Why Backtested Strategy Returns Rarely Survive Contact With Live Markets

Backtesting is a useful research tool, but it is easy to unintentionally overfit a strategy to historical data — tuning parameters until they perform well specifically on the period being tested, at the cost of generalizing to new market conditions.

Slippage and execution assumptions are another common gap. A backtest that assumes fills at the exact quoted price ignores the reality that larger or faster orders move the market, especially in less liquid instruments.

Survivorship bias can also inflate results when a backtest only includes assets that still exist today, quietly excluding the ones that were delisted or collapsed during the test period.

None of this means automated or rules-based strategies are unusable — it means backtested performance should be treated as one data point among several, alongside forward (paper-trading) testing and a clear understanding of the strategy's underlying logic.