What is an example of survivorship bias in investing?

Backtesting a strategy on today's S&P 500 members. Those companies are in the index because they grew and survived; the ones that went bankrupt, were acquired or shrank out of the index are missing. Any strategy tested on that list inherits a head start no investor in the past could have had, because nobody knew then which companies would survive.

Is the Opulence Alpha universe free of survivorship bias?

Not entirely. Its S&P 500 layer largely is: it holds every member since 1996 whose prices could be recovered, including 413 companies that later delisted, though 137 earlier members, mostly gone before 2010, are missing. No delisting return is added. The 740 companies outside the index were chosen from those listed when the universe was built, so smaller companies that failed or were acquired earlier are under-represented; lead results are therefore re-measured on each date's S&P 500 members.

FAQ · Survivorship bias

What is survivorship bias in backtesting?

Reviewed against the platform's code on Sep 26, 2026

Survivorship bias is the error of studying only the companies, funds or strategies that still exist today. Because the failures — bankruptcies, forced delistings, closed funds — have dropped out of the sample, past returns look higher, and risk lower, than investors at the time actually experienced.

Why it matters

A stock screen run on today's index members can only pick companies already known to have survived, so it silently leaves out every company that went bankrupt, was taken over or shrank out of the index. Studies of mutual funds found that dropping funds that closed overstates average returns (Elton, Gruber & Blake, 1996) and can even create the appearance of persistent skill (Brown, Goetzmann, Ibbotson & Ross, 1992).

How it works

The fix is to test on the universe as it stood on each past date: keep delisted securities with their full price history, take index membership from historical records rather than today's list, and include the return on the delisting itself. Databases often omit that return for companies removed for poor performance, where it is typically a large loss (Shumway, 1997). Point-in-time membership lists and delisting codes are the standard tools.

How Opulence Alpha applies it

The Opulence Alpha studies use one fixed universe of 1,767 US stocks. It holds every company in the S&P 500 since 1996 whose prices could be recovered, 1,027 in all, 413 of them since delisted, with full history; delisted companies supply 17.6% of the sample's stock-weeks. Three gaps are stated: 137 earlier members, mostly gone before 2010, could not be recovered; the 740 non-index companies were chosen from those listed at construction, so lead results are re-measured on each date's S&P 500 members; and no delisting return is added.

The universe, delisted companies included →

Related concepts

Point-in-time dataPoint-in-time data records what was known on each date, filing dates and revisions included, so a backtest cannot use the future. Look-ahead bias, explained.Look-ahead biasLook-ahead bias: a backtest using information not yet available on the decision date. Where it comes from, how to prevent it, and how Opulence Alpha avoids it.Backtest overfittingBacktest overfitting is tuning a strategy to noise in past data: the backtest looks strong, live results decay. How to detect it and how to avoid it.

All concepts →Sector behaviour in every market regime →

Questions

What is an example of survivorship bias in investing?: Backtesting a strategy on today's S&P 500 members. Those companies are in the index because they grew and survived; the ones that went bankrupt, were acquired or shrank out of the index are missing. Any strategy tested on that list inherits a head start no investor in the past could have had, because nobody knew then which companies would survive.
How do you avoid survivorship bias in a backtest?: Use the universe as it stood on each test date: keep delisted companies with their full price history, take index membership from historical records rather than today's list, and include the return on the day a company delists. Then check whether a result also holds on a subset defined only by what was known on each date, such as that date's index members.
Is the Opulence Alpha universe free of survivorship bias?: Not entirely. Its S&P 500 layer largely is: it holds every member since 1996 whose prices could be recovered, including 413 companies that later delisted, though 137 earlier members, mostly gone before 2010, are missing. No delisting return is added. The 740 companies outside the index were chosen from those listed when the universe was built, so smaller companies that failed or were acquired earlier are under-represented; lead results are therefore re-measured on each date's S&P 500 members.

References

Brown, S. J., Goetzmann, W., Ibbotson, R. G. & Ross, S. A. (1992). Survivorship Bias in Performance Studies. Review of Financial Studies 5(4).
Elton, E. J., Gruber, M. J. & Blake, C. R. (1996). Survivor Bias and Mutual Fund Performance. Review of Financial Studies 9(4).
Shumway, T. (1997). The Delisting Bias in CRSP Data. Journal of Finance 52(1).

Educational content about research methods. Not investment advice.