Risk-Adjusted Returns Calculator

Five from the methodology page: Sharpe (return per total volatility), Sortino (return per downside volatility), Calmar (return per max drawdown), Information Ratio (excess return per tracking-error vs. benchmark), and Treynor (return per beta). All annualized.

When the strategy has asymmetric returns. Sortino penalizes downside volatility only; Sharpe penalizes both up and down volatility equally. For long-only equity, Sharpe and Sortino track closely. For strategies with positive skew (trend-following, options buying), Sortino is more representative.

Whatever benchmark you'd hold if you didn't have the strategy. For US equity strategies, SPY is typical. For multi-asset, a 60/40 stock/bond mix. For style-specific (small-cap value), a style-matched index (IWM, VTV). Information Ratio against a mismatched benchmark is meaningless.

At least 36 monthly observations for Sharpe to be statistically meaningful (2 sigma SE). For Calmar, you need at least one full drawdown-recovery cycle (typically 3-7 years). Below those thresholds, the metrics are descriptive but not robust. The tool flags low-N estimates.

Yes. A strategy with rare large losses (option selling, short-vol) can show high Sharpe in calm regimes and catastrophic Sharpe when the regime breaks. The methodology page warns about this and recommends pairing risk-adjusted metrics with explicit tail-risk measures (VaR, CVaR, drawdown duration).