Survivorship Bias
Survivorship bias is the error of studying only the visible winners while ignoring everyone who tried the same thing and disappeared from the sample. In fooled-by-randomness, Taleb uses this to explain why performance records, business success stories, and expert reputations often exaggerate skill.
Pattern
- Many participants take high-variance actions.
- A few win by chance, timing, or hidden exposure.
- Observers study the winners.
- The losers are absent, so the strategy looks more reliable than it is.
The result is false learning: the visible sample teaches confidence when it should teach caution.
Where It Appears
- Trading managers with excellent recent returns but hidden tail risk.
- Entrepreneurs whose stories omit the failed cohort.
- Backtests that only include assets or funds that still exist.
- Experts selected for past calls without considering base rates.
- Personal development advice based on unusual outlier paths.
Practical Test
Ask:
- Who is missing from this sample?
- How many tried and failed?
- Did the method cause success, or did the winner merely survive randomness?
- Would this record still look good if the full graveyard were included?
Connections
- illusions-of-competence - Survivorship bias can make weak methods look competent.
- trading-edge - Edge claims must survive full-sample scrutiny.
- reasonable-expectations - Extraordinary records often hide unusual risk or selection effects.
- active-management-as-error-detection - Active managers must distinguish real mistakes from lucky records.
Sources
- fooled-by-randomness - Monkeys, millionaires, performance records, and expert-selection examples.