RISK METRICS
Returns without risk context are meaningless. Learn the metrics that matter.
A +40% return means nothing without knowing the risk taken to achieve it. Risk-adjusted metrics like Sharpe, Sortino, and Max Drawdown let you compare strategies on equal footing.
– RISK-ADJUSTED RETURNS
A risk-adjusted return measures how much return a strategy earned per unit of risk taken, not the raw number alone. A +40% year with deep drawdowns can be worse than a smooth +15% year, because the first is far harder to hold and to fund. Risk-adjusted metrics like Sharpe, Sortino and Calmar express that trade-off, so two strategies can finally be compared on equal footing rather than by headline return.
There is no single number for risk. The useful ones each protect against a different failure mode, which is why you read them together rather than picking one. For the exact formulas, see the performance metrics reference.
Full definitions and formulas: Performance Metrics reference.
Test whether your Sharpe is statistically real with the Deflated Sharpe Ratio calculator.
A risk-adjusted return measures return per unit of risk taken, rather than the raw return alone. A +40% year with deep drawdowns can be worse than a +15% year that was smooth. Ratios like Sharpe, Sortino and Calmar express this trade-off so two strategies can be compared on equal footing.
The Sortino ratio measures return per unit of downside deviation, the volatility of losses only. Unlike the Sharpe ratio, which penalizes all volatility, Sortino ignores upside swings, so it is more representative for strategies whose gains are lumpy but whose losses are controlled.
Downside risk is the part of volatility that comes from losses rather than gains. Total volatility treats a big up day and a big down day as equally bad; downside measures such as downside deviation, drawdown and VaR focus only on the harmful side, which is what actually threatens capital.
No single metric is enough. Sharpe and Sortino summarize risk-adjusted return, maximum drawdown shows the worst loss you would have endured, and VaR estimates day-to-day tail risk. Read them together: a high Sharpe paired with a brutal drawdown still describes a strategy that is hard to hold.
Every metric explained: Sharpe, Sortino, MaxDD, VaR, Calmar. What they measure and how to interpret them.
Is your Sharpe ratio statistically significant? Test for multiple-testing bias with the Bailey-Lopez de Prado method.
How likely is your strategy overfit? Estimate the probability that your backtest is just curve-fitting noise.
How AuditZK computes Sharpe, Sortino, MaxDD, and VaR from daily equity snapshots inside a hardware enclave.
NEXT STEP
Every metric explained here is computed automatically from your exchange data.