Survivorship bias

Survivorship bias is a logical error that occurs when a person focuses on the people or things that survived some process and inadvertently overlooks those that did not because of their lack of visibility. This bias can lead to overly optimistic beliefs because failures are ignored, such as when companies that have gone bankrupt are excluded from analyses of financial performance. It is a form of selection bias.

Overview

Survivorship bias can lead to a significantly distorted view of reality, by making things seem better than they actually are. The term is derived from during World War II, when analysts only examined the damage on returning aircraft, neglecting the fact that planes that had been shot down were not available for assessment. This oversight led to a misunderstanding of which parts of the aircraft needed additional armor. The bias is not limited to wartime analytics but can be found in many areas of life and study, including in the fields of finance, psychology, and history.

Examples

Finance

In finance, survivorship bias is a common problem when analyzing the performance of stocks, mutual funds, or other financial instruments. By only considering those funds that are currently active, analysts may ignore those that have failed and closed, leading to an overestimation of overall performance.

History

In history, survivorship bias can lead to a skewed understanding of historical events or figures, focusing only on those that have "survived" in historical records while ignoring those that have not been documented.

Psychology

In psychology, survivorship bias can affect the way people assess their own chances of success. For example, individuals might overestimate their chances of success in an industry by looking only at the most successful people, without considering the many more who attempted and failed to achieve similar success.

Mitigation

To mitigate survivorship bias, it is important to try to account for all relevant examples, including those that did not survive or succeed. This may involve looking for hidden or overlooked data, considering alternative sources of information, or statistically adjusting for missing data.

See also

• Selection bias
• Confirmation bias
• Availability heuristic