Deflated Sharpe Ratio calculator
You tested ten variants and are showing your best Sharpe. So does everyone. The Deflated Sharpe Ratio (Bailey & López de Prado, 2016) asks the only question that matters: what is the probability your Sharpe beats the best of N lucky draws?
How it works
Testing N variants and reporting the best inflates the expected maximum Sharpe even when every variant is pure noise. The DSR converts your annualized Sharpe to per-observation units, computes the expected maximum Sharpe of N null strategies at your sample size, and returns the probability that your observed Sharpe exceeds it - adjusted for the skewness and kurtosis of your returns.
A DSR above 0.95 is the conventional bar for "probably not selection luck." Most backtests never state N at all, which is why we ask for it on every audit.
Reference: Bailey, D. & López de Prado, M. (2016), "The Deflated Sharpe Ratio: Correcting for Selection Bias, Backtest Overfitting and Non-Normality."