Learn

Common mistakes that make trading backtests look better than they are, and how to test for each one. Every code example runs on synthetic data, and every result shown is hypothetical.

1. Lookahead bias in pandas: 8 ways a backtest sees the future

   Eight common pandas patterns that leak future data into a backtest, each with a fix and a runnable test on synthetic data.

   5 min read

2. Survivorship bias in backtests: delisted stocks and reused tickers

   Why a backtest on today's index members is biased, how to build a point-in-time universe with delisted stocks, and how reused tickers break joins.

   5 min read

3. The deflated Sharpe ratio in Python: how many strategies did you try?

   What the deflated Sharpe ratio corrects for, a short Python implementation checked against the paper's worked example, and how to count your trials.

   5 min read

4. Point-in-time fundamentals: filing dates, acceptance times and restatements

   Why joining fundamentals by period end or filing date leaks into a daily backtest, what SEC acceptance times show, and how to do an as-of join in pandas.

   10 min read

Questions about our products are answered in the help articles.