During World War II, a team of military engineers stared at returning fighter planes covered in bullet holes and thought they'd found the answer.

The damage was everywhere — wings, tail, fuselage. So the logical move seemed obvious: reinforce the areas taking the most hits.

Then a statistician named Abraham Wald looked at the same data and said something that changed the way we think about success, failure, and evidence itself.

He said they were looking at the wrong planes entirely.

This single correction became one of the most famous case studies in statistics. It's called survivorship bias, and once you understand it, you'll start seeing it everywhere — in business advice, gym transformations, "the good old days," and even your investment portfolio.


What Is Survivorship Bias?

Survivorship bias happens when we draw conclusions from the "survivors" of a process while ignoring everyone who didn't make it.

The failures go silent. They disappear from the data. And whatever's left starts to look like the whole story — even though it's only half of it.

It's not that the visible data is fake. It's that it's incomplete in a way that quietly flips the conclusion upside down.


The WWII Bomber Story

Here's what actually happened.

The U.S. military was losing bombers over Europe at an alarming rate. To fix this, they studied the planes that came back from missions, mapping every bullet hole on the fuselage, wings, and tail.

The damage clustered heavily in certain spots. So the initial plan was to add armor exactly where those bullet holes were concentrated.

Abraham Wald, working with the Statistical Research Group, pointed out the fatal flaw in that logic.

The planes they were studying were the ones that survived and made it home. The bullet holes on those planes marked the areas a plane could take damage and still fly.

The planes that got hit in the other spots — the engine, the cockpit, the fuel lines — never made it back at all. Their data was missing because those planes were sitting at the bottom of the English Channel or in a field somewhere in France.

Wald's recommendation was the opposite of what everyone expected: armor the places with no bullet holes, not the places covered in them. Those were the spots that, when hit, meant the plane never came home to be counted.

This reframe likely saved thousands of lives, and it's still taught today as one of the clearest real-world examples of statistical thinking done right.


Example 1: Business Books and "Success Secrets"


Walk into any bookstore and you'll find shelves of books studying wildly successful companies, pulling out the habits, mindsets, and strategies that supposedly made them win.

The problem is the same one Wald identified.

These books study the businesses that survived. They don't study the hundreds of companies that had the exact same habits, mindsets, and strategies — and still failed.

"Take big risks," "trust your gut," "ignore the naysayers" — these traits appear in plenty of failed startups too. We just don't write books about them, because nobody wants to read a case study on a company nobody's heard of.

The survivors get credit for the traits. The traits didn't necessarily cause the survival.


Example 2: The College Dropout Myth


Bill Gates, Mark Zuckerberg, and Steve Jobs all dropped out of college and built massive companies. This story gets repeated so often it's practically folklore now.

But this is survivorship bias wearing a hoodie.

For every dropout who built a trillion-dollar company, there are thousands of dropouts whose startups quietly failed, who took on debt with no degree to fall back on, and whose stories never made it to a graduation-speech slideshow.

We don't hear from them because failure doesn't get invited to give TED talks.

The lesson isn't "dropping out works." The lesson is that we only hear from the tiny percentage where it worked out — and that sample tells us almost nothing about the odds.


Example 3: "They Don't Build Them Like They Used To"


Old stone buildings, ancient cathedrals, centuries-old bridges — they're often held up as proof that older construction was simply better and more durable.

Here's the catch: we're only looking at the buildings still standing.

For every ancient structure still around today, countless others from the same era collapsed, crumbled, or were torn down and forgotten within decades of being built. Nobody preserves a monument to a badly built medieval wall.

What we're actually looking at is a heavily filtered sample — the best-engineered, best-located, or luckiest structures from thousands of years of construction. The ones that survived make the entire era look sturdier than it really was.


Example 4: Mutual Fund and Investing Performance


This one costs people real money.

When financial companies report the "average" performance of mutual funds over the past 20 years, that number often looks impressively strong.

But underperforming funds routinely get shut down, merged into other funds, or quietly removed from the lineup. They vanish from the historical record — and take their bad returns down with them.

What's left in the average is skewed toward the funds that survived long enough to keep reporting numbers. This is well documented in financial research and even has its own name: survivorship bias in fund data.

If you're comparing "20-year fund performance," you may be comparing yourself only to the winners, with the losers erased from the equation entirely.


Why This Bias is so Easy to Fall for

Survivorship bias is sneaky because the surviving data feels complete. It's right in front of you — visible, countable, easy to study.

The missing data, by definition, isn't there to remind you it's missing.

Our brains are also wired to build stories from whatever evidence is available, rather than pausing to ask what evidence might be absent. This is closely tied to how the availability heuristic shapes our judgment — we lean on the examples that are easiest to recall, not the ones that are most representative.


How to Spot Survivorship Bias in Everyday Life

A few habits make this bias much easier to catch before it fools you.

Ask what happened to the group that isn't being shown. If you're looking at successful people, businesses, or products, actively look for the failures that used the same approach.

Question "average" statistics. Ask whether the failures were dropped from the dataset before the average was calculated.

Notice survivor-only sources. Testimonials, before-and-after photos, and "success stories" are almost always survivor data by design — the people who didn't get results rarely submit a testimonial.

Remember the plane. If a pattern in the data seems obvious, ask yourself what Abraham Wald would ask: what's missing from this picture?


FAQs


What is the survivorship bias plane story, in one sentence?

It's the WWII story where the military almost armored the wrong parts of bomber planes because they only studied planes that survived combat, until Abraham Wald pointed out that the missing planes held the real answer.


Who was Abraham Wald?

He was a mathematician and statistician working with the U.S. Statistical Research Group during WWII, known for reframing the aircraft armor problem using survivorship logic.


Is survivorship bias the same as selection bias?

They're related. Survivorship bias is a specific type of selection bias where the "selection" happens because certain data (usually failures) never makes it into the sample at all.


Why is survivorship bias dangerous in real life?

Because it makes patterns look stronger and more reliable than they actually are, leading to bad decisions in business strategy, investing, health choices, and personal goals.


How can I avoid survivorship bias in my own thinking?

Actively search for the failures, not just the successes, before drawing a conclusion. If you can only find winners in your research, that itself is a red flag.


Final Thoughts

The survivorship bias plane story sticks with people because it's such a clean, visual example of a mistake we make constantly, just without bullet holes to make it obvious.

Every "top 10 successful founders" list, every "these habits changed my life" post, and every "average returns" chart is quietly shaped by the same blind spot Wald caught in 1943.

The fix isn't complicated, even if it takes practice. Before trusting a pattern, ask who or what got left out of the data — and whether their absence is the whole reason the pattern exists in the first place.