Definition
Calmar ratio
The Calmar ratio divides annualised return by maximum drawdown, pricing return in units of the worst loss. Formula, the 36-month window, and why short windows inflate it.
The Calmar ratio is compound annual growth rate divided by the absolute maximum drawdown over the same window, conventionally the trailing 36 months. It expresses return in units of the worst peak-to-trough loss a holder actually had to sit through, rather than in units of volatility as the Sharpe ratio does. A 20% CAGR with a 10% maximum drawdown is a Calmar of 2.0; the same 20% with a 40% drawdown is 0.5.
- Also known as
- return drawdown ratio · MAR ratio · drawdown ratio
How it is calculated
Calmar = CAGR ÷ |Maximum drawdown|
Conventional window: trailing 36 months
20% CAGR, 10% max drawdown → 20 ÷ 10 = 2.0
20% CAGR, 40% max drawdown → 20 ÷ 40 = 0.5
Expected max drawdown ∝ σ√T, so a shorter window inflates the ratio.
MAR ratio: the same formula over the full track record.
Worked example: 20% CAGR with a 10% maximum drawdown → 20 ÷ 10 = 2.0; the same 20% with a 40% drawdown → 0.5.
It prices return in the currency of the worst loss
Volatility counts every wobble, up or down, and weights them all the same. Maximum drawdown records one thing: the deepest fall from a peak before a new peak was made, which is the number that decides whether an investor stays in or is forced out. Calmar therefore rewards a strategy whose losses are shallow even if its gains are lumpy, and punishes one that is smooth for years and then loses a third of the account in a month. Read next to Sharpe, the pair says whether risk arrived as steady noise or as a single event. The Sharpe and max drawdown calculator on QANTERION reports Calmar beside Sharpe for the same series.
The window sets the number
Maximum drawdown is a single path-dependent observation, so Calmar is noisier than Sharpe and moves with the window it was measured over. Magdon-Ismail and Atiya show that the expected maximum drawdown of a Brownian path grows with the square root of time, so a 12-month Calmar is mechanically higher than a 36-month one on the same strategy and the two are not comparable. The MAR ratio applies the same formula over the full track record, which is the most demanding version. A strategy that trades more often also exposes itself to more chances of a deep drawdown, so trade frequency belongs in the comparison too.
How it gets misread
Calmar ratios are compared across different windows and across strategies with different trade frequencies as if they were one statistic, which they are not: shorten the window and the ratio rises without anything about the strategy changing. The second mistake is reading a high Calmar as low risk. It may only mean a short, calm history in which the worst drawdown has not happened yet, so a high figure over 12 months says far less than a moderate one over 36.
See also
Sources
- On the Maximum Drawdown of a Brownian Motion — Journal of Applied Probability (author copy, Rensselaer Polytechnic Institute)
- Drawdown Measure in Portfolio Optimization — International Journal of Theoretical and Applied Finance (author copy, Columbia University)
- Drawdown: From Practice to Theory and Back Again — arXiv (Goldberg & Mahmoud; published in Mathematics and Financial Economics)
Definitions are educational. Nothing here is investment advice, and no metric described on this page predicts future results.
Definitions are the easy part
Knowing what drawdown means is not the same as having a system that halts on it. QANTERION applies these limits while a strategy runs.