Look-Ahead Bias
A backtesting flaw where a model is accidentally fed information that would not have been known at the time of the prediction, inflating apparent performance.
Look-ahead bias occurs when a backtest uses data that would not have been available at the moment a trade was simulated. Common culprits include using a stock's closing price to make a decision timestamped before the close, using restated fundamentals instead of the originally reported numbers, or normalising features using statistics computed over the entire dataset (including the future).
The result is a backtest that looks far more profitable than anything achievable in live trading. Because the leak is often subtle, look-ahead bias is one of the most common reasons quant strategies fail in production after promising research results.
endeavr.ai's pipeline enforces point-in-time correctness: features are computed only from data available at the prediction timestamp, and walk-forward windows never overlap into the future. This is what makes the published accuracy numbers trustworthy.