Your Backtest Has a Sharpe Ratio. Does It Have a P-Value?
Most retail backtests stop at "here's my equity curve" and maybe a Sharpe ratio pulled from a half-remembered formula. That's not enough to know whether a strategy's edge is real or noise.
What a rigorous backtest report actually needs
- Sharpe ratio (correctly annualized)
- Sortino ratio (downside deviation only — very different picture on skewed strategies)
- Maximum drawdown
- Profit factor
- Win rate
- A p-value (Student's t-test) — this is the piece almost everyone skips, and it's the one that tells you whether your edge could just be luck
Automatic overfitting flags
The Quant Audit & Backtest Report Generator takes raw backtest output and produces all of the above automatically, with built-in overfitting warnings — so you catch a curve-fit strategy before you catch a real drawdown.
Requirements: Python 3.9+.
Related: if you need to sanity-check the CSV before it goes into the report generator (catch logic bugs like an impossible win rate or look-ahead bias), see the Backtest CSV Auditor: https://theglitchlist.com/product/backtest-csv-auditor/?utm_source=devto&utm_medium=article&utm_campaign=n07_quant_audit
Top comments (0)