Portfolio Performance Metrics Cheat Sheet
Return-Based Metrics
| Metric | Formula | What It Tells You |
|---|---|---|
| Total Return | (End Value − Start Value + Dividends) ÷ Start Value | Overall gain/loss including income |
| Annualized Return (CAGR) | (End Value ÷ Start Value)^(1/Years) − 1 | Smoothed annual growth rate |
| Time-Weighted Return (TWR) | Geometric linking of sub-period returns | Manager performance (removes cash flow effects) |
| Money-Weighted Return (IRR) | Rate that sets NPV of all cash flows = 0 | Investor’s actual experience (includes timing of flows) |
| Excess Return | Portfolio Return − Benchmark Return | Value added above the benchmark |
Risk-Adjusted Return Metrics
These are the metrics that actually matter for comparing portfolio managers and strategies.
| Metric | Formula | Interpretation | Good Value |
|---|---|---|---|
| Sharpe Ratio | (Rp − Rf) ÷ σp | Excess return per unit of total risk | > 1.0 is good; > 2.0 is excellent |
| Sortino Ratio | (Rp − Rf) ÷ Downside Deviation | Excess return per unit of downside risk only | > 1.5 is good; > 3.0 is excellent |
| Treynor Ratio | (Rp − Rf) ÷ βp | Excess return per unit of systematic risk | Higher = better (compare to peers) |
| Information Ratio | (Rp − Rb) ÷ Tracking Error | Active return per unit of active risk | > 0.5 is good; > 1.0 is exceptional |
| Jensen’s Alpha | Rp − [Rf + β(Rm − Rf)] | Return above what CAPM predicts | Positive = outperformance |
| Calmar Ratio | Annualized Return ÷ Max Drawdown | Return per unit of worst-case loss | > 1.0 is solid; > 3.0 is strong |
Risk Metrics
| Metric | Formula | What It Measures |
|---|---|---|
| Standard Deviation | √[Σ(Ri − R̄)² ÷ (n−1)] | Total volatility — how much returns vary |
| Beta | Cov(Rp, Rm) ÷ Var(Rm) | Sensitivity to market movements |
| Max Drawdown | (Trough − Peak) ÷ Peak | Worst peak-to-trough decline |
| Tracking Error | Std Dev of (Rp − Rb) | Consistency of active returns vs. benchmark |
| R-Squared | Correlation² of portfolio vs. benchmark | How much of portfolio variance is explained by the benchmark |
| Value at Risk (VaR) | Maximum loss at a confidence level over a period | Worst expected loss (e.g., 95% confidence, 1-day) |
Metric Selection Guide
| Situation | Best Metric | Why |
|---|---|---|
| Comparing two fund managers | Sharpe Ratio | Universal risk-adjusted return comparison |
| Evaluating downside protection | Sortino Ratio + Max Drawdown | Focuses specifically on harmful volatility |
| Measuring active management skill | Information Ratio + Alpha | Shows value added vs. benchmark per unit of active risk |
| Assessing portfolio risk level | Beta + Standard Deviation | Beta for market risk; StdDev for total risk |
| Reporting to clients | CAGR + Max Drawdown + Sharpe | Clear, intuitive metrics everyone understands |
Key Takeaways
- Risk-adjusted metrics (Sharpe, Sortino, Information Ratio) matter more than raw returns
- Sharpe Ratio above 1.0 and Sortino above 1.5 indicate strong risk-adjusted performance
- Max Drawdown shows the worst-case scenario — essential for understanding real risk
- Use TWR for evaluating managers; use IRR for evaluating your actual investment experience
- Evaluate metrics over at least one full market cycle for meaningful conclusions
Frequently Asked Questions
What is a good Sharpe ratio for a portfolio?
A Sharpe ratio above 1.0 is generally considered good for a diversified portfolio. Above 2.0 is excellent. The S&P 500’s long-term Sharpe ratio is roughly 0.4–0.5. Most actively managed funds struggle to consistently exceed 1.0.
What’s the difference between alpha and excess return?
Excess return is simply portfolio return minus benchmark return. Alpha (Jensen’s Alpha) adjusts for the risk taken — it’s the return above what CAPM would predict given the portfolio’s beta. A portfolio can have positive excess return but negative alpha if it took too much risk.
Why do institutional investors prefer time-weighted returns?
Time-weighted return (TWR) removes the impact of cash inflows and outflows, isolating the manager’s investment decisions. Since the manager doesn’t control when clients add or withdraw money, TWR is the fairest measure of investment skill.
How is max drawdown different from standard deviation?
Standard deviation measures average variability around the mean in both directions. Max drawdown measures the single worst peak-to-trough decline. A portfolio can have low standard deviation but still suffer a brutal drawdown during a crash.
Can I use these metrics for individual stocks?
Yes, but they’re most meaningful for portfolios. Individual stocks have much higher volatility and idiosyncratic risk, which makes metrics like Sharpe ratio less stable. For single stocks, beta and max drawdown are the most useful.