Blogs
Sharpe vs Sortino vs Calmar: How to Measure Risk-Adjusted Returns
Aman Anand
Sharpe vs Sortino vs Calmar: How to Measure Risk-Adjusted Returns
Two trading strategies both return 20% a year. One did it on a smooth, steady equity curve; the other lurched through gut-wrenching 40% drawdowns to get there. On a raw-return basis they look identical — yet any experienced trader would take the first in a heartbeat. The difference is risk-adjusted return: how much reward a strategy earns for every unit of risk it takes.
Raw returns are the most quoted and least useful number in trading. They tell you what happened, not how much danger you were exposed to along the way, and they reward reckless leverage exactly as much as disciplined edge. Risk-adjusted metrics — the Sharpe ratio, the Sortino ratio, the Calmar ratio, and maximum drawdown — are how professionals separate genuine skill from luck wearing a good month.
This guide explains each metric in plain language: how it is calculated, what a good value looks like in 2026, and when to trust it. By the end you will know which ratio to reach for in which situation, the mistakes that quietly make these numbers lie, and how to track them automatically instead of rebuilding spreadsheets after every trade.
Table of Contents
Key Takeaways
Metric | What it measures | Reach for it when |
|---|---|---|
Sharpe ratio | Excess return per unit of total volatility | You want one standardized score to compare any two strategies |
Sortino ratio | Excess return per unit of downside volatility | Upside swings shouldn’t be punished — only losses should |
Calmar ratio | Annual return divided by maximum drawdown | You care most about surviving the worst peak-to-trough loss |
Maximum drawdown | Largest peak-to-trough equity decline | You need to know the pain — and capital — required to hold on |
What are risk-adjusted return metrics?
Risk-adjusted return metrics measure how much profit a strategy generates relative to the risk it takes to generate it. Instead of asking “how much did it make?” they ask “how much did it make per unit of risk?” — turning two strategies with different volatility profiles into directly comparable scores.
This matters because risk and return are not independent. Anyone can double their expected return by doubling their leverage; what they cannot do is double their return without changing their risk. A strategy that earns 15% with a 5% standard deviation is demonstrably better-engineered than one earning 15% with a 25% standard deviation, even though their headline numbers match. Risk-adjusted ratios make that quality visible in a single figure, which is why funds, proprietary desks, and serious retail traders evaluate every system this way before allocating a dollar of capital.
What is the Sharpe ratio and how is it calculated?
The Sharpe ratio is the excess return of a strategy divided by its total volatility. The formula is (Portfolio return − Risk-free rate) ÷ Standard deviation of returns. A higher Sharpe means more reward for each unit of risk; a Sharpe of 1.0 means you earn one unit of excess return for every unit of volatility you accept.
Developed by Nobel laureate William F. Sharpe in 1966, it remains the industry’s default yardstick because it is simple, standardized, and comparable across asset classes. To annualize a Sharpe computed from daily returns, multiply by the square root of the number of trading periods per year (about √252 for daily data). The risk-free rate — typically the yield on short-term Treasury bills, around 4–4.5% in 2026 — represents what you could earn with no risk at all, so the numerator isolates the return your strategy added on top.
Its weakness hides in the denominator: standard deviation treats upside and downside moves identically. A strategy that occasionally surges 8% in a day is penalized by the Sharpe ratio exactly as if it had lost 8%, even though no trader complains about outsized gains. That blind spot is precisely what the Sortino ratio was built to fix.
Sharpe vs Sortino: what is the difference?
The Sortino ratio is a variation of the Sharpe ratio that divides excess return by downside deviation only — the volatility of negative returns — instead of total volatility. The difference matters because it stops punishing a strategy for the one thing traders actually want: large gains.
Mechanically, the Sortino replaces the Sharpe’s standard deviation with the standard deviation of only those returns that fall below a target (usually zero or the risk-free rate). For a strategy with asymmetric returns — small frequent losses and occasional big wins, the profile of most trend-following and breakout systems — the Sortino ratio will be meaningfully higher than the Sharpe, and arguably more honest. For a strategy with symmetric returns, the two converge. As a rule of thumb, when a system’s Sortino sits far above its Sharpe, its volatility is concentrated on the upside, which is exactly where you want it.
What is the Calmar ratio?
The Calmar ratio is a strategy’s annualized return divided by its maximum drawdown over the same period. Where Sharpe and Sortino measure return against the wiggle of the equity curve, Calmar measures it against the single deepest wound — the worst peak-to-trough loss you would have had to endure.
This makes Calmar the metric of psychological and practical survival. A strategy with a 30% annual return and a 15% maximum drawdown has a Calmar of 2.0; one earning the same 30% but suffering a 60% drawdown scores just 0.5 — a warning that the path to those returns runs through capital destruction most accounts could never hold through. Usually calculated over a trailing 36-month window, Calmar is favoured by managed-futures and CTA allocators because drawdown, not standard deviation, is what actually forces traders to abandon a system at the worst possible moment.
Maximum drawdown: the survival metric
Maximum drawdown (MDD) is the largest percentage drop from a peak in account equity to the subsequent trough, before a new peak is reached. It answers the most visceral question in trading: “what is the worst this has ever felt, and how much money was on the line?”
Drawdown matters more than most newcomers expect because recovery is non-linear. A 50% drawdown requires a 100% gain just to break even; a 20% drawdown needs only 25%. The deeper the hole, the more mathematically improbable the climb out — which is why two strategies with identical Sharpe ratios can have wildly different real-world survivability if one quietly carries twice the drawdown. Always read maximum drawdown alongside the ratios above; it is the context that tells you whether a high score was earned safely or on borrowed time.
Drawdown suffered | Gain needed just to break even |
|---|---|
10% | 11.1% |
20% | 25% |
33% | 50% |
50% | 100% |
75% | 300% |
Sharpe vs Sortino vs Calmar: side by side
All three ratios reward return and penalize risk — they simply disagree on what “risk” means. Sharpe defines it as total volatility, Sortino as downside volatility, and Calmar as worst-case drawdown. Reading them together gives a far fuller picture than any one in isolation.
Sharpe ratio | Sortino ratio | Calmar ratio | |
|---|---|---|---|
Risk measure | Total standard deviation | Downside deviation | Maximum drawdown |
Formula | (Return − Risk-free) ÷ Std dev | (Return − Risk-free) ÷ Downside dev | Annual return ÷ Max drawdown |
Penalizes big gains? | Yes | No | No |
Best for | General comparison | Asymmetric / trend strategies | Drawdown-sensitive capital |
Typical window | Annualized | Annualized | 36 months |
Watch out for | Treats upside as risk | Needs enough downside data | Driven by a single worst event |
What is a good Sharpe, Sortino, or Calmar ratio?
As a general benchmark, a Sharpe ratio above 1.0 is considered good, above 2.0 is very good, and above 3.0 is excellent; below 1.0 is usually judged sub-optimal. Sortino ratios run higher than Sharpe for the same strategy, so the bar sits around 2.0+ for good, while a Calmar ratio above 1.0 is solid and 3.0+ is regarded as strong.
Treat these as guidelines, not gospel. Ratios are inflated by short or cherry-picked time windows, by ignoring transaction costs and slippage, and by backtests that never lived through a real bear market. A Sharpe of 3.0 measured over six lucky months means far less than a Sharpe of 1.2 sustained across several years and multiple regimes. Context — sample length, market conditions, and whether costs are included — decides whether a “good” number is actually good.
Rating | Sharpe | Sortino | Calmar |
|---|---|---|---|
Sub-optimal | < 1.0 | < 1.5 | < 1.0 |
Good | 1.0 – 2.0 | 2.0 – 3.0 | 1.0 – 3.0 |
Excellent | > 2.0 | > 3.0 | > 3.0 |
Common mistakes that make these metrics lie
The fastest way to be fooled by a risk-adjusted ratio is to compute it on returns that don’t reflect reality. The most common errors are ignoring trading costs, using too little data, and comparing strategies measured over different periods.
Excluding costs and slippage. A Sharpe ratio built on frictionless backtest fills can collapse once commissions, spread, and slippage are deducted. Always measure ratios on net, after-cost returns.
Too-short samples. Volatility and drawdown need time to reveal themselves. A ratio from three months of a bull market is noise, not signal.
Annualization games. Multiplying a flattering daily Sharpe by √252 can manufacture an impressive headline number from a fragile strategy.
Comparing across regimes. A 2021 Sharpe and a 2022 Sharpe were earned in different worlds; only compare strategies over identical windows.
Reading one metric alone. A great Sharpe paired with a 55% maximum drawdown is a trap. These ratios are designed to be read together.
Where Nvestiq fits
Nvestiq is a no-code, AI-powered platform for building, backtesting, and automating trading strategies without writing a line of code. Risk-adjusted metrics are only useful if you compute them consistently — and that is exactly the kind of bookkeeping that breaks down when you are tracking results by hand in a spreadsheet.
When you backtest a strategy in Nvestiq, the platform reports Sharpe, Sortino, Calmar, and maximum drawdown automatically on net, after-cost returns, so you compare systems on the same honest footing every time. Because the same engine runs your backtest and your live automation, the metrics you optimized against are the metrics you are held to in production — closing the gap between a strategy that looks good on paper and one that holds up when real capital is on the line. Instead of arguing with a spreadsheet, you spend your time on the part that actually matters: the edge.
Frequently Asked Questions
Is a higher Sharpe ratio always better?
Not always. A higher Sharpe is better only when the underlying returns are realistic — measured net of costs, over a long enough sample, and across varied market conditions. A sky-high Sharpe from a few lucky months, or one that ignores slippage, can be far weaker than a modest ratio earned consistently over several years.
What is the difference between the Sharpe and Sortino ratios?
The Sharpe ratio divides excess return by total volatility, penalizing big gains and big losses equally. The Sortino ratio divides by downside volatility only, so it ignores upside swings and rewards strategies whose risk is concentrated in profitable moves — a fairer measure for asymmetric, trend-following systems.
Can the Sharpe ratio be negative?
Yes. The Sharpe ratio turns negative whenever a strategy’s return falls below the risk-free rate. A negative Sharpe means you would have been better off holding risk-free Treasury bills than taking on the strategy’s volatility — a clear signal the system is not adding value.
What is a good Calmar ratio?
A Calmar ratio above 1.0 is generally considered solid, meaning annual return exceeds the worst peak-to-trough drawdown. Above 3.0 is strong and typical of well-managed systematic strategies. Because Calmar is driven by a single worst-case loss, always check it over at least a 36-month window.
Which risk-adjusted metric should I use?
Use them together. Start with the Sharpe ratio for a standardized comparison, switch to the Sortino ratio when a strategy has asymmetric, upside-heavy returns, and lean on the Calmar ratio and maximum drawdown when surviving the worst loss matters most. No single number tells the whole story.