Survivorship Bias

Survivorship Bias in Stock Market Analysis

If you create a stock scanner today and run a historical backtest going back 10 years using a current list of Nifty 50 or small-cap stocks, your backtest results will look phenomenally profitable. But your data is flawed due to survivorship bias.

Why the backtest lies to you:

Your data matrix is only testing companies that are still alive and healthy in 2026. It completely omits all the companies that were part of that index 10 years ago but went bankrupt, defaulted, got delisted, or collapsed along the way (e.g., companies like Kingfisher Airlines or DHFL from past eras).

By removing the "dead bodies" from your historical data, your code is testing a pristine, artificial environment.