What is Sortino Ratio?
Reviewed by Karthikeyan Anbarasan (Founder, TradeLyser) · Updated 19 Oct 2026 · Editorial policy
How downside deviation is worked out
- Choose a target return per period. Here it is 0.5% a month (6% a year); zero is also common.
- For each month, take the shortfall below the target, or zero if the month beat it. Strategy A falls short in five months, by 1.5, 0.5, 2.5, 1.5 and 0.5 points.
- Square the shortfalls, add them and divide by all 12 months: 11.25 ÷ 12 = 0.94.
- Take the square root: downside deviation is 0.97%.
- Divide the excess return, 1.5% − 0.5% = 1.0%, by it: 1.03 a month, or about 3.6 annualised with √12.
Conventions differ. Some calculations divide by the number of months below target instead of all months, or use zero as the target, and the result changes. Use one method for every strategy you compare.
Worked example: an option-selling book
Illustrative. An option-selling strategy on ₹10,00,000 earns 2% (₹20,000) in each of 11 months and loses 12% (−₹1,20,000) in one month. The target is zero. Over 12 months the average is (22 − 12) ÷ 12 = 0.83% a month, ₹1,00,000 in total on a simple basis.
| Measured over | Average monthly return | Downside deviation | Annualised Sortino |
|---|---|---|---|
| The first 11 months only | 2.0% | 0% | Not defined: no month below target |
| All 12 months | 0.83% | 3.46% | 0.83 |
| 12 months, losing month capped at −6% and other months at 1.6% | 0.97% | 1.73% | 1.9 |
Before the bad month the ratio cannot be computed, because there is no downside to divide by; a tool that shows a very large figure there is showing the absence of a test, not safety. One month at −12% puts downside deviation at √(144 ÷ 12) = 3.46%. In the third row a tighter exit rule gives up some monthly income but halves the worst month, and Sortino more than doubles.
Sortino vs Sharpe
Sharpe
Sortino
Downside deviation ignores moves above the target, so it is usually smaller than the standard deviation, and Sortino usually comes out higher than Sharpe on the same returns. Compare Sortino figures with each other, not with a Sharpe ratio.
Reading it with care
Tracking Sortino in your journal
- Record month-end equity for each strategy, after charges, and convert it to a monthly return on the capital set aside for it.
- Fix one target, zero or a risk-free rate, and one convention for the divisor, and write both in your trading plan.
- Recompute Sortino on a rolling 12- or 24-month window each month, alongside max drawdown and the number of months below target.
- Tag the months that fell below target with what caused them: a single event, a run of small losses or a broken rule.
If Sortino is high but only one or two months sit below target, treat the figure as provisional until more losing months have been seen. A rising Sortino alongside a deepening max drawdown is a sign to look at the losing months themselves rather than the ratio.
Reference guide
| Context | Value | Reading |
|---|---|---|
| Sortino far above Sharpe | Volatility mostly on the upside | Losing periods are small or rare next to the good ones; check max drawdown too |
| Sortino only a little above Sharpe | Losses as large as gains | Downside swings make up much of the volatility |
| Only a few periods below target | Unstable figure | Two or three bad months set the whole denominator, and one more can move it a lot |
Questions
What is a good Sortino ratio?
There is no universal cutoff, and values depend on the target and the period used. Compare strategies with the same target, the same return frequency and the same date range.
What is the difference between the Sortino and Sharpe ratios?
Sharpe divides excess return by the volatility of all returns. Sortino divides it by downside deviation, which counts only returns below a target, so large up months do not count against it.
What target return should I use for Sortino?
Zero or a risk-free rate are both common. Use the same target for every strategy you compare, and state it next to the number.
Why is my Sortino ratio so much higher than my Sharpe ratio?
Downside deviation leaves out every period above the target, so it is smaller than the standard deviation. The more of your volatility comes from good months, the wider the gap.
How is the Sortino ratio calculated?
Subtract the target from the average return, then divide by downside deviation: the square root of the average squared shortfall below the target, with periods above it counted as zero. A monthly figure is commonly annualised by multiplying by √12.
Can the Sortino ratio be infinite?
If no period falls below the target, downside deviation is zero and the ratio is not defined. Some tools show a very large number instead. Either way it means the record has not yet been tested by a losing period.