Fault Tolerant

Imagine you’ve spent years building a portfolio.

You have $100,000 invested.

Then one morning, one of your largest investments has a catastrophic problem.

Maybe the company discovers accounting fraud.

Maybe its main product fails.

Maybe a competitor makes its business model obsolete.

Maybe the company simply goes bankrupt.

Your investment falls 80%. Or 90%. Or even all the way to zero.

Now here’s the important question: How much of your $100,000 disappears with it?

If that investment represented 5% of your portfolio, a complete loss would cost you roughly $5,000.

Painful? Absolutely. Fatal to the portfolio? Probably not.

But if that investment represented 70% of your portfolio, the same failure could destroy $70,000.

Same company. Same failure. Completely different outcome.

That’s the idea behind diversification.

And if you’re a software engineer, there’s a concept you already know that makes this surprisingly easy to understand: It’s called fault tolerance.

Imagine you run an application.

Everything depends on one database server. No replica. No backup. No failover. No redundancy.

If that server dies, the application dies. You have created a single point of failure.

The system might work perfectly for years. But its reliability is ultimately limited by one component.

Now look at an investment portfolio.

Suppose you have $100,000 invested entirely in one company’s stock.

That company might be fantastic. Its products might be excellent. Its management might be brilliant. Its financial statements might look beautiful.

But your entire financial outcome now depends heavily on one company continuing to perform well.

That’s concentration. And concentration creates a single point of failure.

In software, we design systems so one component can fail without bringing everything down.

A diversified portfolio tries to do something similar: One investment can fail without becoming a portfolio-level disaster.

Remember Maya and Leo?

In the first book, they learned about the power of starting early. In the second book, they faced the problem of trying to predict the market.

Now they’re facing a different problem. Both have $100,000 to invest.

Leo has found a company he absolutely loves. He’s researched the business. He understands the industry. He believes the company’s future is enormous. So Leo invests $80,000 in it.

The remaining $20,000 is spread across several other investments.

Maya takes a different approach. She likes some of the same companies Leo likes. But she doesn’t want one company to determine the outcome of her entire portfolio. So she spreads her $100,000 across many investments.

“If I’m right about this company, I’ll make a lot more money.”

“True. But what happens if you’re wrong?”

That’s the question diversification is designed to address.

Let’s make the difference visible. Imagine both portfolios contain $100,000.

Now suppose Company A falls 50%.

Leo loses: $80,000 × 50% = $40,000. His portfolio falls from $100,000 → $60,000 (-40% overall).

Maya loses: $20,000 × 50% = $10,000. Her portfolio falls from $100,000 → $90,000 (-10% overall).

Neither portfolio is immune to the decline. But Maya’s portfolio has something Leo’s doesn’t have: damage containment.

The failure of one component doesn’t dominate the entire system.

Diversification sounds complicated because people often explain it with complicated portfolios. The underlying idea is much simpler.

Diversification means spreading your investment exposure so that the poor performance of one investment has a limited effect on the overall portfolio.

Think about a box of eggs. If you put every egg into one basket and drop it, you have a problem. If you spread the eggs across several baskets, dropping one basket doesn’t destroy all of them.

Investing works differently in many important ways, but the underlying risk-management idea is similar: You’re reducing your dependence on one outcome.

Here’s a simple mental model you can use to inspect your portfolio: It’s called the Zero Test.

Suppose your portfolio is worth $100,000.

You own $5,000 of Company A. If Company A goes to zero: $100,000 → approximately $95,000 (a 5% portfolio loss).

Now suppose you own $40,000 of Company A. If it goes to zero: $100,000 → approximately $60,000 (a 40% portfolio loss).

Same company. Same failure. Different position size.

The Zero Test doesn’t tell you what percentage you should own. It gives you something more useful: a way to understand your exposure.

There’s a simple relationship hiding underneath the Zero Test.

If an investment represents p% of your portfolio and falls L%, then—ignoring interactions with the rest of the portfolio—the approximate portfolio impact is:

For example: A position represents 10% of your portfolio and falls 50%. Approximate portfolio impact: 10% × 50% = 5% portfolio loss.

Now increase the position to 40%. 40% × 50% = 20% portfolio loss. The same investment decline now produces a quadrupled portfolio loss.

This is why position size matters. A great investment can still become a dangerous portfolio position if you own too much of it.

Leo isn’t necessarily making a bad investment. That’s important.

He may actually be correct about Company A. The company could perform incredibly well. That’s precisely what makes concentration so tempting.

If you put $80,000 into an investment and it doubles, you’ve made $80,000. If you put only $20,000 into it and it doubles, you’ve made $20,000.

Concentration increases your exposure to the upside. But it also increases your exposure to the downside. That’s the trade-off.

Diversification isn’t a machine that creates extra returns. It’s a way of managing how much one mistake can hurt you.

Not all investment risks are the same.

Imagine Company A develops a major accounting problem. Company B doesn’t. Company C doesn’t. Company D doesn’t.

If you own only Company A, the problem directly hits your portfolio.

If you own Companies A through D, the problem with Company A is still painful—but it doesn’t automatically affect the others in the same way.

This is often called company-specific risk or idiosyncratic risk.

• Management problems
• Fraud
• Product failure
• Lawsuits
• Technological disruption
• Loss of a major customer
• Bankruptcy

Owning multiple investments can reduce your dependence on any one company’s outcome. But there’s another category.

Imagine every major stock market falls 30%.

You own 100 different stocks. You’re diversified. But you’re still likely to experience a significant decline.

Why? Because the problem isn’t one company. It’s the broader market.

Economic recessions, financial crises, interest-rate shocks, geopolitical events, and other large forces can affect many investments simultaneously.

Diversification cannot make market risk disappear.

Diversification can reduce concentration risk. It cannot eliminate investing risk.

That’s an important idea because diversification is sometimes sold as if it were financial armor. It isn’t. It’s more like a shock absorber.

The road can still be rough. The goal is to reduce how violently one bump affects the entire vehicle.

Here’s another subtle problem. Suppose someone tells you: “I own 30 stocks. I’m diversified.”

Maybe. But let’s look underneath the hood.

Imagine those 30 companies are all technology companies, and many of them depend on the same trends: cloud spending, semiconductor demand, advertising, artificial intelligence, consumer technology.

You’ve diversified across companies. But perhaps not nearly as much across sources of risk.

Now imagine owning 30 companies where 25 of them tend to move together. You have 30 holdings. But economically, you may still have a much smaller number of independent bets.

This is where the concept of correlation becomes useful.

Correlation is simply a way of describing how two things tend to move relative to each other.

If two investments frequently move in the same direction at the same time, they have higher correlation. If they behave more independently, their correlation is lower.

You don’t need to become a statistics expert to understand the practical implication. Consider two portfolios:

Both portfolios contain 30 investments. But they don’t necessarily have the same diversification.

Counting investments isn’t the same as measuring diversification. Thirty versions of the same bet are still, in an important sense, one big bet.

Remember Maya? She isn’t trying to predict which company will become the next superstar.

Instead, she’s asking a different question: “What happens if I’m wrong?”

That’s a powerful shift. Leo’s strategy depends heavily on being right about Company A.

Maya’s strategy is designed so that being wrong about one company doesn’t destroy her entire plan.

That’s what makes diversification powerful: It doesn’t require perfect prediction. In fact, it’s partly a response to the fact that prediction is difficult.

You don’t know which company will dominate in ten years, which industry will struggle, or which business model will become obsolete.

So instead of betting everything on one forecast, you can design the portfolio so that multiple outcomes remain survivable.

Diversification doesn’t stop at individual stocks. You can think about diversification at several levels:

Each layer can introduce a different source of diversification.

But don’t turn this into a checklist that says: “Own exactly X% of this and Y% of that.” We’re building the mental model first.

Here’s where things get interesting: If diversification is good, shouldn’t you just own everything?

Not necessarily. Imagine you own 2,000 investments, each representing 0.05% of your portfolio.

You’ve certainly reduced dependence on any one investment. But now the portfolio can become difficult to understand.

More isn’t automatically better. The goal isn’t the maximum number of investments. The goal is the appropriate distribution of risk.

Let’s bring everything together with a simple framework. Look at your portfolio. Find your largest positions. Then ask three questions:

For example: You might own five different technology companies. But if all five depend heavily on the same industry trend, they may experience similar problems.

The objective isn’t to eliminate every possible connection. That’s impossible. The objective is to understand the connections you actually have.

Leo opens his portfolio. Company A represents 80%. He runs the Zero Test: Company A → $0. His $100,000 portfolio becomes approximately $20,000.

Leo suddenly sees something he hadn’t considered: He had effectively made his financial future heavily dependent on Company A.

Maya runs the same test. Her largest position represents 10%. If it goes to zero: $100,000 → approximately $90,000. She wouldn’t be happy. But her entire financial plan hasn’t collapsed.

“So you think your investments won’t fail?”

“No. I think some of them probably will. That’s why I don’t want one failure to decide everything.”

And that’s the heart of fault tolerance.

Investing is often discussed as if the objective were simply: “Find the investment with the highest return.” But that’s incomplete.

A portfolio exists over time. And over long periods, unexpected things happen: Companies fail, industries change, technologies become obsolete, markets crash, predictions turn out to be wrong.

The investor’s job isn’t necessarily to predict every failure. It’s to build a system that can survive being wrong.

Think about software again: A good production system isn’t designed around “Nothing will ever fail.” It’s designed around “Things will fail. What happens when they do?”

That’s fault tolerance. And a resilient portfolio follows the same philosophy.

If you remember only five things from this book, remember these:

Want to see how much one investment can affect your portfolio? Experiment with different position sizes, portfolio values, and hypothetical investment losses:

Try something simple. Start with a $100,000 portfolio. Give one investment 5%. Then 10%. Then 25%. Then 50%. Finally, run the Zero Test. Watch what happens. The numbers make the concept surprisingly obvious.

You don’t build a reliable system by assuming every component will work forever. You build it by planning for failure.

Investing is no different. A company can fail. An industry can struggle. A prediction can be wrong. A market can crash.

You can’t control all of those things. But you can control how much power any one of them has over your portfolio.

A strong portfolio isn’t one where nothing ever goes wrong. It’s one where something can go wrong without everything going wrong.

That’s fault tolerance. And that’s how you build a portfolio designed to survive the unexpected.

We’ve now built three pieces of the system: The Money Loop taught us how money can grow on top of money. Buy the Average showed us how a simple investing rule can reduce the need to predict short-term market movements. Fault Tolerant showed us how diversification can reduce our dependence on any single investment.

But there’s another problem. It doesn’t require a company to collapse. It doesn’t require a market crash. It doesn’t even require you to make a bad investment. Your investments can perform reasonably well... and money can still quietly disappear.