Sortino Ratio: Downside Risk, Minimum Acceptable Return, and Limits

What is Sortino Ratio?

The Sortino ratio measures return above a selected target relative to downside deviation.

A common form is:

1Sortino Ratio
2=
3Average Return - Minimum Acceptable Return
4------------------------------------------
5             Downside Deviation

The target is often called the minimum acceptable return, or MAR.

Unlike the Sharpe Ratio, which uses total Volatility, the Sortino ratio focuses on returns that fall below the chosen target.

That makes it especially useful when the analytical question is:

1How much return was earned relative to harmful shortfall variation?

The result depends critically on how the target and downside deviation are defined.

A simple Sortino example

Suppose a portfolio has:

1Average annualized return: 10%
2Minimum acceptable return:  4%
3Annualized downside deviation: 8%

Then:

1Sortino Ratio
2= (10% - 4%) / 8%
3= 0.75

The number is unitless.

A higher positive Sortino ratio means more return above the selected target per unit of measured downside deviation, all else equal.

But the ratio is not meaningful without knowing the target and calculation convention.

The minimum acceptable return changes the metric

The MAR is not a cosmetic input.

It defines what counts as a shortfall.

One analyst might use:

1MAR = 0%

Another might use the risk-free rate.

A pension plan could use an actuarial hurdle.

An investor could use a required return tied to spending needs.

The same return series can therefore produce different Sortino ratios under different MAR choices.

That means a statement such as:

1"The Sortino ratio is 1.2"

is incomplete unless the target and methodology are understood.

How downside deviation works

A common downside-deviation calculation focuses on shortfalls below the target:

1Downside Deviation
2=
3sqrt[ Σ min(0, r_t - MAR)^2 / n ]

where:

• r_t is the periodic return;
• MAR is the periodic minimum acceptable return; and
• n is the number of observations under the chosen convention.

Returns above the target contribute zero shortfall in this formulation.

That is the key difference from standard deviation, where both positive and negative deviations around the average increase measured volatility.

Implementations can vary, including denominator and annualization choices, so provider methodology matters.

Sortino ratio does not simply mean "negative volatility only"

A common shorthand says the Sortino ratio penalizes only negative returns.

That can be misleading.

The target is the chosen MAR, not necessarily zero.

If the target return is 5% annually, a positive return below that target can still count as a shortfall under a properly aligned periodic calculation.

The more precise statement is:

1Sortino penalizes returns below the selected target.

That target can be zero, the risk-free rate, or another required return.

Sortino versus Sharpe ratio

The Sharpe Ratio uses total standard deviation.

The Sortino ratio uses downside deviation relative to a target.

If an investment has large upside jumps but relatively controlled downside shortfalls, the Sharpe ratio can penalize those upside moves while the Sortino ratio does not.

That can produce a much higher Sortino ratio than Sharpe ratio.

The difference can be informative because it signals asymmetry in the return distribution.

But it does not automatically prove that the investment is superior.

A strategy can have favorable downside deviation and still carry rare crash risk that does not appear often enough in the sample.

Sortino ratio is not maximum drawdown

Maximum Drawdown measures the largest observed peak-to-trough decline over a time window.

Sortino ratio measures return relative to the distribution of below-target periodic returns.

Those are different dimensions of risk.

For example, a strategy can experience one severe crash and many otherwise strong periods.

Depending on the sample length and return frequency, that isolated event may not dominate downside deviation as much as the investor's lived drawdown experience suggests.

A useful review therefore considers both:

1Sortino Ratio   -> downside-adjusted return efficiency
2Maximum Drawdown -> worst observed peak-to-trough path loss

Neither replaces the other.

Return frequency affects the calculation

Daily, weekly, and monthly return series can produce different downside-deviation estimates.

A sharp intramonth loss that recovers before month-end may disappear from a monthly series.

Likewise, a monthly series has fewer observations, which can make estimates less stable.

The target return also needs to be converted to the same periodic frequency before shortfalls are measured.

Mixing an annual MAR directly with monthly returns creates a mismatched calculation.

Frequency consistency matters just as much for Sortino as it does for Sharpe.

Annualization requires a convention

A periodic Sortino ratio is often annualized, commonly using square-root-of-time scaling for downside deviation under simplifying assumptions.

But return processes can be serially correlated, volatility can change over time, and downside events can cluster.

Therefore, annualized Sortino should be understood as a methodological convention, not an exact transformation that is universally valid.

When comparing products, confirm that they use compatible return frequencies, targets, and annualization methods.

Few downside observations can make the ratio unstable

Imagine a strategy with three years of monthly data and only two months below the selected target.

Its estimated downside deviation may look very low.

That can produce an unusually high Sortino ratio.

But the estimate may be fragile because very few adverse observations are carrying most of the denominator.

A longer or less favorable sample can change the result dramatically.

This issue is especially important for strategies with infrequent losses, smoothed marks, or nonlinear payoff structures.

Option-like strategies can fool simple summaries

A strategy that repeatedly earns small gains while occasionally suffering a very large loss can appear attractive for long stretches.

If the sample contains few crash events, both volatility- and downside-based ratios can understate the economic significance of the tail exposure.

Examples can include some short-volatility, leveraged carry, and credit strategies.

The lesson is not that Sortino is useless.

It is that no one ratio captures every shape of downside risk.

Investors should inspect the return distribution, drawdowns, leverage, liquidity, and scenario behavior.

Sortino ratio does not establish alpha

A high Sortino ratio does not prove manager skill or Alpha.

A portfolio may earn attractive downside-adjusted returns because it is exposed to compensated systematic risks.

Alpha asks whether returns remain after accounting for a selected benchmark or model.

Sortino asks how return above a selected target compares with downside deviation.

Those are different questions.

Fees and financing still matter

Gross backtest returns can generate an attractive Sortino ratio while net investor returns do not.

Management fees, transaction costs, spreads, financing expenses, taxes, and market impact can reduce the numerator.

If a strategy relies on leverage, stressed financing costs can also arrive at the same time downside risk is rising.

For meaningful comparisons, use returns measured on a consistent net or gross basis and make the convention explicit.

Negative Sortino ratios need context

When average return falls below the MAR, the numerator becomes negative.

As with negative Sharpe ratios, mechanical ranking can become unintuitive because a larger denominator can make a negative ratio look numerically closer to zero.

A negative Sortino ratio should therefore be interpreted alongside the actual average return, target return, downside deviation, and drawdown record.

Do not reduce the analysis to "less negative is always better."

How investors should use the Sortino ratio

A disciplined Sortino review asks:

1. What MAR or target return was used?
2. Was the target converted to the same frequency as returns?
3. How exactly was downside deviation calculated?
4. How many below-target observations exist?
5. How was the ratio annualized?
6. Are returns liquid and frequently marked?
7. Are fees, trading costs, and financing included?
8. What does maximum drawdown show?
9. How different is the Sharpe ratio, and what explains the gap?
10. Does the strategy have skewed or option-like tail risk?

Grizzly Bulls' Models research can be evaluated with these distinctions without turning this encyclopedia page into a live Sortino calculator. The Macroeconomic Conditions Index can add separate regime context, but it does not define the denominator or MAR used here.

Sources and further reading

• CFA Institute, 2026, Portfolio Performance Evaluation: https://www.cfainstitute.org/insights/professional-learning/refresher-readings/2026/portfolio-performance-evaluation
• CFA Institute Research and Policy Center, Risk-Adjusted Performance Measures: https://rpc.cfainstitute.org/topics/performance-attribution/risk-adjusted-performance
• CFA Institute, 2026, Analysis of Active Portfolio Management: https://www.cfainstitute.org/insights/professional-learning/refresher-readings/2026/analysis-active-portfolio-management
• CFA Institute, Portfolio Development and Construction Practical Skills Module: https://www.cfainstitute.org/programs/cfa-program/candidate-resources/practical-skills-modules/portfolio-development-and-construction

Explore more topics in the Financial Research Encyclopedia.