Portfolio analysis · Downside risk measurement
Sortino Ratio: Measuring Risk-Adjusted Return Without Penalizing Gains
The Sharpe ratio treats every swing in your portfolio the same way — a month of unexpected gains gets penalized just like a month of unexpected losses. The Sortino ratio fixes this by only counting the volatility that actually hurts investors: returns that fall below your target. This guide explains the formula, works through a step-by-step calculation, compares Sortino to Sharpe, and shows you what ratio levels signal strong risk-adjusted performance.
Sortino Ratio = (Portfolio Return - Risk-Free Rate) / Downside Deviation. It measures excess return per unit of harmful downside risk only.Developed by Frank Sortino in the 1980s, it improves on the Sharpe ratio by penalizing only downside volatility — the harmful kind — rather than all return variability.
The Sortino Ratio Formula
The formula has three components. The key difference from the Sharpe ratio is in the denominator: instead of total standard deviation, it uses downside deviation only.
- Rp (Portfolio Return)
- The total return of the portfolio over the measurement period. Use the same time horizon consistently — monthly or annual.
- Rf (Risk-Free Rate / MAR)
- The minimum acceptable return (MAR) — the threshold below which a return is considered "bad." Commonly set to 0% (any loss is bad), the 3-month T-bill yield, or inflation. The choice of MAR directly affects the calculated ratio.
- σd (Downside Deviation)
- The standard deviation of returns that fall below the MAR only. For each period: if return < MAR, the difference is squared and included; if return ≥ MAR, it contributes zero. Take the square root of the average of all squared shortfalls (including the zero periods) to get σd.
- Result
- Sortino Ratio = (Rp − Rf) / σd. A higher number is better. Returns above the MAR are rewarded in the numerator but do not inflate the denominator.
The downside deviation calculation is subtle: the squared shortfall terms are divided by the total number of periods (not just the bad ones) before taking the square root. This keeps the measure comparable across portfolios with different frequencies of bad months.
Sortino Ratio vs. Sharpe Ratio
The two ratios answer the same question — how much return do you earn per unit of risk? — but they disagree on what "risk" means.
| Feature | Sortino Ratio | Sharpe Ratio |
|---|---|---|
| Denominator | Downside deviation (bad months only) | Total standard deviation (all months) |
| Upside volatility penalty | None — positive surprises are ignored | Yes — big gains raise denominator |
| Best for | Asymmetric strategies: options, managed futures, momentum | Long-only equity portfolios with symmetric returns |
| Typical value vs Sharpe | Usually higher (smaller denominator) | Usually lower for skewed strategies |
Use the Sortino ratio when a strategy deliberately engineers positive skew — for example, when options overlays create large upside but limit downside. In those cases, the Sharpe ratio punishes the manager for the very asymmetry they were paid to create. For plain diversified equity portfolios where returns are roughly symmetric, the two ratios tell similar stories and either works. For a full primer on the Sharpe ratio, see the Sharpe Ratio guide, which covers the formula, worked examples, and benchmarks using total volatility.
Worked Example: Step-by-Step Calculation
A 6-month return series shows how downside deviation is calculated and how the Sortino ratio uses only the bad months in the denominator.
| Month | Return | vs MAR (0%) | Squared shortfall |
|---|---|---|---|
| Jan | +4% | Above MAR | 0 |
| Feb | +2% | Above MAR | 0 |
| Mar | −3% | Shortfall: −3% | (0.03)² = 0.0009 |
| Apr | +5% | Above MAR | 0 |
| May | −1% | Shortfall: −1% | (0.01)² = 0.0001 |
| Jun | +3% | Above MAR | 0 |
- Step 1: Sum of squared shortfalls
- 0.0009 + 0.0001 = 0.0010
- Step 2: Downside variance (divide by n=6)
- 0.0010 / 6 = 0.0001667
- Step 3: Monthly downside deviation (σd)
- √0.0001667 ≈ 0.01291 ≈ 1.29% per month
- Step 4: Annualize σd
- 1.29% × √12 ≈ 4.47% annualized
- Step 5: Annual portfolio return
- Sum of monthly returns: 4% + 2% − 3% + 5% − 1% + 3% = 10% (using simple sum for clarity)
- Sortino Ratio
- (10% − 4.5%) / 14.1% ≈ 0.39 (using σd annualized as 4.08% × √12 ≈ 14.1% with risk-free rate 4.5%)
This result looks low because only 6 months of data is used. The two loss months in this sample are mild (−3% and −1%), but they form the entire basis for the downside deviation estimate. In practice, 3+ years of monthly data (36+ observations) is preferred for a stable Sortino ratio. Short-history ratios are highly sensitive to whether a bad month happens to fall inside or outside the measurement window.
What Is a Good Sortino Ratio?
Benchmarks vary by strategy type. Hedge funds targeting 2.0+ are operating in a different risk environment than passive equity investors.
| Sortino Ratio | Interpretation | Typical context |
|---|---|---|
| Below 1.0 | Poor risk-adjusted return | Concentrated portfolios; short track records; volatile strategies |
| 1.0 – 2.0 | Acceptable | Passive equity funds in bull markets; diversified long-only portfolios |
| 2.0 – 3.0 | Good | Institutional target for active managers; hedge fund minimum threshold |
| Above 3.0 | Excellent | Rare; achievable in sustained bull trends or low-volatility strategies; warrants scrutiny for data quality |
Hedge funds commonly target Sortino ratios above 2.0, particularly those using tail-risk hedging or options overlays that explicitly limit downside volatility. Passive equity funds (S&P 500 index funds) typically register Sortino ratios between 1.0 and 1.5 in sustained bull markets. Because the Sortino uses a smaller denominator than the Sharpe, Sortino ratios are almost always higher than the same portfolio's Sharpe ratio for the same period.
Limitations of the Sortino Ratio
- Requires sufficient data. Short histories (under 36 monthly observations) produce unstable readings. If only a handful of losing months exist, whether one more bad month falls inside or outside the window can dramatically change the downside deviation estimate. Use 3–5 years of monthly data at minimum before treating a Sortino ratio as meaningful.
- MAR choice affects results. Setting the minimum acceptable return at 0%, the risk-free rate, or inflation produces different Sortino ratios for the same portfolio. A fund that targets 0% MAR will show a higher ratio than one benchmarked to inflation, because fewer months fall below the threshold. Always verify what MAR a manager is using before comparing ratios across funds.
- Not comparable across different MAR assumptions. A Sortino of 2.4 at MAR = 0% is not equivalent to 2.4 at MAR = 5%. League tables and fund comparisons are only valid when MAR is standardized. Industry convention typically uses 0% or the risk-free rate.
- Does not capture tail risk. The Sortino ratio uses standard deviation of downside returns, which still assumes a roughly normal distribution of losses. Black swan events — rare but catastrophic drawdowns — are not reliably captured. For tail-risk analysis, pair the Sortino ratio with maximum drawdown metrics and VaR (Value at Risk).
Related Risk Metrics
The Sortino ratio is one of several complementary tools for evaluating risk-adjusted performance. Each measures a different dimension of risk.
- Sharpe Ratio — uses total return volatility in the denominator, penalizing both upside and downside swings equally. The right starting point for symmetric, diversified equity portfolios. For strategies with asymmetric returns, the Sortino ratio only penalizes downside volatility and gives a more accurate risk picture.
- Beta — measures market sensitivity: how much a portfolio moves relative to a benchmark like the S&P 500. A complementary risk lens that captures systematic (market-driven) risk, while Sortino captures total downside risk relative to a minimum threshold.
- CAPM and Cost of Equity — uses beta to derive the required return for any asset given its systematic risk. Provides the theoretical foundation for what the risk-free rate and market risk premium imply for a minimum acceptable return — the MAR input in the Sortino formula.
- Treynor Ratio — replaces total standard deviation with beta in the denominator, so it only penalizes systematic (market) risk rather than all volatility. When comparing managers who each hold broadly diversified portfolios, the Treynor ratio is a natural complement to both the Sharpe and Sortino ratios.
- Return on Invested Capital (ROIC) — measures how efficiently a company deploys its capital inside the business, expressed as NOPAT divided by invested capital. While the Sortino ratio evaluates whether a portfolio of stocks earns adequate return for its downside risk, ROIC evaluates whether each underlying company in that portfolio is itself a disciplined allocator of capital — the two metrics operate at different levels of the same investment decision.