Sharpe, Sortino, and Information Ratio

Three risk-adjusted return measures sit on every quant's dashboard.

Sharpe ratio is excess return per unit of total volatility:

Sortino is similar but only penalizes downside volatility:

It's preferred when returns are heavily skewed and upside variance shouldn't be punished.

Information ratio (IR) measures active return per unit of tracking error:

where $R_b$ is the benchmark. It's the right measure when you're judged on outperforming an index.

A daily Sharpe of $1$ annualizes to $\sqrt{252} \approx 16$. Annualizing also magnifies the noise in your estimate — small samples produce wildly variable Sharpe estimates, so confidence intervals matter.