Survivorship bias is the logical error of drawing conclusions from a population that has already been filtered for success — looking only at the people, companies, or strategies that survived a given selection process and ignoring the larger set that did not.
The clearest illustration is statistician Abraham Wald's wartime analysis of bomber armor: the U.S. military, observing where returning planes had been hit, proposed reinforcing those areas. Wald argued the opposite — the planes that returned were the ones that could take hits in those places. The armor needed to go where the returning planes were not damaged, because the planes hit there had been shot down and never returned to be counted.
The bias underwrites a great deal of bad strategic advice in business, investing, and personal development. Books, talks, and case studies tend to over-sample from successful outliers and produce confident generalizations about behaviors common to "what works," when many of those same behaviors are equally common to ventures that failed. The fix is structural: include the failures, or at minimum acknowledge that the visible sample has been selected by the very outcome the conclusions are about.