Calmar Ratio: Return Against Max Drawdown

The Calmar ratio divides a strategy's compound annual growth rate by its maximum drawdown, giving you a single number for return earned per unit of pain endured. A Calmar ratio of 2.0 means the strategy returned twice as much, annually, as the worst peak-to-trough loss it suffered. It matters because CAGR by itself tells you nothing about how deep the hole got before the strategy climbed out.

What is the Calmar ratio?

The Calmar ratio is a risk-adjusted performance measure built from two numbers you already track: compound annual growth rate and maximum drawdown. Divide the first by the absolute value of the second and you get the ratio. The name comes from California Managed Accounts Reports, the newsletter that popularized it in the late 1980s for evaluating commodity trading advisors.

The original version used a trailing 36-month window, and a lot of tearsheets still default to that period. There's nothing sacred about three years though. You can compute it over any stretch of the equity curve, as long as you're honest about which window you're using and consistent when you compare two strategies.

How to calculate the calmar ratio

The formula is CAGR divided by max drawdown, expressed as a positive number: Calmar = CAGR / |Max Drawdown|. Say a strategy compounded at 24% annually over three years and its worst drawdown during that stretch was 12%. Calmar ratio: 24 / 12 = 2.0.

Now compare a second strategy that compounded at 36% but drew down 30% at its lowest point. Calmar ratio: 36 / 30 = 1.2. The second strategy made more money on paper. The first strategy made more money per unit of drawdown risk, and that's the number the Calmar ratio is built to surface.

The Calmar ratio punishes the drawdown that CAGR happily ignores.

Why CAGR alone hides the real story

CAGR tells you where the equity curve ended up. It says nothing about the road it traveled to get there. Two strategies can post identical 30% annual returns while one lost 8% at its worst point and the other lost 45%. On a CAGR-only tearsheet, they look interchangeable. On a Calmar ratio basis, they are not close.

Drawdown math is unforgiving in a way that raw return numbers disguise. A 20% drawdown needs a 25% gain to get back to breakeven. A 50% drawdown needs a 100% gain. A 45% drawdown needs roughly 82%. The deeper the hole, the more disproportionate the climb, and CAGR never shows you the hole, only the final altitude.

This is the exact blind spot the Calmar ratio is designed to close. A strategy with a high CAGR and a shallow max drawdown has room to survive a bad stretch and keep compounding. A strategy with a high CAGR sitting on a deep drawdown is one bad month away from erasing years of gains, and the return number alone won't warn you.

Calmar ratio vs Sharpe ratio

Both are risk-adjusted return measures, but they punish different things. The Sharpe ratio divides excess return by the standard deviation of returns, so it treats every wiggle in the equity curve, up or down, as risk. A strategy with choppy but shallow volatility can post a mediocre Sharpe ratio even if it never suffers a serious drawdown.

The Calmar ratio only cares about the single worst peak-to-trough loss. It ignores day-to-day noise entirely and zeroes in on the one number that determines whether a trader (or a client) can actually stomach holding the strategy through its worst stretch. For strategies with fat-tailed, infrequent losses, like many trend-following or options-selling systems, Calmar often tells a more honest story than Sharpe, because Sharpe can be flattered by long quiet periods that mask a single catastrophic tail event.

Neither ratio replaces the other. A strategy with a strong Sharpe and a weak Calmar has smooth returns punctuated by one brutal drawdown. A strategy with a strong Calmar and a weak Sharpe is bumpy day to day but never falls off a cliff. Knowing which one you're looking at changes what kind of trader can actually run it.

Where the calmar ratio falls short

Max drawdown is a single historical event, not a distribution. A strategy that has only run for two years might simply not have encountered its worst-case scenario yet, and its Calmar ratio will look better than it deserves to. The ratio is only as reliable as the sample of market conditions the strategy has actually traded through.

The ratio also treats one bad drawdown identically whether it happened once in ten years or is a recurring pattern every eighteen months. Two strategies with the same Calmar ratio can have very different drawdown frequency, and frequency matters just as much as depth when you're the one watching the account. Look at the full drawdown history, not just the worst single number, before you trust a Calmar ratio on its own.