Glossary
Survivorship bias
The distortion that arises when only the winners that still exist are counted, and the failures that disappeared are left out.
On this page
Survivorship bias is the error of drawing conclusions from the things that survived some process while ignoring those that did not. In investing, the funds, companies and strategies that failed tend to vanish from the records. What remains looks better than the full history really was.
Where it appears
- Fund performance. Poorly performing mutual funds are closed or merged into others, and their records drop out of many databases. The average return of funds existing today overstates what investors actually earned.
- Hedge fund indexes. Reporting is voluntary. Funds stop reporting when results turn bad, and new funds add their good early history after the fact.
- Backtests. Testing a strategy on today's index members assumes you knew years ago which companies would last. Bankrupt and delisted stocks are missing.
- Stock market history. The US market's long record is that of one of the most successful countries. Markets in Russia and China were wiped out entirely in the twentieth century.
- Success stories. Studying the habits of famous investors ignores the many who did the same things and failed.
The S&P SPIVA reports count this way. Over 20 years, well over half of US stock funds have typically been merged or liquidated.
How to guard against it
- Ask what is missing. Does the data include the dead funds, the delisted stocks, the abandoned strategies?
- Use survivorship-free data in backtests: index membership as it was at each date, with delisting returns included.
- Judge a group from its starting line-up, not its current one.
- Be wary of track records selected after the fact. A firm that launched ten funds and advertises the best two is showing survivors.
The related look-ahead bias uses information that was not available at the time, such as restated earnings.
