Survivorship Bias: Meaning, Examples and the WWII Planes Story

By the IQ Revealed Team · Updated November 4, 2026

Survivorship bias is the error of judging a whole group by the members that made it through some filter, while the ones that failed are out of sight. The funds still open, the companies still trading, the buildings still standing and the studies that got published all look better than the full set they came from. Also called survivor bias, it is a form of selection bias that often works like one of the cognitive biases: the evidence in front of you feels complete when it is not.

Survivorship bias at a glance

What happens: a selection process removes some cases, and conclusions are drawn from the cases that remain.
Classic example: Abraham Wald's Second World War method for estimating aircraft vulnerability from the damage on planes that came back.
Other names: survivor bias, survival bias; a special case of selection bias.
Where it bites: investing, business advice, medicine, history and published research.
The fix: find the full starting group and ask what happened to the cases that dropped out.

The Second World War planes: what Abraham Wald actually did

Abraham Wald was born in 1902 in Kolozsvár, now Cluj-Napoca in Romania, earned his doctorate in mathematics at the University of Vienna in 1931 and left Austria for the United States in 1938. During the war he worked with the Statistical Research Group at Columbia University, a team of statisticians solving problems for the US military, whose history W. Allen Wallis later described in 1980.

One of those problems was aircraft vulnerability. Wald developed a method for estimating, from the damage seen on aircraft that survived, how likely a hit to each part of a plane was to bring it down. The core idea is the survivorship insight: planes hit in places that are usually fatal do not come back to be counted, so areas that rarely show damage on returning planes can be exactly the areas that need protection. According to Marc Mangel and Francisco Samaniego, who condensed and re-examined the work in 1984, the memoranda were used in the Second World War and later in Korea and Vietnam, and were reissued by the Center for Naval Analyses.

The version that circulates online, in which officers propose armouring the parts with the most bullet holes and Wald tells them to armour the parts with none, is a simplification. Wald's memoranda are technical statistics, not a story about one meeting, and the widely shared drawing of a bomber covered in red dots is a modern illustration, not a figure from his work. The lesson survives the simplification: the missing planes carry the information.

Examples of survivorship bias

Investment funds. Funds that close or merge are mostly the poor performers, so a database of surviving funds overstates what investors earned. Edwin Elton, Martin Gruber and Christopher Blake tracked every fund that existed at the end of 1976, including the returns of funds that later merged, to measure the bias precisely. Burton Malkiel, using data on all equity funds existing each year from 1971 to 1991, found that survivorship bias was more important than earlier studies had estimated.
Spurious winning streaks. Stephen Brown, William Goetzmann, Roger Ibbotson and Stephen Ross showed in 1992 that dropping the losers from a sample affects not only average performance but also the relative rankings of managers over time, so it can make skill look more persistent than it is.
Start-up success stories. Famous founders who left university are visible; the far larger number who left and failed are not. In Eurostat's figures for firms born in 2015, the share still active five years later ranged from about a third in Portugal and Norway to about two thirds in the Netherlands.
“They don't make them like they used to.” Old houses, furniture and tools that are still around are the ones that were built to last; the flimsy ones were thrown away long ago.
Published research. Robert Rosenthal named the “file drawer problem” in 1979: if studies with non-significant results stay unpublished, the journals hold a biased sample of the research that was actually done.
Medical studies. A related trap, immortal time bias, was first identified in the 1970s in studies of heart transplant survival. Samy Suissa showed in 2008 that observational studies can make drugs look far more effective than they are when patients must survive long enough to be counted as treated.

Why the error is so easy to make

Survivors are what we see, talk about and write about. Failures leave fewer records, fewer books and fewer interviews. That makes survivorship bias a close cousin of the availability heuristic: the examples that come to mind are the ones that survived to be remembered. And once a success story is known, hindsight bias makes the path to it look inevitable, which hides how many people took the same path and failed.

Survivorship bias and intelligence

Lists of the habits of geniuses, the study routines of top students or the reading lists of billionaires are classic survivor samples. They describe people who succeeded, but without the people who had the same habits and did not succeed, they cannot show that the habits caused anything. The same caution applies to claims about raising your IQ: a method promoted by its success stories, with no record of how everyone who tried it fared, has not been tested.

Being clever does not make you immune. Spotting survivorship bias is less about raw reasoning power and more about a habit of asking what the data leave out, the kind of question our pages on critical thinking and confirmation bias cover.

How to avoid survivorship bias

Find the starting line. Before judging the winners, count everyone who started: every fund launched, firm founded, patient enrolled or study run.
Ask where the failures went. Were they closed, merged, deleted, never published or never asked? The reason they left often explains the bias.
Compare survivors with non-survivors. A trait only matters if it is more common among those who made it than among those who did not.
Be wary of advice from winners only. Success stories are real, but on their own they cannot tell you the odds.

Trust and scope notes

This page summarises published research and official statistics for general information. It is not investment, business or medical advice. The IQ Revealed test is an online reasoning test of 40 original items for adults aged 18 and over; it does not measure survivorship bias or any other thinking bias. Our guide to how accurate IQ tests are explains what an online test can and cannot tell you.

IQ Revealed is not affiliated with, endorsed by or connected to the researchers, universities, journals, publishers, statistical offices or research organisations named on this page. Names are used only to identify the work being described.

Frequently asked questions

What is survivorship bias?＋

Survivorship bias, also called survivor bias, is the error of drawing conclusions only from the people, companies or things that made it through some selection process, while ignoring the ones that did not. Because the failures have dropped out of view, what remains looks more successful, more durable or more consistent than the whole group really was.

What is the survivorship bias plane story?＋

During the Second World War the statistician Abraham Wald, working for the Statistical Research Group at Columbia University, developed a method for estimating how vulnerable aircraft were, using the damage on planes that came back. The key insight is that hits on returning planes show where a plane can be hit and still return; areas that rarely show damage on survivors may be the ones where a hit brings a plane down. The popular retelling is a simplification of a set of technical memoranda, and the well-known drawing of a bomber covered in red dots is a modern illustration, not a figure from Wald's work.

What are examples of survivorship bias?＋

Common examples include judging investment funds only by those still open, since funds that close tend to be the poor performers; studying successful college dropouts and ignoring the many whose businesses failed; believing old buildings or tools were better made because the badly made ones have long since gone; and reading only the studies that were published, when studies with null results often stay in the file drawer.

Who discovered survivorship bias?＋

No single person discovered it; it is a form of selection bias that statisticians have long recognised. Abraham Wald's wartime work on aircraft survivability is the best-known early example. In finance it was studied in detail in papers such as Stephen Brown, William Goetzmann, Roger Ibbotson and Stephen Ross's 1992 Survivorship Bias in Performance Studies.

Is survivorship bias a cognitive bias?＋

It is usually classed as a selection bias, a flaw in which data you look at, rather than a quirk of perception. It becomes a thinking error when people treat the visible survivors as if they were the whole group. That is why it often appears alongside cognitive biases such as hindsight bias and the availability heuristic.

How does survivorship bias affect investing?＋

If a database keeps only funds that still exist, the average return looks better than investors actually earned, because funds that closed or merged were mostly the weaker ones. Edwin Elton, Martin Gruber and Christopher Blake tracked every fund that existed at the end of 1976 to measure this precisely, and Burton Malkiel concluded in 1995 that survivorship bias was more important than earlier studies had estimated.

How can you avoid survivorship bias?＋

Ask what is missing from the data and why. Find the full starting group, not just those still standing: every fund launched, every company founded, every patient who started a treatment, every study run. Compare survivors with non-survivors on the factor you care about, and be wary of advice drawn only from the winners.