Survivorship bias is the error of studying only the winners and ignoring everyone who failed along the way. When the losers disappear from the data, the winners look more skilled, the strategy looks more reliable, and the odds of success look far better than they really are. It’s one of the most expensive cognitive traps in investing, because the evidence that would correct it has, by definition, vanished.
The WWII planes that started it all
The classic example comes from World War II. American bombers were returning from missions riddled with bullet holes, and the military wanted to add armor. The holes clustered on the wings and fuselage, so the obvious move was to armor those spots. Statistician Abraham Wald pointed out the flaw: the planes being studied were the ones that survived. The bullet holes showed where a plane could be hit and still make it home. The missing holes, on the engines and cockpit, marked where the planes that never returned had been hit. They armored the spots with no holes.
Every investing dataset has the same problem. We study the survivors because they’re the only ones left to study, and we forget to ask what the missing data would have told us.
Where it hides in investing
Mutual fund performance is the purest example. Databases track the funds that exist today, but the funds that performed terribly were closed or merged out of existence, taking their bad records with them. Studies of fund performance that ignore dead funds overstate average returns significantly. The industry looks far more skilled than it is because its failures are literally deleted from the record.
The same bias infects stock market history. “Stocks return about 10% a year” is computed from markets that survived, the United States above all. Investors in Russia in 1917 or China in 1949 saw their markets go to zero, and those outcomes don’t appear in the long-run charts. Backtested trading strategies have it worst of all: anyone can mine historical data until they find a pattern that “worked,” but the thousands of patterns that failed never get published.
The dropout billionaire fallacy
Outside markets, survivorship bias powers the most seductive bad advice in business. Bill Gates, Steve Jobs, and Mark Zuckerberg all dropped out of college and became billionaires, so dropping out starts to look like a strategy. What’s invisible is the denominator: the hundreds of thousands of dropouts whose companies failed and who never appear in a single profile. The winners write the memoirs, give the commencement speeches, and get the documentaries. The losers are silent, not because they have nothing to teach, but because nobody interviews them.
Nassim Nicholas Taleb built much of Fooled by Randomness around this exact trap: we mistake luck for skill because the lucky are the only ones still visible. The fund manager with five great years gets the magazine cover; the hundred managers who flipped the same coin and lost get forgotten. Enough people take enough risks and someone was bound to look like a genius.
How to see through it
The antidote is to always ask what’s missing. When you see an impressive track record, ask how many others tried the same thing and failed. When you read a success story, ask about the base rate: what percentage of people who took that path actually succeeded? When you evaluate a fund or strategy, ask whether dead competitors were included in the numbers.
Practically, this means preferring data that includes the graveyard: survivorship-bias-free databases, full-cycle track records, and strategies tested on data the creator hadn’t seen. And it means treating every vivid winner’s story, the dropout billionaire, the trader who called the crash, the fund that never had a down year, as a sample of one drawn from an unknown number of attempts. The winners can still teach you something. Just remember you’re reading the planes that made it home.
You must be logged in to post a comment.