Glossary
Deflated Sharpe ratio
The probability that a strategy's true Sharpe ratio is above what the best of N unskilled trials is expected to show. It corrects a result for selection, short samples and non-normal returns (Bailey and López de Prado, 2014).
The deflated Sharpe ratio is a probabilistic Sharpe ratio whose benchmark is the expected best Sharpe ratio of N independent trials with no skill. It needs the number of trials, the sample length, and the skewness and kurtosis of returns.
A value near 1 says the result is unlikely to be selection alone; a value near 0.5 says it is about what searching would produce.
Example: an annual Sharpe ratio of 1.2 over 60 months, skew -0.5 and raw kurtosis 5 (a normal distribution has 3). Against zero the probability is 99.0%. As the best of 100 trials, the benchmark is 1.13 and the deflated value is 55.3%.
Sources
- The Deflated Sharpe Ratio: Correcting for Selection Bias, Backtest Overfitting, and Non-Normality. The Journal of Portfolio Management 40(5), 94-107 (David H. Bailey, Marcos López de Prado), 2014. Accessed 9 Nov 2026