How the Sortino ratio works
By Jude Wallis
The Sortino ratio measures excess return per unit of downside risk. Subtract the target from the return, then divide by the deviation of below-target returns only. A 10 percent return against a 3 percent target with 8 percent downside deviation gives 0.875.
Sortino ratio
0.875
Excess return over the target, divided by downside deviation only.
- Sortino
- 0.8750
Put this on a class page: one iframe, free, for Google Sites, Canvas, WordPress or Notion.
On this page
In short
- Excess return sits on top: 10 percent minus a 3 percent target is 7 points.
- Downside deviation sits underneath, computed from below-target returns only, so upside swings are ignored.
- 7 points over 8 points of downside deviation is a Sortino ratio of 0.875.
- Halving the downside deviation to 4 points doubles the ratio to 1.75 on an unchanged return.
- The ratio only compares against another Sortino built on the same target.
It only counts the risk you actually mind
Standard deviation treats a return far above average as identical in risk to one far below it. No investor experiences those the same way, and for strategies that are deliberately lopsided the symmetric measure gives a misleading answer.
Sortino fixes that by building the denominator from below-target returns only. Everything above the target contributes nothing to the risk figure, so a strategy with occasional very good months is not punished for having them. On this portfolio, 7 points of excess over 8 points of downside deviation gives 0.875.
The target is an input, and it matters
The number on top is return minus a target you choose. Use zero and you are measuring excess over breaking even. Use the risk-free rate and you are close to a Sharpe numerator. Use a required return and you are measuring against the hurdle you actually care about.
All three are legitimate and they produce different ratios from identical data, so a Sortino quoted without its target is not comparable to anything. State it, and use the same one on both sides of any comparison.
Small changes in the denominator move it a long way
Because the ratio divides by downside deviation, the denominator carries most of the variation between funds. A portfolio with the same 7 points of excess but only 4 points of downside deviation scores 1.75, twice as high, with no improvement in return at all.
That sensitivity is why the inputs should travel with the answer. Two funds quoting 0.875 and 1.75 might differ entirely in how consistently they lose, which is exactly what the measure is designed to surface.
Reading it next to Sharpe
Sortino is usually the higher of the two on the same data, because downside deviation is smaller than total volatility. That does not make it a better score, only a different question. Sortino against Sharpe sets the two denominators side by side, how the Sharpe ratio works covers the older measure, and standard deviation of returns covers the statistic underneath both. The Sortino ratio calculator shows the excess and the denominator rather than the ratio alone. This is educational material, not financial advice.
Worked examples
10 percent against a 3 percent target
A portfolio returns 10 percent, the target is 3 percent, and downside deviation is 8 percent. What is the Sortino ratio?
- Excess return is 10 minus 3, which is 7 points.
- Divide by 8 points of downside deviation: 0.875.
The Sortino ratio is 0.875, so each point of downside risk earned a little under a point of excess return.
The same return, half the downside
A second portfolio also returns 10 percent against a 3 percent target, but its downside deviation is 4 percent.
- The excess is unchanged at 7 points.
- Divide by 4 instead of 8: the ratio is 1.75.
1.75, exactly double, from consistency rather than from return. Nothing on the top of the fraction moved.
Common questions
How is downside deviation calculated?
From the returns that fell below the target, squared and averaged over the full period, then square rooted.
Can Sortino be negative?
Yes, when the return is below the target. Negative risk-adjusted ratios rank poorly performing funds in a confusing order.
Is a higher Sortino always better?
Only against the same target and over the same period. Outside those conditions the two numbers are not comparable.
Is this financial advice?
No. It is educational material about a risk-adjusted return measure.
Keep reading
This page is educational material, not financial advice. The figures come from the formula shown and assume the inputs you enter hold for the whole term. Your own rate, fees, taxes and timing will differ, so treat the output as arithmetic to check a decision against, not as a recommendation.