Look-ahead Bias

Accidentally using information in a backtest that wouldn’t have been available at the time.

1 min

Why it matters

It creates fake profitability.

Decision use: treat this term as a risk input. If you can’t observe it, you can’t size for it.

Common trap

Using revised data or future-knowledge signals unknowingly.

Example

Using final CPI values in a backtest where only the initial print was known.

Operational cue: if you can’t point to it on the chart, in the DOM, or in your broker logs, treat it as a story and trade smaller.

Notes

Look-ahead bias is silent. You must design tests to avoid it.

If you see this repeatedly, you are in a different regime than your model assumes.
Regime change usually means position size change.
When in doubt: trade smaller, or don’t trade.