Noon Barbari

Validation & overfitting

What is the deflated Sharpe ratio?

The deflated Sharpe ratio (Bailey & López de Prado, 2014) answers: given how many strategy variants were tried, how likely is it that this Sharpe ratio is just the luckiest draw from zero-skill noise? It penalizes the observed Sharpe for the number of trials, the track record's length, and the return distribution's skew and fat tails.

It matters because optimizers try thousands of combinations — the best of 1,000 random strategies has an impressive Sharpe by construction. A deflated Sharpe probability above ~95% says the result would be unusual even accounting for the search; below that, the "edge" is statistically indistinguishable from selection luck.

Deflated Sharpe calculatorSharpe ratio (glossary)

The fastest answer is a test

Most 'does X work?' questions take a minute to answer empirically — on real data, with an out-of-sample check, free.

Backtest a strategy free

More in this topic

Educational content, not financial advice. Backtested and historical figures describe past periods only; past performance does not guarantee future results.