Risk Metrics
Risk metrics quantify how much an investment's value might swing or fall, and how much return was earned relative to that risk — essential context alongside any return figure on its own.
Beta, β
Covariance of the stock's returns with the benchmark ÷ Variance of the benchmark's returns
Shows how sensitive a stock is to overall market movements. A beta above 1 means larger swings than the market; below 1 means smaller swings.
Strengths: Useful for managing a portfolio's overall sensitivity to market moves.
Watch out for: A statistical measure based on past data — the relationship may not hold in the future, and it says nothing about company-specific risk.
Combine with: Alpha and volatility.
Volatility / Standard Deviation
Standard deviation of returns over a period (often annualized)
The most basic statistical measure of how much price swings around, up or down.
Strengths: An objective, comparable measure of risk.
Watch out for: Treats upside swings and downside swings as equally 'risky,' which doesn't match most investors' intuition.
Combine with: Sortino ratio (which isolates downside risk specifically).
Sharpe Ratio
(Portfolio return − Risk-free rate) ÷ Volatility (standard deviation)
The classic risk-adjusted return measure — how much excess return was earned per unit of risk taken.
Strengths: Allows fair comparison between investments or managers with different risk levels.
Watch out for: Treats upside and downside volatility equally, so it can understate results for investments whose returns are skewed favorably (more upside surprises than downside).
Combine with: Sortino ratio and maximum drawdown.
Sortino Ratio
(Portfolio return − Risk-free rate) ÷ Downside deviation (standard deviation of negative returns only)
An improved version of the Sharpe ratio that only treats downside volatility as 'risk.'
Strengths: Better matches the intuition that upside volatility is welcome, not a risk to penalize.
Watch out for: Less widely known than the Sharpe ratio, and the underlying data required is somewhat more involved.
Combine with: Sharpe ratio (compare both to see the gap) and maximum drawdown.
Value at Risk, VaR
Estimated statistically (e.g. from historical return distributions) as the maximum expected loss over a given period at a given confidence level (e.g. 95%).
Answers 'how much could I plausibly lose, and with what probability?' — a risk-management tool widely used by institutional investors.
Strengths: Expresses loss risk in a concrete, intuitive amount or percentage.
Watch out for: Doesn't capture 'tail risk' — losses beyond the chosen confidence level — a limitation exposed dramatically during the 2008 financial crisis.
Combine with: Maximum drawdown and stress testing.
Maximum Drawdown
The largest percentage decline from a peak to a subsequent trough over a given period
Shows the worst decline an investment has actually experienced historically — an intuitive, real-world risk measure.
Strengths: Grounded in actual experience, which makes it easy to compare against how much loss an investor could personally tolerate.
Watch out for: A historical figure — it doesn't guarantee an even larger drawdown won't happen in the future.
Combine with: VaR and the time it took to recover from the drawdown.
Alpha
Actual portfolio return − [Risk-free rate + β × (Market return − Risk-free rate)] (excess return based on the CAPM)
Shows the value added by a manager or strategy that can't be explained simply by exposure to the overall market (beta).
Strengths: Helps judge whether active management is genuinely adding value beyond the market.
Watch out for: Requires an accurate benchmark to calculate correctly, and results can shift depending on the time period and assumptions used.
Combine with: Beta and the information ratio.
Tracking Error
Standard deviation of the difference between portfolio returns and benchmark returns
Shows how much a portfolio's returns deviate from its benchmark index.
Strengths: Reveals how closely an index fund tracks its benchmark, or how distinctive an active manager's positioning is.
Watch out for: A high tracking error isn't inherently good or bad on its own — it's only meaningful alongside whether alpha was positive or negative.
Combine with: Alpha and the information ratio.
Correlation Coefficient
Covariance of two assets' returns ÷ (product of their standard deviations); ranges from −1 to 1
Shows how closely two stocks or asset classes move together.
Strengths: One of the most important inputs for judging diversification benefits — combining low-correlation assets tends to reduce overall portfolio risk.
Watch out for: Correlations aren't stable — they tend to spike higher across many assets during a market crash, exactly when diversification is needed most.
Combine with: Beta and volatility.