediately changing the rules destroys the clean test.

The original strategy never gets the chance to reveal what it really is.

A shadow portfolio needs frozen rules for a defined period.

That means no repairs simply because recent performance is uncomfortable.

The strategy should be allowed to succeed or fail under the exact conditions it originally accepted.

Shadow portfolios test behaviour, not just return

Return matters, but it is not the only useful observation.

A strategy can make money while behaving badly.

It may take excessive risk.

It may repeatedly select illiquid companies.

It may depend on one industry.

It may produce large drawdowns before recovering.

It may turn over the portfolio constantly.

It may become impossible to follow during market stress.

A shadow portfolio can reveal these operational weaknesses.

I can observe questions such as:

• Are the selections actually tradable?

• Are the prices realistic?

• Does the system concentrate without noticing?

• Does it keep changing its opinion?

• Are the signals stable between updates?

• Does the portfolio behave sensibly during stress?

• Would a real investor be able to follow it?

These details may be difficult to appreciate in a clean historical summary.

Real timing exposes hidden assumptions

Backtests often simplify timing.

They may assume that financial statements become available instantly.

They may buy at a convenient closing price.

They may rebalance exactly when intended.

They may assume that every data source updates reliably.

In real time, these assumptions face friction.

A filing can arrive late.

A provider can fail.

A price can be stale.

A stock may be unavailable through the investor’s broker.

A currency conversion may be missing.

The market may move before the strategy can act.

Shadow portfolios reveal whether the system works as an actual process rather than merely as a mathematical idea.

That is critical.

A strategy is not useful if it cannot be executed.

The strategy can no longer select its test

Historical research offers many possible periods.

If one period looks bad, it is tempting to emphasize another.

A shadow portfolio receives whatever market arrives next.

It cannot choose an easier decade.

It cannot avoid an unexpected recession.

It cannot request lower interest rates.

It cannot remove a difficult quarter.

The sequence is imposed from outside.

That makes the evidence harder to manipulate.

The strategy must survive the conditions it is given.

Shadow testing reveals emotional pressure

A backtest compresses years into seconds.

A chart may show a major decline followed by a recovery.

The recovery makes the decline look temporary and manageable.

Living through that decline is different.

During a shadow portfolio, the ending is unknown.

A position can fall for weeks or months.

The strategy may look obviously wrong.

Other investments may outperform.

The desire to interfere becomes strong.

This is valuable information.

A system that looks easy to follow historically may be psychologically difficult in real time.

Shadow testing exposes the gap between theoretical discipline and actual discipline.

It can reveal whether signals are stable

A strategy may perform well historically while producing unstable rankings.

Small data updates might cause companies to jump rapidly from the top of the list to the bottom.

One revised filing could completely change the portfolio.

That instability creates practical problems.

It may lead to excessive trading, transaction costs, and inconsistent decisions.

A shadow portfolio lets me observe how recommendations evolve from one update to the next.

A durable system should not require constant movement unless the underlying evidence genuinely changes.

A stock-selection model that changes its mind every few days may be reacting to noise rather than business fundamentals.

Shadow portfolios can compare strategy families

One shadow portfolio is useful.

Several can be more informative.

Different strategy families can be tested side by side.

For example:

• A value-heavy strategy

• A quality-focused strategy

• A balanced strategy

• A defensive strategy

• A growth-at-a-reasonable-price strategy

Each portfolio follows frozen rules.

The goal is not simply to crown whichever one produces the highest short-term return.

The comparison can reveal:

• Which strategy behaves most consistently

• Which one survives different conditions

• Which one experiences extreme drawdowns

• Which one depends on a narrow theme

• Which one requires frequent intervention

• Which one continues working without modification

This creates a tournament based on survival, not just excitement.

The highest-returning shadow may still be fragile

A shadow portfolio can also get lucky.

One strategy may select a company that suddenly rises several hundred percent.

That can dominate the result.

The portfolio finishes first, but the underlying process may not be broadly reliable.

The same caution used with backtests should still apply.

I would want to know:

• How many holdings contributed?

• Was the result dependent on one stock?

• Did the strategy take unusual risk?

• Did similar versions behave well?

• Was the performance consistent?

• Did it survive adverse periods?

Forward performance is stronger evidence than optimized historical performance.

It is still not proof.

Longevity can matter more than the highest return

My goal is not necessarily to trust the shadow portfolio with the largest gain after a few months.

I am more interested in which strategy can continue functioning cleanly for the longest period.

A strategy earns trust when it:

• Follows its frozen rules

• Produces investable selections

• Avoids catastrophic failures

• Behaves sensibly across changing markets

• Maintains acceptable performance

• Does not require constant repair

• Survives longer than competing strategies

Longevity suggests that the system may have captured something durable.

A spectacular short-term result may be luck.

Years of coherent behaviour are harder to dismiss.

Failure becomes useful instead of embarrassing

A failed shadow portfolio is not wasted effort.

It reveals a weakness before real capital is exposed.

Perhaps the strategy was too concentrated.

Perhaps it relied on poor-quality data.

Perhaps the ranking system was unstable.

Perhaps it worked only during one market regime.

Perhaps the portfolio rules allowed several companies with the same hidden risk.

These are valuable discoveries.

The purpose of shadow testing is not to make every strategy look good.

It is to let weak strategies fail safely.

In nature, unsuccessful variations disappear.

The ecosystem learns through survival.

A research process can use the same principle.

The rules must be archived

For shadow evidence to mean anything, the exact strategy must be preserved.

That includes:

• Metric definitions

• Weights

• Filters

• Rebalancing rules

• Position-sizing rules

• Data sources

• Portfolio constraints

• The date the rules were frozen

Otherwise, the system can slowly change while keeping the same name.

The resulting performance would combine several different strategies.

That would make the evidence difficult to interpret.

Each meaningful rule change should create a new version.

The old version continues or ends.

The new version begins its own test.

This creates a clean family history.

A shadow portfolio should not become a disguised backtest

There is still a temptation to interfere.

Suppose a strategy performs poorly for three months.

I create a revised version.

Then another.

Soon I have dozens of new shadow portfolios, each reacting to the latest disappointment.

That recreates overfitting in real time.

The process becomes a constant chase after recent conditions.

To avoid this, new versions should be created carefully.

A change should have a clear reason beyond recent performance.

The original version should remain visible.

Failures should not be erased.

The system needs a memory.

Real money should come later

A successful shadow portfolio does not automatically deserve real capital.

It earns the right to face further scrutiny.

I would still want to examine:

• The length of the test

• The number of decisions made

• The range of market conditions experienced

• Drawdowns

• Concentration

• Liquidity

• Trading costs

• Data reliability

• Whether the strategy can be followed consistently

Trust should accumulate gradually.

The system should move from idea, to backtest, to sealed validation, to shadow portfolio, and only later to limited real-money use.

Each stage should make failure more expensive only after the strategy survives the previous one.

What a backtest cannot show

A backtest can show what would have happened under a set of historical assumptions.

A shadow portfolio can show whether the entire research process continues functioning when the future is unknown.

It tests the strategy.

It tests the data pipeline.

It tests execution.

It tests portfolio construction.

It even tests the researcher’s willingness to leave the rules alone.

That is why shadow portfolios are more than paper trading.

They are a truth serum.

They force the strategy to stop explaining the past and begin confronting the future.

A backtest can make a strategy look intelligent.

A shadow portfolio gives that strategy the chance to prove it can survive without hindsight.

Subscribe now

Share

Share Fungal Intelligence for Stocks

Leave a comment

They are related, but they require different kinds of reasoning.

Stock selection asks:

Which companies appear attractive?

Portfolio allocation asks:

How should those companies be combined?

Many investment systems blur these two jobs together.

The highest-ranked company receives the largest position. The second-highest receives less. Lower-ranked companies receive smaller allocations or are excluded entirely.

That sounds logical.

But it quietly assumes that a slightly better score justifies a significantly larger financial commitment.

I do not think that assumption should be made automatically.

A ranking is not a measurement of certainty

Suppose my system gives one company a score of 84 and another a score of 80.

The first company ranks higher.

But what does the four-point difference actually mean?

It might mean the first company is genuinely more attractive.

It might also result from:

• A small difference in one valuation ratio

• A recently reported quarter

• Better data coverage

• A temporary price movement

• An arbitrary scoring threshold

• Noise inside the financial data

The score helps organize candidates.

It does not necessarily tell me that the first company deserves twice as much capital.

Rankings create order.

They do not automatically measure confidence with enough precision to determine position size.

Selection searches for attractive components

The purpose of stock selection is to narrow a large market into a manageable group of candidates.

My system may begin with thousands of companies.

It can classify them, apply financial rules, compare metrics, and reject businesses that do not fit the strategy.

This process might examine:

• Valuation

• Cash generation

• Growth

• Debt

• Profitability

• Dilution

• Deterioration

• Data quality

• Trading liquidity

At the end, the system may identify several companies that appear investable.

That is an important accomplishment.

But it does not finish the portfolio.

A company can pass every individual test and still be a poor addition because of what is already owned.

Allocation examines relationships

Portfolio allocation begins where stock selection ends.

It asks how the selected businesses interact.

Do several companies depend on the same commodity?

Are they concentrated in one country?

Do they all benefit from low interest rates?

Are several holdings exposed to the same customers or economic cycle?

Would one new position make the portfolio more resilient, or merely increase an existing risk?

The best-ranked stock may not be the best next addition.

A lower-ranked company may improve the portfolio because it brings a different source of return.

This is a systems problem.

The value of a component depends partly on the system it enters.

The best player does not always improve the team

Imagine building a sports team using only individual rankings.

You select the ten highest-scoring players available.

But every player performs the same role.

You have excellent individuals and a dysfunctional team.

A portfolio can fail in the same way.

The ten highest-ranked stocks may all be:

• Small technology companies

• Highly cyclical manufacturers

• Commodity producers

• Banks

• Expensive growth businesses

• Companies from one country

Each business may be attractive when evaluated alone.

Together, they may produce a concentrated and fragile portfolio.

Selection identifies strong players.

Allocation builds the team.

A high score can hide shared risk

Suppose several oil producers rank highly because oil prices have risen, cash flow is strong, and valuations appear low.

A ranking system may place all of them near the top.

Buying the top five could look disciplined because every company passed the same rules.

But the positions share a major dependency.

If oil prices decline, all five may weaken together.

The ranking captured company attractiveness under current conditions.

It did not automatically manage portfolio-level exposure.

A portfolio needs rules that can say:

These companies may all be attractive, but we do not need all of them.

Position sizing creates consequences

A stock score is an opinion.

Position size determines the consequence if that opinion is wrong.

That distinction matters.

I may believe a company is attractive and still limit its allocation because:

• The business is highly cyclical

• The data is incomplete

• The stock is illiquid

• The company has significant debt

• The portfolio already has similar exposure

• The strategy itself is still unproven

Position sizing should reflect the possibility of error, not just the attractiveness of the opportunity.

The more uncertain the system is, the more dangerous aggressive sizing becomes.

Why I am cautious about conviction weighting

Conviction weighting gives larger positions to the ideas an investor believes in most.

This can produce excellent results when the investor is correct.

It can also magnify the cost of being confidently wrong.

The problem is that confidence is not always well calibrated.

People often feel most confident when:

• A story is easy to understand

• Recent performance has been strong

• Other investors agree

• The company is familiar

• The thesis supports an existing belief

• The stock has already made them money

None of these guarantees that the investment is safer.

A systematic score can also create false confidence if its historical precision is exaggerated.

Giving the highest-ranked stock the largest position may turn a small modelling error into a major portfolio loss.

Equal weighting is a useful starting point

This is why equal weighting appeals to me, especially during the early development of the system.

If ten companies are selected, each receives approximately the same allocation.

Equal weighting is not based on the belief that every company is identical.

It is based on humility about my ability to rank their future performance precisely.

The system may be good enough to identify a group of attractive companies without being good enough to predict exactly which one will perform best.

Equal weighting allows selection to contribute while limiting the damage caused by ranking errors.

It also prevents one position from controlling the entire outcome.

Equal weighting does not solve everything

An equally weighted portfolio can still be badly concentrated.

Ten equal positions in ten gold miners remain one large commodity bet.

Ten equal positions in one country remain geographically concentrated.

Equal weighting controls company-level position size.

It does not automatically control shared economic exposure.

Allocation still needs constraints involving:

• Countries

• Industries

• Business models

• Currencies

• Liquidity

• Economic sensitivities

• Issuer overlap

Equal weighting is a foundation, not a complete portfolio-construction system.

Selection rules can change without changing allocation rules

Separating the two layers makes the system easier to improve.

The stock-selection model may change over time.

I may add a better measure of cash quality.

I may improve how dilution is calculated.

I may create different scoring rules for different business types.

Those changes should not automatically rewrite the logic governing portfolio concentration.

Likewise, I may improve allocation rules without changing how companies are evaluated.

Perhaps the portfolio eventually needs stronger country limits or better measures of shared risk.

The two systems can develop independently.

This is similar to programming.

One module identifies valid candidates.

Another module decides how those candidates are assembled.

Clear boundaries make the system easier to test, repair, and understand.

It makes failures easier to diagnose

Suppose the portfolio performs poorly.

What went wrong?

Did the system select weak companies?

Or did it select reasonable companies but combine them badly?

Those are different failures.

If stock selection and allocation are mixed together, the cause becomes harder to identify.

A large loss could come from:

• Bad company analysis

• Excessive position sizing

• Country concentration

• Industry concentration

• Correlated risks

• Poor rebalancing

• One dominant holding

Separating the layers creates a clearer audit trail.

The selection system can be judged by whether its candidates behave as expected.

The allocation system can be judged by whether it controls risk and builds a resilient combination.

Allocation should not repair bad selection

Portfolio construction cannot transform a bad company into a good investment.

Diversifying across many weak businesses does not create quality.

Position sizing can limit damage, but it cannot create a durable advantage where none exists.

This is why stock selection still matters.

The goal is not to assemble random companies with low correlation.

The system should first identify businesses that meet reasonable financial and valuation standards.

Allocation then decides how to combine those qualified candidates.

Good ingredients still matter.

The recipe simply matters too.

Selection should not ignore the existing portfolio

The separation between selection and allocation does not mean the two systems never communicate.

The selection model can produce a ranked list.

The allocation layer can then consider the current portfolio and choose among those candidates.

For example, the highest-ranked company might be another American industrial business.

The portfolio may already contain two similar companies.

A slightly lower-ranked Canadian insurer or Australian consumer company might provide a more useful addition.

The lower-ranked stock is not necessarily better in isolation.

It may be better for the system.

This is an important difference.

Cash can be an allocation decision

Sometimes the correct allocation is not to buy another stock immediately.

A ranking system will almost always produce a first-place company.

That does not mean the opportunity is attractive enough to deserve capital.

If valuations are poor, data quality is weak, or suitable diversification is unavailable, holding cash may be more honest than forcing a position.

Selection ranks what is available.

Allocation decides whether the available choices improve the portfolio.

The existence of a winner does not prove the existence of a good investment.

The portfolio needs its own objective

A stock-selection system might aim to identify companies with strong expected returns.

The portfolio has a broader job.

It must balance:

• Return potential

• Survival

• Diversification

• Liquidity

• Drawdown risk

• Country exposure

• Uncertainty

• The ability to continue operating after mistakes

The highest-returning collection in a backtest may not be the most useful portfolio in reality.

A system that survives can keep compounding.

A system destroyed by one concentrated error cannot benefit from future opportunities.

Allocation should therefore optimize for more than maximum historical return.

My current approach

My project is still developing, so I do not want position sizing to pretend that the system knows more than it does.

My starting structure is simple:

• Select a small group of companies that pass the relevant rules

• Aim for roughly eight to ten holdings

• Limit exposure to any single issuer

• Spread positions across several countries

• Avoid filling the portfolio with businesses driven by one shared risk

• Use approximately equal position sizes

• Do not use leverage

• Do not let a backtest alone justify aggressive allocation

This structure may change as the system earns evidence.

But complexity should arrive only when it proves that it improves decisions.

More intelligence can be added later

In the future, allocation could become more adaptive.

It might consider:

• Estimated downside

• Market regime

• Company-specific uncertainty

• Correlation under stress

• Liquidity

• Confidence in the underlying data

• The historical survival of each strategy family

But those additions should be treated cautiously.

A sophisticated allocation model can overfit just as easily as a stock-selection model.

It can create the illusion that risk is being measured precisely when the future remains uncertain.

The allocator should earn complexity through evidence.

Until then, simple constraints may be safer.

Two separate questions

When evaluating an investment, I want to ask two questions in order.

First:

Does this company deserve consideration?

Then:

Does this position improve the portfolio, and how much capital should it receive?

The first question evaluates the business.

The second evaluates the system.

A good stock can be a bad addition.

A slightly lower-ranked stock can play a more valuable role.

And a strong opinion does not automatically deserve a large position.

Stock selection finds the ingredients.

Portfolio allocation decides how much of each ingredient belongs in the final mixture.

Treating those as separate jobs helps prevent one attractive company—or one imperfect ranking—from controlling the survival of the entire portfolio.

Subscribe now

Share

Share Fungal Intelligence for Stocks

But if every plant is the same species, the entire field may depend on the same soil conditions, the same weather, and the same resistance to disease.

One threat can spread through everything.

A portfolio can have the same weakness.

It may contain ten, twenty, or even fifty stocks and still behave like a monoculture.

The company names are different.

The underlying dependency is not.

What is a monoculture?

A monoculture is an environment dominated by one type of organism.

This approach can be efficient.

When every plant has similar needs, the land can be managed using the same tools, schedules, and processes.

The harvest may also be highly productive when conditions are favourable.

But efficiency creates concentration.

If every plant shares the same weakness, one disease, pest, drought, or temperature change can damage the entire system.

There are few alternative organisms available to slow the threat or continue functioning under different conditions.

The system becomes optimized for one environment.

It becomes fragile when that environment changes.

Portfolios can become financial monocultures

An investor might own several companies and believe the portfolio is diversified.

But perhaps every company is:

• A rapidly growing technology business

• Dependent on cheap financing

• Sensitive to consumer spending

• Exposed to the same commodity

• Located in one country

• Valued using optimistic future expectations

• Benefiting from the same economic cycle

These stocks may operate in different industries.

They may sell different products.

They may even appear under different sector labels.

But they still depend on similar conditions.

When those conditions weaken, the whole portfolio may decline together.

The investor owned many stocks.

Economically, the investor owned one environment.

Success can hide concentration

Monocultures often appear strongest during favourable periods.

When the environment perfectly matches the organism, growth can be fast and efficient.

The same thing happens in markets.

Suppose low interest rates support expensive growth stocks.

Companies can borrow cheaply.

Investors are willing to wait years for future profits.

High valuations seem reasonable because alternative returns are low.

A portfolio filled with these companies may perform extremely well.

That success can make the investor believe the portfolio is strong.

But the system may simply be highly adapted to one season.

When rates rise, financing becomes more expensive and distant profits become less valuable.

Several companies can weaken at once.

The problem was not necessarily that every company was bad.

The problem was that they all needed the same weather.

Diversification is not about counting names

Owning more stocks does not automatically create a resilient portfolio.

Ten oil producers are not ten independent sources of return.

Neither are ten regional banks exposed to the same housing market.

A portfolio of software companies, semiconductor companies, online retailers, and speculative biotechnology firms may look varied.

Yet all of them may depend on abundant capital and high investor risk tolerance.

Counting ticker symbols measures the number of positions.

It does not measure the number of independent risks.

A better question is:

How many different economic conditions can this portfolio survive?

Different sectors can share the same root

Sector diversification is useful, but it is not enough.

Consider a portfolio containing:

• A homebuilder

• A mortgage lender

• A furniture retailer

• A building-material company

• A real estate website

• A utility serving a rapidly growing suburb

These businesses belong to different sectors.

But they all depend partly on housing activity.

If home sales fall, mortgage demand weakens, construction slows, and consumer spending on furniture declines.

Different branches can still grow from the same root.

This is why I want to understand the forces behind each company rather than relying only on its label.

Valuation can create a monoculture

Companies do not need to share an industry to share a risk.

They may all be priced using the same assumptions.

Suppose a portfolio contains businesses from technology, healthcare, consumer products, and clean energy.

Each company is expected to grow rapidly for many years.

Each trades at a high valuation.

Each requires investors to remain confident about profits far in the future.

The industries are different.

The valuation risk is similar.

If expectations fall, the entire portfolio can decline even when the businesses continue operating normally.

The shared vulnerability was not the product.

It was the price paid for the expected future.

Investors can create monocultures through one philosophy

An investment philosophy can also become too narrow.

A strict value investor may own only statistically cheap companies.

A growth investor may own only rapidly expanding companies.

A dividend investor may own only high-yield stocks.

A momentum investor may own whatever has recently risen the most.

Each approach may contain useful ideas.

But when every holding is selected for the same reason, the portfolio may inherit the same weakness repeatedly.

Cheap companies may be cheap because their businesses are deteriorating.

Fast-growing companies may depend on optimistic expectations.

High-yield stocks may be concentrated in mature, indebted industries.

Momentum stocks may all reverse when market leadership changes.

A philosophy helps create discipline.

It should not become blindness.

Nature uses diversity as insurance

Healthy ecosystems often contain organisms with different roles and survival strategies.

Some grow quickly.

Others grow slowly but live longer.

Some tolerate drought.

Others thrive in wet conditions.

Some recycle dead material.

Others provide shelter, food, or protection.

Not every species succeeds at the same time.

That is part of the strength of the system.

When one population weakens, another may continue functioning.

Diversity creates redundancy and alternative pathways.

A portfolio can benefit from the same principle.

Different companies may respond differently to:

• Economic growth

• Recessions

• Inflation

• Interest rates

• Commodity prices

• Currency movements

• Consumer confidence

• Government policy

The goal is not to find stocks that never fall.

It is to avoid requiring every holding to succeed under exactly the same conditions.

Diversity has a cost

A diverse ecosystem is not always the fastest-growing system during perfect conditions.

A monoculture may produce a larger short-term harvest because every resource is directed toward one optimized crop.

Portfolio diversification can also reduce performance when one theme is dominating the market.

If technology stocks are rising rapidly, owning banks, utilities, industrials, or consumer businesses may feel like a mistake.

The concentrated investor may outperform.

The diversified investor may appear overly cautious.

But diversification is not designed to win every season.

It is designed to reduce dependence on predicting which season comes next.

That protection often feels unnecessary until the environment changes.

False diversification is especially dangerous

The most dangerous portfolio may not be the one that openly owns a single stock.

At least that concentration is visible.

False diversification creates comfort without reducing much risk.

An investor sees many names, sectors, and positions and assumes the portfolio is protected.

But hidden dependencies remain.

This can encourage larger risks because the investor believes those risks have already been spread out.

The portfolio may be diversified by appearance but concentrated by cause.

A system should therefore examine shared drivers such as:

• Revenue sources

• Customer groups

• Financing needs

• Geographic exposure

• Currency exposure

• Commodity sensitivity

• Interest-rate sensitivity

• Valuation assumptions

• Market liquidity

• Regulatory dependence

These connections matter more than the number of rows in a brokerage account.

Countries can behave like separate environments

One reason I am interested in companies from several countries is that different markets do not always move for exactly the same reasons.

Countries can have different:

• Economic cycles

• Interest-rate policies

• Currencies

• Natural resources

• Industry structures

• Political risks

• Consumer conditions

Geographic diversification does not eliminate risk.

A global recession can affect nearly everyone.

Countries can also be more connected than they initially appear.

But spreading exposure across several environments may reduce dependence on one national outcome.

A portfolio containing two companies from each of several countries may be more resilient than one dominated by a single market—provided those companies also represent different economic risks.

Equal weighting can limit monoculture damage

Even a diversified portfolio can become dominated by one theme if its largest positions share the same exposure.

Position sizing matters.

Suppose five different types of companies each receive small allocations, while one technology company receives half the portfolio.

The portfolio’s fate still depends heavily on technology.

Equal weighting is one simple way to limit this problem.

It does not guarantee true diversification.

Ten equal positions can still share the same risk.

But it prevents one stock from turning the rest of the portfolio into decoration.

Until I have strong evidence that one opportunity deserves more capital, equal weighting provides a useful boundary against excessive confidence.

Not all correlation is visible in advance

Investors often measure diversification using historical correlation.

That can be helpful.

But correlation is not permanent.

Two stocks may behave differently during normal conditions and then fall together during a crisis.

When markets become stressed, investors sell risky assets, lenders reduce credit, and customers cut spending.

Companies that appeared independent may suddenly share the same pressure.

Historical numbers may underestimate relationships that emerge only during extreme conditions.

This is why diversification should be based on economic reasoning as well as statistics.

I want to know not only whether two stocks moved together in the past, but also why they might fail together in the future.

A resilient portfolio needs different roles

A portfolio can be designed like an ecosystem.

Each holding should contribute more than another ticker.

One company may provide steady cash flow.

Another may offer long-term growth.

One may benefit from inflation.

Another may have low debt and defensive demand.

One may operate in a resource-rich country.

Another may benefit from technological change.

The portfolio does not need a complicated story for every position.

But each addition should answer a question:

What does this company add that the portfolio does not already have?

If the answer is merely “more exposure to the same successful theme,” the system may be becoming a monoculture.

Concentration is not automatically wrong

There are investors who understand a small number of businesses deeply and deliberately maintain concentrated portfolios.

That approach can succeed.

Concentration can increase returns when the analysis is correct.

It may also be appropriate when the investor has unusual knowledge, strong conviction, and the ability to tolerate large losses.

But concentration should be recognized honestly.

A portfolio should not claim the protection of diversification while behaving like one large bet.

My project is not built around the assumption that I can reliably identify a single perfect company.

It is built around controlling the damage caused by mistakes.

That makes avoiding hidden monocultures especially important.

The goal is not maximum variety

Diversification should not become random collection.

Owning weak companies solely because they belong to different industries does not strengthen a portfolio.

A resilient ecosystem still needs healthy organisms.

The goal is to find strong or attractively priced companies whose sources of success are not identical.

Quality and diversity must work together.

Too much concentration creates fragility.

Too much careless variety creates mediocrity.

Portfolio construction is the search for a useful balance.

The lesson from the forest

A forest containing many species can absorb shocks in ways a single-crop field cannot.

Some organisms will struggle.

Some will adapt.

Others may benefit from the change.

The system survives because its future does not depend on one response.

A portfolio should aim for the same kind of resilience.

It should not require one industry, country, economic regime, or investment style to remain permanently dominant.

A collection of stocks becomes a portfolio only when the relationships between them are understood.

Many companies can still represent one bet.

Many species create a better chance of surviving many seasons.

That is why monocultures fail in nature—and why they can fail just as dangerously inside a portfolio.

Subscribe now

Share

Share Fungal Intelligence for Stocks

Leave a comment