Survivorship Bias

The error of drawing conclusions from the survivors of a selection process while ignoring those who didn't survive—producing a systematically distorted picture. Studying only successful companies to find 'the keys to success' ignores all the companies with the same keys that failed. WWII statisticians who studied returning bombers to find where to add armor were studying the WRONG planes—the planes that needed armor in those spots never came back. Survivorship bias is ubiquitous: motivational advice from successful people ignores the millions who followed the same advice and failed. Published research ignores null results. Visible buildings survive because they were well-built; we never see the equally well-built ones that burned down.

When to use it

When studying success factors based only on successful cases; when motivational advice seems to work for famous examples but not for you (the failures are invisible); when historical analysis is based on what survived to the present; when any selection process might be hiding the information that didn't pass through the filter.

How it can help

For any analysis of success: actively seek the non-survivors. When studying what makes companies successful: study failed companies with the same characteristics. When evaluating advice from successful people: find people who followed the same advice and didn't succeed. When analyzing historical data: ask 'what's missing from this dataset because it didn't survive to be included?' The correction is simple but requires effort: expand your sample to include failures, not just successes. Every 'secret of success' should be tested against the failure base.