Sortino Ratio - Returns Adjusted Only for Downside

By EC Assets Research Team, Performance Analytics · Published 28 May 2026

Sortino Ratio — The Sortino ratio is a risk-adjusted return measure that penalises only downside volatility, treating upside variation as desirable rather than risky. It modifies the Sharpe ratio by replacing total standard deviation with downside deviation below a target return.

Definition

The Sortino ratio takes the same numerator as the Sharpe ratio - excess return over a benchmark or minimum acceptable return - but changes the denominator. Where Sharpe divides by total standard deviation (treating positive and negative deviations as equally undesirable), Sortino divides by downside deviation (counting only returns below the target).

Formally, given a return series and a target return T (often zero, or the risk-free rate, or a hurdle rate):

The change matters when return distributions are skewed. A strategy with infrequent large upside spikes and modest downside is penalised by Sharpe for being too volatile, even though all the volatility is in the direction investors want. Sortino corrects that asymmetry.

The ratio was introduced by Frank Sortino in the early 1980s as a refinement of post-modern portfolio theory, which argues that variance is the wrong risk measure when investors care only about losses.

How Sortino Differs From Sharpe

Two distinct philosophies are at play.

For symmetric return distributions (close to normal), Sortino and Sharpe rank strategies similarly. For asymmetric distributions - option-selling strategies, trend-following CTAs, distressed credit - they can disagree sharply, sometimes producing the opposite ranking.

The choice of target is the second philosophical input. Common conventions are T = 0 (penalise any negative return), T = risk-free rate (penalise underperformance versus cash), or T = hurdle (penalise underperformance versus a contractual benchmark). The ratio is not interpretable in isolation - the target must be stated explicitly.

Worked Example

Consider two strategies with the following monthly statistics over 24 months. The risk-free rate is 0.3% per month and the target T is set to zero.

Strategy A wins on Sharpe; Strategy B wins on Sortino. The reason: B's total volatility was driven by upside spikes (which Sharpe penalises but Sortino does not), while its downside deviation was actually lower than A's. Whether B is genuinely the better strategy depends entirely on whether the upside spikes are repeatable - a question Sortino does not answer.

When Sortino Applies - and Where It Misleads

Useful for:

• Strategies with intentionally asymmetric payoffs (long options, trend-following, distressed)
• Investors whose mandate forbids losses below a threshold but is indifferent to upside magnitude
• Goal-based wealth management, where the goal is to clear a hurdle (retirement income, pension liability) rather than minimise variance

Misleading when:

• The downside sample is small. With monthly data over five years, only 60 observations exist; if only ten are negative, the downside deviation estimate is statistically unreliable.
• The target T is unspecified or inconsistently chosen across managers.
• The strategy's upside is not repeatable. A backtest with one lucky 30% month produces a flattering Sortino that will not persist.
• Tail events are present. A strategy with capped upside and unbounded downside (selling far-out-of-the-money puts) shows attractive Sortino in calm periods and is misleading until a crash arrives.

Why Sortino Matters

Three institutional contexts demand the Sortino frame.

First, liability-driven investing. A pension fund whose liabilities discount at a hurdle rate cares about returns falling below that hurdle, not about returns that exceed it. Sharpe treats outperformance as risk; Sortino treats it as performance. For LDI mandates, Sortino is conceptually correct.

Second, option-based strategies. A long-volatility tail-hedge fund has high total volatility but most of it occurs during crisis spikes when it is delivering the strategy's entire raison d'être. Reporting it on Sharpe will systematically understate its quality. Allocators evaluating such mandates routinely run Sortino in parallel.

Third, manager communication. When a manager produces a Sharpe of 0.8 and a Sortino of 1.8, the gap is itself diagnostic. It signals that the manager's volatility is predominantly upside - useful information for the allocator, particularly if the divergence is consistent across multiple periods.

Sortino does not replace Sharpe. The right way to use it is alongside Sharpe, alongside maximum drawdown, alongside the actual return distribution, with the target T explicitly stated. Used that way, it sharpens the picture rather than re-arranging it.

Sortino vs Sharpe in Practice

Consider three managers, all with the same Sharpe ratio of 0.8:

All three look identical on Sharpe because the symmetric standard deviation is the same. But Sortino reveals stark differences: Manager B has positive convexity (volatility mostly from upside surprises), while Manager C has negative convexity (volatility from rare large losses).

The Sortino metric is more revealing for strategies with asymmetric return distributions: option-selling strategies, distressed credit, merger arbitrage, certain insurance-like strategies. Sharpe treats their upside vol identically to downside vol, missing the asymmetry that defines their actual risk.

When NOT to Use Sortino

Sortino is not universally better than Sharpe. Three caveats:

• Sortino requires more data than Sharpe (need enough downside observations to estimate downside deviation reliably). For short track records (3-5 years), Sharpe is more robust.
• Sortino can be gamed by managers using strategies with hidden tail risk (long calm periods, then catastrophic loss). The downside deviation looks low until the tail event arrives.
• The "minimum acceptable return" threshold in Sortino is somewhat arbitrary. Different choices (0%, risk-free, benchmark return) produce different Sortino values.

References

1. Bodie, Z., Kane, A., & Marcus, A. J. (2021). Investments (12th ed.). McGraw-Hill.
2. Sharpe, W. F. (1994). The Sharpe Ratio. Journal of Portfolio Management, 21(1).
3. CFA Institute. Portfolio Management: Performance Evaluation. CFA Program Curriculum.