backtesting · 8 min read
Sharpe vs Sortino: Which Risk-Adjusted Metric Should You Trust?
Both measure return per unit of risk. But they define 'risk' differently — and that difference changes which strategies look good. Here's when each one lies to you.
By Quantinger Research
Return Alone Means Nothing
Two strategies both return 40% a year. The first does it with a smooth, steady climb. The second does it with stomach-churning swings — up 60%, down 40%, up 50% — that would have you checking your account at 3 a.m. and probably abandoning the strategy at the worst possible moment.
They have the same return. They are not the same strategy. The second is far riskier, and risk-adjusted metrics exist to make that difference visible in a single number.
The two most common are the Sharpe ratio and the Sortino ratio. They're closely related, they're both useful, and the difference between them tells you something important about how a strategy actually behaves.
The Sharpe Ratio: Return Per Unit of Total Volatility
The Sharpe ratio is the classic. It measures how much excess return a strategy generates per unit of total volatility:
Sharpe = (Strategy Return − Risk-Free Rate) / Standard Deviation of Returns
The numerator is your return above what you could earn risk-free (a treasury yield, say). The denominator is the standard deviation of your returns — a measure of how much your returns bounce around their average.
A higher Sharpe means more return for the volatility you endured. Rough industry benchmarks: below 1.0 is mediocre, above 1.0 is good, above 2.0 is very good, above 3.0 is excellent (and worth double-checking for overfitting).
The Sharpe ratio's strength is that it's universal and comparable. You can stack any two strategies side by side and the higher Sharpe generally gave you more return per unit of risk.
The Flaw Hiding in the Sharpe Ratio
Here's the problem. The Sharpe ratio uses total standard deviation — it treats all volatility as equally bad. Upside volatility and downside volatility count the same.
Think about what that means. A strategy that occasionally has explosive winning months gets penalized for that "volatility" exactly as if those were losing months. The Sharpe ratio punishes big gains.
But no trader actually minds upside volatility. Nobody loses sleep because their account jumped 30% in a month. The risk we actually care about is downside — the losses, the drawdowns, the months that hurt. By treating upside and downside identically, the Sharpe ratio can make a strategy with healthy upside surprises look worse than a flatline strategy that never gains much but also never spikes.
The Sortino Ratio: Only Downside Counts
The Sortino ratio fixes this by changing the denominator. Instead of total standard deviation, it uses downside deviation — the volatility of only the negative returns:
Sortino = (Strategy Return − Risk-Free Rate) / Downside Deviation
By measuring only the returns that fell below a target (usually zero or the risk-free rate), the Sortino ratio ignores upside swings entirely. It asks: "For the pain I actually experienced — the losses — how much return did I get?"
This is closer to how traders intuitively think about risk. A strategy with big upside and small, controlled downside will have a Sortino ratio noticeably higher than its Sharpe. That gap is informative: it tells you the strategy's volatility is mostly the good kind.
Reading the Gap Between Them
The relationship between a strategy's Sharpe and Sortino tells a story:
Sortino much higher than Sharpe. The strategy's volatility is mostly upside. Its big swings are gains, not losses. This is a desirable profile — the "risk" the Sharpe ratio penalized is actually the strategy occasionally winning big. Many trend-following strategies look like this: lots of small losses and occasional huge wins, which inflates total volatility (hurting Sharpe) but is mostly upside (helping Sortino).
Sharpe and Sortino similar. The strategy's gains and losses are roughly symmetric in their volatility. Neither tail dominates.
Sortino barely higher than Sharpe, both low. The strategy has meaningful downside volatility — real losing periods that drag both metrics down. This is the profile to be cautious about.
Which Should You Actually Use?
Neither alone. Use both, and understand what each is telling you.
Use the Sharpe ratio for broad, standardized comparison — it's the industry default, so it lets you benchmark against funds, other strategies, and published numbers. Just remember it penalizes upside.
Use the Sortino ratio to understand the character of a strategy's risk — specifically, whether its volatility is the kind that hurts (downside) or the kind that helps (upside). For strategies with asymmetric return profiles, especially trend-following, the Sortino is the fairer judge.
A strategy that looks mediocre by Sharpe but strong by Sortino isn't being dishonest — it's a strategy whose "risk" is mostly its tendency to win big occasionally. That's exactly the kind of strategy the Sharpe ratio is worst at evaluating.
The Metrics Neither One Captures
Both ratios share blind spots worth naming:
They don't show drawdown. A strategy can have a great Sharpe and still have suffered a single 40% drawdown that would have caused most traders to quit. Standard deviation and max drawdown are different things. Always look at max drawdown alongside these ratios — pair them with the Calmar ratio (return divided by max drawdown) for the full picture.
They assume normal-ish distributions. Both ratios are most meaningful when returns are roughly normally distributed. Strategies with rare, extreme outcomes (selling options, for instance) can show beautiful Sharpe ratios right up until the catastrophic loss that the metric never saw coming. A high Sharpe on a strategy with hidden tail risk is a trap.
They're period-dependent. A strategy's Sharpe over a calm year and a volatile year can differ wildly. Always note the period and prefer ratios computed over multiple market regimes.
The Bottom Line
The Sharpe ratio measures return per unit of total volatility — comprehensive but unfair to strategies with healthy upside. The Sortino ratio measures return per unit of downside volatility — closer to the risk traders actually care about. The gap between them reveals whether a strategy's swings are mostly gains or mostly losses.
Use both. Pair them with max drawdown. And never trust a single risk-adjusted number without knowing the shape of the distribution behind it — because the most dangerous strategies are the ones that look smooth right up until they don't.
See both ratios on every backtest: Quantinger reports Sharpe, Sortino, and max drawdown together so you see the full risk picture, not one flattering number.