Glossary
Calmar Ratio
Return divided by maximum drawdown — the metric that maps cleanest to "how much pain to make this money." Underused outside CTA evaluation.
Sentivue Capital··3 min read
The Calmar ratio is annualized return divided by maximum drawdown. It expresses how much you earn per unit of worst-case loss — the closest single number to "how much pain to make this money."
Formula
Calmar = R_annualized / |MaxDrawdown|
Both terms expressed as percentages. Most implementations use a trailing 36-month window (the original Young, 1991 formulation).
Why Calmar earns its keep
- Drawdown is what allocators feel. Volatility is theoretical until it actualizes as a peak-to-trough loss. Calmar prices in lived experience.
- Robust to volatility scaling. A strategy and its 2× leveraged twin produce the same Sharpe but different Calmars — Calmar correctly punishes the levered version's deeper drawdown.
- Standard in CTA evaluation. Managed futures allocators screen on Calmar before Sharpe.
Common traps
- Sample-period sensitivity. A strategy with no major drawdown in the window scores artificially well. Always compare Calmar across strategies on identical windows.
- Single-data-point denominator. Max drawdown is one observation. Two near-identical strategies can have wildly different Calmars if one happened to have its worst loss inside the window and the other just outside.
- Mixing definitions. Some shops use peak-to-trough from a closed-trade equity curve, others from mark-to-market. Pick one and document it.