Glossary

    Glossary

    Survivorship Bias

    Survivorship bias is judging performance from the survivors while the failures have been deleted from the sample: the funds that closed, the stocks that were delisted, the strategies that blew up. The graveyard is invisible, which is why the average fund looks better than the options you can actually buy.

    BehaviouralLast reviewed 2026-10-08

    What it is

    Survivorship bias is the error of drawing conclusions from the sample that survived while ignoring the sample that did not. In markets the failures are deleted systematically: closed funds vanish from databases, delisted stocks vanish from indexes, and the strategies that blew up vanish from the backtests. What remains is a roster of winners being measured as if it were the full population.

    The US numbers are material. Roughly 5% to 7% of active mutual funds close in a typical year, and a fund database that scrubs them overstates the average active fund's return by roughly half a point to a point a year. The index reconstitution machine does the same work at the stock level: the S&P 500 adds winners and removes the companies that stopped qualifying, so today's index holds yesterday's successes by construction.

    Key takeaways

    • Survivorship bias measures only the winners; the closed funds and delisted stocks are deleted from the sample.
    • Roughly 5% to 7% of active funds close in a typical year, and the databases scrub them out.
    • Scrubbing closures overstates the average active fund's return by roughly half a point to a point a year.
    • Index reconstitution adds winners and removes failures, so today's index holds yesterday's successes by construction.
    • The fix: use total-market and index comparisons, and ask any backtest what happened to the failures.

    How it works

    The mechanism is a missing denominator. Every statistic computed on survivors alone (average fund return, average stock return, average startup outcome) excludes the entries that failed and left the sample, and the excluded entries are systematically the worst performers. A fund database that removes closed funds keeps only the ones that lived long enough to be measured; a backtest of a stock screen run on today's tickers silently excludes every company that was delisted, acquired at a loss, or bankrupted along the way. The result is an average computed on a roster pre-sorted for success.

    The fund version has a number. Roughly 5% to 7% of active US funds close in a typical year (mergers and liquidations), and the cumulative effect over a decade removes roughly a third of the starting roster. Studies that re-add the closed funds find the average active fund's return falls by roughly 0.5% to 1% a year, which compounds to a material gap over a career: the database version of the active-management industry is roughly a point a year better than the actual one.

    The index version runs in the winners' favor. The S&P 500 adds companies as they grow into the qualification criteria and removes them when they fall out, so the index continuously promotes yesterday's winners; the Dow's roster of forever companies has lost Kodak, Sears, and GE along the way. Holding the index is the only strategy that automatically inherits the survivors without having to pick them.

    Why it matters

    Survivorship bias inflates every expectation built on survivors: the fund track record, the backtest, the newsletter picks, the friend's crypto wins. A backtest run on today's S&P 500 constituents is a test of the companies that survived the last twenty years, which is roughly like testing a parachute design on the people who landed. The SPIVA scorecards, which compare live funds against benchmarks each year, are the industry's own admission that the survivors are not the population.

    I watched a colleague named Brent (yes, that Brent) backtest a dividend screen on current tickers, find a 14% annual return since 2000, and size a portfolio to it. The screen had silently excluded every company that met the criteria and then went bankrupt; re-run with the full historical universe, the return fell to roughly 9%. The 5-point gap was the graveyard, and it was invisible in the tool.

    The highest-leverage thing most Americans can do is prefer total-market index comparisons (which carry their dead with them) and ask one question of any backtest: what happened to the failures? The index-fund entry has Bogle's arithmetic on why this matters, and the compare investment fees calculator runs the survivorship-adjusted comparison before you size anything.

    Real-world examples

    The fund database, one decade

    A fund database lists 1,000 active funds in 2015. Roughly a third of them close over the following decade (mergers and liquidations), and the database version of the average fund return since 2015 is computed on the 670 survivors. Re-adding the closed funds (whose final years were, by definition, bad) drops the average active fund's return by roughly 0.5% to 1% a year. Over the decade on a $100,000 investment, that is roughly $60,000 to $120,000 of reported performance that the survivors-only version never had to earn.

    The backtest on today's tickers

    A screen run on today's S&P 500 constituents since 2000 shows a 14% annual return: the test population is the set of companies that survived the dot-com crash, the financial crisis, and the pandemic. Re-run on the full historical universe including the companies that were removed (WorldCom-era names, Lehman-era names, and the rest), the return falls to roughly 9%. The 5-point gap is the dead companies, deleted by construction. Testing a strategy on survivors is testing the parachute on the people who landed.

    The index that carries its dead

    The S&P 500 added the growth winners as they qualified and removed the failures along the way; the Dow 30 has replaced Kodak, Sears, and General Electric over the decades. An investor who bought the index inherited the survivors automatically, with no research required and the failures removed by committee. An investor who picked forever companies individually inherited the same deletions without the committee, which is the difference between the machine and the memory. The reconstitution is the feature that looks like a bug until the graveyard is priced.

    Common mistakes

    Mistake: Trusting fund track records and database averages at face value.

    Why it bites you: The databases scrub closed funds, overstating the average active fund by roughly half a point to a point a year. The graveyard never files a return.

    Mistake: Backtesting a strategy on today's tickers.

    Why it bites you: The test population is the survivors, and the delisted failures are excluded by construction. Brent's 14% backtest fell to 9% once the dead companies were re-added.

    Mistake: Copying the visible winners (newsletters, influencers, a friend's picks).

    Why it bites you: The sample of people still talking about their wins excludes the ones who stopped. The survivor's strategy is measured after the failure rate has already done its work.

    Mistake: Assuming the index's past constituents were today's.

    Why it bites you: Reconstitution continuously swaps failures for winners, so historical index returns include deletions a cherry-picked list never had. The index carries its dead; the personal list does not.

    Ivy's take

    I spent an afternoon with the fund-closure data and a decade of SPIVA scorecards so you do not have to. The graveyard is the most underpriced dataset in finance: every average you have ever been sold is computed on the survivors, and the survivors are pre-sorted for success. My two questions for any performance claim are now mechanical: where are the failures, and who stopped talking? A backtest on today's tickers tested the parachute on the people who landed.

    The index is the only strategy that carries its dead with it, which is a feature dressed as a bug. Brent backtested a screen to 14% and sized a portfolio on the ghosts. The ghosts did not invest.

    Related jargon, defined

    Fund closure
    A merger or liquidation that removes a fund from databases; roughly 5% to 7% of funds a year.
    Index reconstitution
    The periodic swap of index constituents, adding qualifiers and removing failures.
    Backtest
    A simulated historical performance test; biased when run on survivors only.
    SPIVA
    S&P's scorecards comparing live active funds against their benchmarks after fees.
    Total-market index
    An index holding essentially the whole market, which carries its failures as they decline.
    Selection bias
    The broader error of sampling non-randomly; survivorship bias is its market version.

    Frequently asked questions

    What is survivorship bias in investing?+

    It is drawing conclusions from the funds, stocks, or strategies that survived while ignoring the failures deleted from the sample. The result is an average computed on a pre-sorted roster of winners.

    How much does it inflate fund performance?+

    Studies that re-add closed funds find the average active fund's return drops by roughly 0.5% to 1% a year. Over a decade on $100,000, that is a six-figure difference in reported performance.

    Why does the S&P 500 benefit from survivorship bias?+

    Reconstitution adds companies as they grow into the criteria and removes them when they fall out, so the index continuously promotes winners. Buying the index inherits the survivors without having to pick them.

    How do I avoid survivorship bias in a backtest?+

    Run it on the full historical universe, including companies that were later delisted or removed, and ask what happened to the failures. A screen on today's tickers is a test of the survivors.

    Does survivorship bias affect stock picking?+

    Yes: the visible winners (the stocks everyone discusses) are the ones that survived, and the delisted failures are out of the conversation. The friend's picks are measured after the graveyard has done its work.

    Disclaimer: This information is general in nature and does not constitute financial or legal advice. Always consult a qualified professional for your specific situation.

    Keep learning

    Ivy Sinclair-Wren

    Ivy Sinclair-Wren

    Financial Chaos Analyst

    Connect on LinkedIn

    Ivy Sinclair-Wren is a Financial Chaos Analyst covering investing, AI, wealth psychology, and the emotional consequences of opening finance apps during market crashes. Based in Melbourne, she specializes in demystifying the US tax code and helping users navigate the intersection of spreadsheet logic and human irrationality.