Sortino Ratio
Also written Sortino measure · Downside risk-adjusted return
Excess return over the risk-free rate divided by semi-standard deviation — a Sharpe-style ratio that counts only downside volatility as risk.
In plain language
The Sharpe ratio punishes a portfolio for any volatility, including months when returns jumped well above average. Most investors do not lose sleep over those months.
The Sortino ratio keeps Sharpe's numerator — return in excess of the risk-free rate — but replaces the denominator with a measure of downside volatility only: the semi-standard deviation, built from returns that fall below the mean (or below a chosen target).
The workbook's framing: it appeals to investors who see risk as the chance of losing money, rather than as uncertainty around the expected return. As with every risk-adjusted ratio, higher is better.
How it works
The workbook separates two families of risk measure:
- Total risk — variance and standard deviation, which count deviations on both sides of the mean.
- Downside risk — semi-variance (dispersion of returns below the mean), target semi-variance (dispersion below a target return), and their square roots.
Sortino divides excess return by the downside version.
The workbook adds a caution about semi-variance itself: if returns are symmetrically distributed, semi-variance is simply proportional to variance and adds no insight; and asymmetric distributions are very difficult to forecast. For those reasons it says semi-variance is not so popularly used in practice.
The formula
Sortino ratio = (Rp − Rf) ÷ Semi-standard deviation of the portfolio
Sharpe ratio = (Rp − Rf) ÷ Standard deviation of the portfolio
A worked example
Illustrative figures. Two multi-cap PMS strategies over the same three years. Risk-free rate 6%.
| Strategy A | Strategy B | |
|---|---|---|
| Annualised return | 14% | 14% |
| Standard deviation | 16% | 16% |
| Semi-standard deviation | 9% | 13% |
| Sharpe = (14 − 6) ÷ 16 | 0.50 | 0.50 |
| Sortino = (14 − 6) ÷ semi-SD | 8 ÷ 9 = 0.89 | 8 ÷ 13 = 0.62 |
Sharpe calls them identical. Sortino does not. Strategy A's volatility comes mainly from sharp up months; Strategy B's comes mainly from sharp down months.
For a retired client drawing ₹60,000 a month from a ₹1 crore portfolio, that distinction is not academic. A bad year in B forces sales at depressed prices to fund withdrawals. The Sortino ratio is the measure that notices.
Why NISM asks about it
Chapter 20 (Performance Measurement and Evaluation of Portfolio Managers) introduces downside risk in section 20.3.1 and the Sortino ratio in section 20.4.4, alongside Sharpe, Treynor, the Information ratio and M². Questions typically ask which ratio uses which risk measure, or which ratio suits an investor who defines risk as the possibility of loss.
Common exam traps
- Sortino = semi-standard deviation; Sharpe = standard deviation; Treynor = beta; Information ratio = tracking error. Matching questions turn on exactly this.
- The numerator is the same as Sharpe's — excess over the risk-free rate, not over a benchmark.
- Semi-variance vs target semi-variance: below the mean vs below a chosen target. The workbook's footnote says the mean can be replaced with any threshold the investor regards as the downside.
- The workbook's footnote describes semi-standard deviation loosely, as "the average of all deviations of actual returns below the mean". Treat it as the downside counterpart of standard deviation; the exam asks what it captures, not how to compute it.
- For symmetric returns, Sortino adds little — semi-variance is then proportional to variance.
Where this is taught
Free preparation for NISM Series XXI-BRelated terms
- Sharpe ratioReturn earned above the risk-free rate divided by standard deviation — how much reward an investment produced for each unit of total risk its holder had to live with.
- Standard deviationA measure of how far returns typically stray from their own average — the standard statistic for total risk, counting company-specific and market-wide causes alike.
- Treynor ratioRisk premium per unit of market risk — the return a scheme earned above the risk-free rate, divided by its beta rather than by its standard deviation.
- Information RatioActive return over the benchmark divided by tracking error — how much outperformance a manager delivers for each unit of risk taken by deviating from the index.
- M² measureA portfolio's return re-scaled to the market's level of risk, so it can be compared with the market return directly, in percentage points.
- Semi varianceA downside-risk measure that looks only at the dispersion of returns below the mean (or a target return), ignoring the upside swings that standard deviation counts equally alongside the downside ones.