Survivorship Bias
We draw conclusions from visible successes while ignoring the invisible failures.
Origin & History
The term was popularized by Abraham Wald's analysis during WWII. Wald, a statistician working with the US military, was asked to analyze bullet holes in returning bombers to determine where to add armor. He pointed out the crucial flaw: the sample consisted only of planes that survived. The planes with holes in engines and cockpits never returned. Reinforcing the visible damage would have been optimizing for the wrong thing. Wald's insight became a foundational concept in statistics and decision-making.
Real-World Examples
Military analysis of bullet holes on returning planes showed most damage on wings and tails — leading to plans to reinforce those areas. Abraham Wald argued the opposite: reinforce engine and cockpit, where surviving planes showed no damage.
Business culture celebrates the bold risk-takers who succeeded, producing biographies, TED talks, and case studies about their traits. The equally bold risk-takers who failed leave no such literature — producing a skewed picture of what traits lead to success.
Fund managers who outperformed for 10 years are held up as evidence that skill exists in markets. The funds that underperformed and were closed don't appear in the dataset — making the average of 'surviving' funds misleadingly positive.
Why It Matters
Survivorship Bias is especially dangerous because it is invisible by definition: the missing data cannot be observed. The corrective habit is to always ask 'what isn't in this sample?' before drawing conclusions. In business: how many companies tried this strategy and failed? In medicine: how many people took this treatment and didn't recover? In education: how many people followed this path and it didn't work? The question 'where are the planes that didn't come back?' is one of the most important questions in any empirical analysis.
Related Laws
Can You Spot Survivorship Bias in the Wild?
Play Mind Traps — 40 psychology laws, one real scenario each. Free, no login.
Play the Game — Free →Frequently Asked Questions
The logical error of drawing conclusions only from subjects that passed a selection filter — ignoring those that didn't, because they're no longer visible.
The concept was rigorously demonstrated by statistician Abraham Wald during WWII in his analysis of bullet holes in returning bombers.
Want a deeper dive?
Read Full Article on the Blog →