Survivorship Bias, explained.
Survivorship bias is a selection error: drawing conclusions from cases that remain visible while overlooking cases removed by failure, exit, or filtering.
Why it happens
The visible group answers a question about survivors. It may not answer the question you intended about everyone who began. The missing cases can change both the average result and the apparent causes of success.
Survivorship bias arises when we draw conclusions from the cases that remain visible while overlooking those that disappeared or were filtered out.
Look for this pattern
Identify what allowed a case into the sample and what removed others. Reconstruct the starting population before treating a common feature of successful cases as a recipe for success.
A worked example
Studying successful businesses
You interview thriving firms and find that many took ambitious risks.
Firms that took similar risks and closed down are absent from the interviews.
Compare success and failure among all firms that took the risk before claiming the behavior caused success.
Where this idea is useful
Studying successful organizations alone cannot tell us whether their common habits caused success.
A common misconception
“A pattern among winners explains how to win.”
It could also be common among failures. A pattern's predictive value requires information about the comparison group.
What this explanation leaves out
- Not every selected sample creates the same distortion. The selection process and the missing cases determine the direction and size of the bias.
Is survivorship bias always an overly optimistic estimate?
Often, but not necessarily. The selection process determines the direction of the distortion. The central issue is that omitted cases differ in ways relevant to the conclusion.
Who is missing from the evidence you can see, and why did they disappear?
Associated thinkers
Associations marked provisional are awaiting source review.