#1 RealTest Backtests
Supercharge Your Trading Now
Reduce drawdown, build diversification, or speed up your development time.
The Sharpe ratio divides excess return by total volatility, so it penalises a big up month exactly as hard as a big down month. The Sortino ratio divides by downside deviation only. Because most systematic strategies are asymmetric, Sortino usually reads higher and is the closer match to what traders actually fear, which is drawdown, not upside.
When evaluating trading strategies, risk-adjusted returns matter more than raw returns. Two metrics dominate this analysis: the Sharpe ratio and the Sortino ratio. While both measure return per unit of risk, they differ fundamentally in how they define 'risk' and that difference can completely change how you evaluate your trading edge.
This guide breaks down both metrics, explains when each is most useful, and shows you which one better captures the reality of systematic trading performance.
The Sharpe ratio, developed by Nobel laureate William Sharpe in 1966, measures excess return per unit of total volatility. The formula is:
Sharpe Ratio = (Portfolio Return - Risk-Free Rate) / Standard Deviation of Returns
The Sharpe ratio treats all volatility equally, whether returns swing up or down. A strategy that delivers +5% one month and +3% the next is penalized the same as one that delivers +5% then -3%, even though most traders only care about downside moves.
The Sharpe ratio works well for strategies with symmetric return distributions, but it has a critical flaw: it penalizes upside volatility just as much as downside risk.
The Sortino ratio, introduced by Frank Sortino in the 1980s, refines the Sharpe ratio by focusing exclusively on downside deviation. The formula is:
Sortino Ratio = (Portfolio Return - Target Return) / Downside Deviation
Instead of using total standard deviation, the Sortino ratio only measures volatility of negative returns below a target threshold (often 0% or the risk-free rate). This makes it a more realistic measure for traders who don't mind upside volatility but want to minimize drawdowns.
The Sortino ratio is particularly valuable for mean reversion strategies and other approaches with asymmetric return profiles.
| Feature | Sharpe Ratio | Sortino Ratio |
|---|---|---|
| Risk Measure | Total volatility (standard deviation) | Downside deviation only |
| Upside Volatility | Penalized | Ignored |
| Best For | Symmetric return distributions | Asymmetric returns, skewed strategies |
| Trader Preference | Academic standard | Practical trader focus |
| Calculation Complexity | Simple | Slightly more complex |
Most systematic traders care about one thing: avoiding large drawdowns. The Sortino ratio aligns with this goal by ignoring upside volatility and focusing exclusively on downside risk.
Consider two hypothetical strategies over 12 months:
Strategy A (Momentum):
Strategy B (Mean Reversion):
By Sharpe ratio, Strategy B looks slightly better (0.33 vs 0.31). But the Sortino ratio reveals Strategy A has superior downside protection (0.63 vs 0.40). It experiences large upside moves that inflate its standard deviation but doesn't suffer proportionally large drawdowns.
For traders focused on capital preservation, Strategy A is the better choice, but you'd only know this by examining the Sortino ratio.
Both ratios are sensitive to the measurement period. A strategy might show a Sharpe of 2.0 over 3 years but only 0.8 during the most recent 12 months. Always examine multiple timeframes and consider regime changes.
Leverage amplifies both returns and risk proportionally, leaving the Sharpe and Sortino ratios unchanged. A 2x leveraged strategy with a Sharpe of 1.5 will still show 1.5, even though absolute risk has doubled. Always evaluate leverage separately.
A Sharpe ratio of 1.2 in equities isn't directly comparable to 1.2 in crypto or forex due to different volatility regimes and correlation structures. Context matters.
Neither ratio fully captures tail risk. A strategy can have a strong Sortino ratio but still experience a catastrophic 40% drawdown. Always pair these metrics with maximum drawdown analysis.
Most backtesting platforms calculate Sharpe and Sortino ratios automatically, but understanding the math helps you interpret results correctly.
If you're backtesting strategies in RealTest, both metrics are included in the standard performance report. Our Complete RealTest Strategy Bundle includes 11 fully-backtested strategies with detailed Sharpe and Sortino analysis.
While Sharpe and Sortino ratios are essential, they shouldn't be your only evaluation criteria. Consider these complementary metrics:
For a deeper dive into performance metrics, check out Jim Simons' 47 trading wisdoms, which emphasize the importance of robust statistical analysis.
Here's a step-by-step framework for using these ratios in your strategy development:
Generate a complete equity curve with at least 3-5 years of data (more for lower-frequency strategies). Ensure your backtest accounts for realistic costs, slippage, and position sizing.
Don't rely on just one metric. Calculate Sharpe, Sortino, and maximum drawdown as your core trio.
Plot a histogram of your returns. If it's roughly symmetric (bell curve), Sharpe is reliable. If it's skewed with a long right tail (many small losses, few large wins), Sortino is more informative.
How does your strategy compare to buy-and-hold SPY? To other systematic strategies in the same asset class? Context is critical.
Recalculate ratios for different market regimes: bull markets, bear markets, high volatility periods. A strategy with a 2.0 Sortino in bull markets but 0.3 in bear markets needs refinement.
Our Low Drawdown Nasdaq Mean Reversion Strategy demonstrates why Sortino matters:
The higher Sortino ratio reveals that most volatility comes from upside moves, while drawdowns remain controlled. This is exactly what mean reversion traders want to see.
Momentum strategies often show different patterns. A typical cross-sectional momentum strategy might show:
Again, the Sortino ratio is higher, indicating that large winning streaks create upside volatility that the Sharpe ratio penalizes unfairly.
For most systematic traders, the Sortino ratio is the superior metric because it aligns with how traders actually think about risk. You don't lose sleep over a +8% month followed by a +2% month. You lose sleep over drawdowns.
However, the best approach is to use both ratios together:
Always combine these ratios with maximum drawdown, profit factor, and win rate analysis for a complete picture.
If you're serious about systematic trading, you need proper backtesting tools that calculate these metrics automatically:
Our Beginner to Pro RealTest Bundle includes three strategies with complete performance breakdowns, showing you exactly how to interpret Sharpe and Sortino ratios in real backtests.
For those building strategies from scratch, explore our 11 best sources for algorithmic trading ideas to find strategies worth testing.
A strategy returning 30% annually with a maximum drawdown of 45% is far less attractive than one returning 18% with a 12% drawdown. The Sharpe and Sortino ratios quantify this intuition.
The Sortino ratio's focus on downside deviation makes it particularly valuable for traders who prioritize capital preservation and smooth equity curves. The Sharpe ratio remains the academic standard and is useful for symmetric strategies.
Use both metrics, understand their limitations, and always combine them with drawdown analysis and other performance measures. The goal isn't to maximize a single number. It's to build robust, tradeable strategies that survive real market conditions.
Ready to see these metrics in action? Explore our Complete RealTest Strategy Bundle with 11 professionally backtested strategies, each with detailed Sharpe and Sortino analysis, maximum drawdown data, and full transparency into what works in systematic trading.
The Sharpe ratio, developed by William Sharpe in 1966, measures excess return per unit of total volatility. It is calculated as portfolio return minus the risk-free rate, divided by the standard deviation of returns. Because it uses total standard deviation, it treats upside and downside movement as equally undesirable.
The Sortino ratio, introduced by Frank Sortino in the 1980s, measures excess return per unit of downside deviation. It is calculated as portfolio return minus a target return, divided by the deviation of returns that fall below that target. Returns above the target do not count as risk.
Both divide excess return by a measure of risk, but they define risk differently. Sharpe uses total volatility, so a strong up month raises measured risk. Sortino uses downside deviation only, so upside volatility is ignored. Sharpe suits symmetric return distributions, Sortino suits skewed ones.
For most systematic traders the Sortino ratio is the closer match, because traders lose sleep over drawdowns rather than over large winning months. The practical approach is to read both. Similar values mean symmetric returns, a much higher Sortino means useful upside volatility.
Below 1.0 is poor, meaning the strategy barely compensates for its volatility. Between 1.0 and 2.0 is good, 2.0 to 3.0 is very good, and above 3.0 is exceptional and rare in retail trading. The same bands apply when reading a Sortino ratio.
Yes. Because it penalises upside volatility, a strategy with occasional large wins can show a weak Sharpe while having well controlled drawdowns. In the article's example, the strategy with the better Sharpe had the worse downside profile, and only the Sortino ratio revealed it.
No. Leverage scales returns and risk in proportion, so both ratios stay the same. A 2x leveraged strategy with a Sharpe of 1.5 still reports 1.5 even though the absolute risk has doubled. Leverage has to be evaluated separately, alongside maximum drawdown.
Calculate the average periodic return, subtract your target return, then calculate the downside deviation using only the returns below that target. Divide the excess return by the downside deviation. Annualise if needed, multiplying by the square root of 252 for daily data or 12 for monthly.
Yes, and this is the main reason to read both metrics. Two strategies can share a Sharpe ratio while one gets its volatility from large gains and the other from large losses. The Sortino ratio separates them, because only the second one is penalised.
Maximum drawdown and drawdown duration first, because neither ratio captures tail risk. Then the Calmar ratio (return over maximum drawdown), the Omega ratio, profit factor, win rate and average win versus average loss. No single number describes a strategy.
Most do. RealTest reports both Sharpe and Sortino in its standard performance report, and Python with pandas or numpy gives full control over custom variants. The value of knowing the maths is in interpreting the output rather than producing it.