The Calmar ratio compares an investment or strategy’s annualized return with the magnitude of its maximum historical drawdown over a defined period. It asks how much annualized return was earned per unit of the worst observed peak-to-trough decline.
A higher ratio indicates more historical return relative to maximum drawdown, but it does not predict future performance or show the frequency, duration, or cause of losses.
$$
\operatorname{CalmarRatio}
=
\frac{
\operatorname{AnnualizedReturn}
}{
\left|\operatorname{MaximumDrawdown}\right|
}
$$
Maximum drawdown is calculated from a cumulative wealth or net asset value series. If \(V_t\) is value at time \(t\) and \(M_t\) is the prior running peak:
$$
M_t = \max_{s \leq t} V_s
$$
$$
D_t = \frac{V_t-M_t}{M_t}
$$
Maximum drawdown is the most negative \(D_t\) in the measurement window.
Key Takeaways
- The numerator and denominator must cover the same lookback period.
- Annualized return should state whether it is compounded return, arithmetic average, or another convention.
- Maximum drawdown depends on observation frequency, valuation data, fees, external cash flows, and the selected start and end dates.
- A ratio can improve because return rises or because the historical window drops an old drawdown.
- The denominator is one extreme observed path statistic, so estimates can be unstable.
- Calmar should be reviewed with volatility, downside deviation, recovery time, liquidity, leverage, and tail-risk measures.
Worked Example
Assume a strategy has:
- annualized compounded return: 18%
- maximum drawdown: -12%
- common three-year measurement window
$$
\operatorname{CalmarRatio}
=
\frac{18\%}{12\%}
=
1.5
$$
The strategy generated 1.5 units of annualized return for each unit of its worst historical percentage drawdown during that window.
This does not mean the strategy will earn 1.5 times its future drawdown, nor that a drawdown is limited to 12%.
Calculating Annualized Return
For beginning value \(V_0\), ending value \(V_T\), and \(Y\) years:
$$
\operatorname{AnnualizedReturn}
=
\left(
\frac{V_T}{V_0}
\right)^{1/Y}
-1
$$
This compounded approach differs from averaging periodic returns. A report should also state whether performance is:
- gross or net of fees
- adjusted for distributions
- in the investor’s base currency
- based on actual or backtested results
- affected by subscriptions and withdrawals
Why the Measurement Window Matters
Calmar ratios are often reported over a multi-year window, but there is no universal period for every use. Short and long windows create trade-offs:
- A short window may contain no meaningful stress.
- A long window may combine different managers, leverage, mandates, or market regimes.
- A rolling window can change sharply when a large drawdown enters or leaves.
- A strategy with a short live history may appear strong because it has not encountered a severe market.
Always disclose the dates, frequency, and whether the drawdown was fully recovered by the end of the period.
Calmar vs. Sharpe and Sortino
| Ratio | Return numerator | Risk denominator | Main focus |
|---|
| Calmar | Usually annualized return | Maximum historical drawdown | Return relative to worst observed peak-to-trough decline |
| Sharpe Ratio | Excess return | Standard deviation | Return relative to total volatility |
| Sortino Ratio | Return above a target | Downside deviation | Return relative to below-target variability |
The rankings can differ. A strategy with smooth returns and one deep drawdown may look favorable on Sharpe but unfavorable on Calmar. A volatile strategy with strong upside moves may have a weak Sharpe ratio but a less severe drawdown.
Drawdown Calculation Choices
Results depend on:
- daily, weekly, or monthly observations
- price return versus total return
- gross versus net performance
- treatment of external cash flows
- account-level versus composite results
- intraday losses omitted from closing data
- valuation smoothing for illiquid assets
- currency conversion
Monthly data can materially understate a drawdown that begins and recovers within a month. Illiquid or appraisal-based assets can also appear smoother than executable market values.
How to Evaluate a Calmar Ratio
Ask:
- Which dates define the lookback period?
- Is return annualized using compounding?
- Are the drawdown and return based on the same net-of-fee series?
- What observation frequency is used?
- Was the worst drawdown recovered?
- How long did recovery take?
- Did leverage, mandate, or management change?
- Is performance actual, hypothetical, backtested, or a composite?
- Does the history include stressed markets?
- What do other downside and liquidity measures show?
Risks and Limitations
- Single extreme observation: one drawdown determines the denominator.
- Window sensitivity: the ratio can change without current performance changing.
- No duration measure: two equal drawdowns can have very different recovery periods.
- No probability: the ratio does not estimate how likely a future drawdown is.
- Historical only: larger unseen losses remain possible.
- Return sensitivity: a negative or near-zero annualized return makes interpretation difficult.
- Zero denominator: no observed drawdown can make the ratio undefined or misleading.
- Smoothing: infrequent or model-based valuations can inflate the ratio.
Common Mistakes
- Using maximum loss in one period instead of peak-to-trough drawdown.
- Mixing a one-year return with a three-year drawdown.
- Comparing gross and net results.
- Ignoring frequency and intraperiod losses.
- Treating a higher ratio as a guarantee of better future risk-adjusted performance.
- Comparing a short live history with a long seasoned history.
- Omitting failed or discontinued strategies.
- Failing to disclose backtested or hypothetical performance.
Authoritative Context
CFTC and NFA materials define drawdown for specified commodity-pool and trading-advisor disclosures. They do not prescribe a universal Calmar-ratio methodology for all funds or portfolios.
- Ulcer Index: Uses all observed drawdowns, so both their depth and persistence affect the measure.
- Sharpe Ratio: Compares excess return with total return volatility instead of maximum drawdown.
- Sortino Ratio: Compares excess return with downside deviation below a selected target.
- Downside Risk: Provides the broader framework for shortfall, tail-loss, and drawdown analysis.
- Standard Deviation: Measures periodic return dispersion rather than the worst cumulative decline.
FAQs
What is a good Calmar ratio?
There is no universal good value. Interpretation depends on strategy, period, fees, leverage, liquidity, data frequency, and whether the history includes stress.
Can the Calmar ratio be negative?
Yes. If annualized return is negative and maximum drawdown is expressed as a positive magnitude in the denominator, the ratio is negative.
Does the Calmar ratio measure drawdown duration?
No. It uses maximum drawdown magnitude. Recovery time and time under water should be reviewed separately.
Educational Use
This article provides general financial education. It is not personalized investment, fund-selection, trading, performance, statistical, tax, legal, or risk-management advice.