Updated 2026-07-25
Survivorship bias is the error of studying only the cases that made it to the end. Backtest a strategy across the coins listed today and you have quietly excluded every token that was delisted, collapsed or lost its market, which is precisely where the losses were. The result is a study of survivors presented as a study of the market.
Crypto is unusually exposed to this. Listings churn constantly, venues delist, and the public record of trader performance is filtered by who chose to keep posting. The fix is to reconstruct the universe as it existed at each point in time, including the assets that later disappeared, and to count the attempts that failed as well as the ones that worked. It sits alongside overfitting as a reason a strategy can look convincing on paper and still have no edge.