Understand survivorship bias — why it's dangerous to learn only from successes and visible cases while missing the “dead” data.
Where the unseen holes are
Abraham Wald's planes
Survivorship bias is when we draw conclusions only from the cases that “survived to reach us”, forgetting the ones that didn't survive and are therefore invisible.
The classic story is World War II. The military studied bombers returning from missions, marked where they had the most bullet holes (wings, tail, fuselage), and wanted to add armor exactly there. Logical? No.
The mathematician Abraham Wald pointed out the mistake: they were looking ONLY at the planes that came back. If a plane with a shot-up wing still made it home, a hit to the wing wasn't fatal. And the places with almost NO holes (the engines, for example) looked clean precisely because planes hit there did NOT come back and never made it into the sample. The armor had to go where the survivors had no holes.
This turns intuition upside down: what matters is not only the visible part of the data but also the invisible one — those who “didn't make it back”.
Lesson notes
Abraham Wald's planes
Survivorship bias is when we draw conclusions only from the cases that “survived to reach us”, forgetting the ones that didn't survive and are therefore invisible.
The classic story is World War II. The military studied bombers returning from missions, marked where they had the most bullet holes (wings, tail, fuselage), and wanted to add armor exactly there. Logical? No.
The mathematician Abraham Wald pointed out the mistake: they were looking ONLY at the planes that came back. If a plane with a shot-up wing still made it home, a hit to the wing wasn't fatal. And the places with almost NO holes (the engines, for example) looked clean precisely because planes hit there did NOT come back and never made it into the sample. The armor had to go where the survivors had no holes.
This turns intuition upside down: what matters is not only the visible part of the data but also the invisible one — those who “didn't make it back”.
Billionaire college dropouts
This error is everywhere in success stories.
— “Drop out of college — Jobs, Gates, and Zuckerberg dropped out and became billionaires!” We see three famous “survivors” and don't see the millions who dropped out and didn't get rich. You can't conclude from successful college dropouts that dropping out pays off.
— “They used to build things to last: look at the old houses that have stood for centuries!” Only the best and sturdiest buildings have come down to us; thousands of bad ones collapsed or were torn down long ago. We admire the survivors and think that “everything used to be better made”.
— “This investor/trader always calls the market!” Out of thousands of players, someone will guess right many times in a row by pure chance — and he's the one shown as a guru. The losers aren't shown.
The antidote is to always ask: “Who am I NOT seeing here? Where are the ones who didn't succeed?” Success is visible; failure usually stays silent.