The Calmar Ratio: CAGR Over Maximum Drawdown

Direct Answer

The Calmar Ratio divides a strategy's compound annual growth rate (CAGR) by the absolute value of its maximum drawdown, usually measured over a trailing 36-month window. A strategy earning 22% CAGR with a 15% max drawdown has a Calmar Ratio of about 1.47; a strategy earning a higher 28% CAGR but a 35% max drawdown scores only 0.80 — worse, once the worst-case pain behind the return is priced in.

The Calmar Ratio does not replace the Sharpe or Sortino ratios, and it is not a universal "best" performance metric. It answers one specific, intuitive question — how large was the annualized return relative to the single worst loss the strategy actually produced — that traders in trend-following, leveraged, and crypto strategies tend to care about more than a statistical volatility figure, because that worst drawdown is the number that determines whether they can psychologically and financially stay in the position.

Key Takeaways

• Calmar Ratio = CAGR ÷ |Maximum Drawdown|, conventionally computed over a trailing 36-month (3-year) window.
• A higher raw CAGR does not mean a better Calmar Ratio — a larger max drawdown can outweigh it entirely.
• Calmar uses one worst-case number (max drawdown), not a statistical dispersion measure like Sharpe and Sortino use.
• That makes Calmar intuitive but also sensitive to a single outlier period and to the lookback window chosen.
• Calmar and Sharpe can diverge sharply for strategies with low routine volatility but one severe drawdown.
• Always restate the CAGR and drawdown window used — a Calmar figure without its window is not comparable to another.

The Formula and the 3-Year Window Convention

The Calmar Ratio was introduced by Terry W. Young in a 1991 Futures magazine article, developed for evaluating commodity trading advisors (CTAs) whose managed futures track records often included exactly the kind of long, quiet stretches punctuated by sharp equity-curve declines that a simple average-return figure obscures. The name comes from the newsletter Young published it in, California Managed Accounts Reports, commonly abbreviated CMAR — "Calmar."

Calmar Ratio = CAGR ÷ |Maximum Drawdown|

Both inputs are measured over the same period. CAGR — compound annual growth rate — is the smoothed, geometric annualized return over that period; maximum drawdown is the largest percentage decline from a prior equity peak to a subsequent trough within the same period. Taking the absolute value of the drawdown keeps the ratio's sign tied to the sign of CAGR: a positive CAGR divided by a drawdown expressed as a positive magnitude yields an intuitively readable positive ratio.

The standard institutional convention, inherited directly from Young's original methodology, is a trailing 36-month window, recalculated monthly as the window rolls forward. Three years is long enough to have a reasonable chance of containing at least one real drawdown episode — a strategy evaluated only over a calm 6- or 12-month stretch may simply never have been tested — while staying short enough that the figure still reflects how the strategy behaves under roughly current market conditions, rather than a decade-old regime.

Shorter and longer windows are also used informally throughout the industry: a 12-month Calmar for a newer track record, a 5-year or since-inception Calmar for a long-running fund, or a Calmar recalculated over a specific historical stress period to answer a narrower question. None of these is wrong on its own, but none of them is directly comparable to a 36-month figure without saying so — the window is part of the number, not a footnote.

Practical checklist

• Confirm both CAGR and maximum drawdown are measured over the identical window before dividing one by the other.
• Default to a trailing 36-month window unless there's a specific reason to use a different one.
• State the window explicitly next to any Calmar figure you publish or compare.
• Treat a Calmar computed over a track record shorter than the intended window as provisional, not final.
• Recalculate on a rolling basis as new months of data arrive, rather than anchoring to a single fixed historical window indefinitely.

Common mistake: quoting a Calmar Ratio without stating the window it was computed over. A ratio calculated over a fortunate recent 12 months and a ratio calculated over a full 36-month cycle that included a real drawdown can look identical on the page while describing very different levels of tested risk.

Worked Example: Higher Return Isn't Always the Better Strategy

Consider two systematic strategies, both evaluated over the same trailing 36-month window.

Strategy A

CAGR (trailing 36 months) = 22%
Maximum drawdown (trailing 36 months) = 15%
Calmar Ratio = 0.22 ÷ 0.15 ≈ 1.47

Strategy B

CAGR (trailing 36 months) = 28%
Maximum drawdown (trailing 36 months) = 35%
Calmar Ratio = 0.28 ÷ 0.35 = 0.80

Judged on CAGR alone, Strategy B looks like the better strategy — a 28% annualized return beats a 22% one by a wide margin. But Strategy B's Calmar Ratio of 0.80 is worse than Strategy A's 1.47, and the reason is visible directly in the inputs: to earn that extra 6 points of annualized return, Strategy B's equity curve fell 35% from its peak at some point during the window, more than twice the depth of Strategy A's 15% drawdown, and disproportionately harder to recover from — a 35% drawdown needs roughly a 54% gain just to get back to the prior peak, while a 15% drawdown needs about 18%.

This is the reframing the Calmar Ratio forces: "which strategy is actually better" depends on what a trader would have had to sit through to earn the return, not just the size of the return itself. A trader running Strategy B lived through a stretch where more than a third of the account's peak value was gone, with no guarantee at the time that the drawdown wouldn't deepen further — a psychologically and financially harder position to hold than Strategy A's 15% dip, even though Strategy B's long-run number looks more impressive in a performance summary. For capital that has to survive the drawdown to capture the eventual return — most real trading and investing capital — the higher-Calmar strategy is frequently the more usable one, even at a lower headline CAGR.

None of this means Strategy A is automatically preferable in every case. A trader with a longer horizon, more risk tolerance, or the ability to add capital during a drawdown might rationally prefer Strategy B's higher raw return despite the deeper drawdown. What the Calmar Ratio does is make that tradeoff visible and comparable, instead of leaving it implicit in a CAGR figure that says nothing about the path taken to get there.

Why Calmar Matters for Trend-Following and Leveraged/Crypto Strategies

The Calmar Ratio was built for exactly the return pattern that trend-following strategies produce: long stretches of small, grinding gains or flat performance, interrupted by occasional sharp trend moves that drive most of the cumulative return, and occasional sharp countertrend moves or choppy regimes that drive most of the cumulative pain. A CAGR figure alone can look excellent for a trend-following CTA while saying nothing about how deep the equity curve dipped during the multi-month chop that preceded the profitable trend — and that dip is exactly what determines whether an allocator or trader stays invested long enough to capture the eventual gain.

Leveraged and crypto strategies raise the stakes on the same problem. Crypto assets carry substantially higher baseline volatility than most traditional asset classes, and leverage mechanically amplifies both the return and the drawdown from any given underlying price move. A leveraged crypto strategy can post a CAGR figure north of 100% while having produced that return by surviving — or not surviving — a 60% or 70% drawdown along the way. A return figure presented without its drawdown is close to meaningless for judging whether the strategy is actually usable, because a drawdown of that size is large enough to trigger margin calls, forced liquidations, or simple behavioral capitulation long before the eventual recovery arrives. Calmar puts the return and the worst realized pain on the same axis, in a single number that is far harder to present misleadingly than a CAGR figure in isolation.

Practical checklist

• Never evaluate a trend-following or leveraged strategy's CAGR without its matching max drawdown alongside it.
• Treat a very high CAGR paired with an unreported or vague drawdown figure as a red flag, not a selling point.
• For crypto and leveraged strategies, ask whether the reported max drawdown would have triggered a margin call or forced liquidation in practice.
• Recognize that a strategy's calmest historical stretch is not representative of its full risk profile — the drawdown-bearing periods are.
• Use Calmar as a first-pass filter, then examine the underlying equity curve directly before committing capital.

Common mistake: being drawn to a trend-following or crypto strategy's headline CAGR without asking what maximum drawdown produced it — the return and the drawdown are not independent facts, they are two views of the same underlying equity curve.

Calmar vs. Sharpe and Sortino

All three ratios divide a return figure by some measure of risk, but they define "risk" differently, and that difference has real consequences for how each one ranks the same strategy.

Sharpe Ratio  = (Return − Risk-Free Rate) ÷ Standard Deviation of Returns
Sortino Ratio = (Return − Minimum Acceptable Return) ÷ Downside Deviation
Calmar Ratio  = CAGR ÷ |Maximum Drawdown|

The Sharpe Ratio penalizes both upside and downside volatility equally, using the standard deviation of every return in the sample. The Sortino Ratio refines this by penalizing only downside deviation — volatility below a chosen minimum acceptable return — so an asset that swings upward sharply isn't penalized the same way one that swings downward is. Both are statistical measures built from every data point in the return series, which makes them relatively stable and less sensitive to any single period.

Calmar breaks from that pattern entirely. Instead of a dispersion measure calculated across the whole series, its denominator is a single realized number: the worst peak-to-trough decline that actually happened. That makes Calmar more intuitive for traders who think in terms of "how much pain did I have to sit through" — it answers that question literally, in the same units a trader experiences a drawdown in, rather than in a monthly standard deviation figure that has to be mentally translated into what a bad month actually felt like. But it comes at a real cost: because the denominator is one number rather than an average across many, Calmar can be dominated by a single outlier bad period that may never repeat, or that understates a future drawdown the historical record simply hasn't produced yet. It is also more sensitive to the lookback window than Sharpe or Sortino tend to be — extending or shortening the window can add or drop the one episode the entire ratio is built on, while Sharpe and Sortino, averaging across many periods, usually shift more gradually as the window changes.

A concrete case where Calmar and Sharpe diverge

Picture a strategy that sells short-dated volatility or runs a funding-rate carry position: on most days it produces small, steady gains with very low day-to-day standard deviation, which is exactly the kind of return series that produces a strong Sharpe Ratio — perhaps 1.8 or higher. Then, once during the 36-month window, a sudden volatility spike or funding-rate reversal forces the position to be closed at a loss, and the account draws down 40% in the space of a single bad week before the strategy resumes its normal low-volatility operation. Because that one week is a small fraction of the total sample, it barely moves the annualized standard deviation — the Sharpe Ratio stays strong. But that same week is, by definition, the single worst peak-to-trough decline in the entire period, so it sets the Calmar Ratio's denominator directly: a 40% max drawdown against even a healthy CAGR produces a weak Calmar figure. The strategy's Sharpe says "smooth and low-risk"; its Calmar says "one bad week nearly took out four in ten dollars at the peak." Both are true simultaneously, describing different aspects of the same equity curve — which is exactly why relying on either ratio alone, rather than reading them together, misses part of the risk picture. This pattern is common enough in short-volatility and options-selling strategies that it has an informal industry description: strong Sharpe right up until the tail event that Calmar is specifically built to expose.

Practical checklist

• Read Calmar alongside Sharpe and Sortino, not as a replacement for either.
• Watch for strategies with a strong Sharpe but a weak Calmar — that combination is a signature of tail risk that day-to-day volatility measures understate.
• Recompute Calmar over more than one window length before trusting a single figure, since it can shift more than Sharpe or Sortino when the window changes.
• Remember that Calmar's denominator is one historical event, not a statistical average — treat it accordingly when weighing how confident to be in the figure.
• Pair Calmar with a look at how long the strategy stayed underwater, since Calmar alone says nothing about drawdown duration, only depth.

Common mistake: assuming Calmar and Sharpe will always agree on which strategy is better. They frequently diverge, and the direction of the divergence — a good Sharpe with a poor Calmar — is itself informative about the shape of the risk a strategy is carrying.

Misconceptions Versus Reality

Common Mistakes

Most Calmar Ratio errors come from treating the number as self-explanatory rather than as a figure that depends heavily on exactly how and over what period it was calculated.

• Computing max drawdown over an unrepresentative short window. A strategy live-tracked or backtested only during a calm bull run that never produced a real drawdown will show an artificially low max drawdown and an inflated Calmar Ratio that says more about the window than the strategy's actual risk.
• Not restating the CAGR window used to compute Calmar next to the number. "Calmar Ratio: 1.47" means very little without specifying whether that's a trailing 36-month figure, a since-inception figure, or a cherry-picked favorable stretch — the window is not optional context, it's part of what the number means.
• Mixing backtested and live-traded data in the same drawdown calculation without disclosure. A backtest's drawdown and a live track record's drawdown are not measuring the same kind of risk, since a backtest can't fully capture execution slippage, liquidity constraints, or behavioral decisions made under real losses.
• Comparing Calmar Ratios computed over different periods as if they were directly comparable. Two strategies each showing "Calmar 1.5" mean different things if one number reflects a period that included a genuine crash and the other reflects a period that didn't.
• Treating Calmar in isolation instead of pairing it with Sharpe, Sortino, or drawdown duration. Each ratio answers a different question; using only one leaves real risk characteristics — volatility shape, recovery time — unexamined.

Risks, Limitations, and Exceptions

• Calmar Ratio is driven by a single historical episode — the worst drawdown — so it can be dominated by one non-repeatable event or fail to reflect a worse drawdown the track record simply hasn't produced yet.
• The ratio is more sensitive to the chosen lookback window than statistically averaged measures like Sharpe and Sortino, since extending or shrinking the window can add or remove the single episode the whole figure rests on.
• A short track record produces a provisional Calmar figure that may change substantially once a genuine drawdown cycle occurs within the measurement window.
• Calmar says nothing about drawdown duration, frequency of smaller drawdowns, or the shape of day-to-day volatility — all real risk characteristics it does not capture.
• Backtested drawdowns can understate real-world drawdowns once execution slippage, liquidity constraints, and behavioral factors during actual losses are accounted for.
• Leverage, margin requirements, and forced liquidation thresholds can make a historical max drawdown irrelevant if the account would have been liquidated before that drawdown fully played out.
• A Calmar Ratio computed from a small number of trades or a short live history carries wide statistical uncertainty even when the number itself looks precise.

Practical Implementation Checklist

1. Calculate CAGR and maximum drawdown over the same, clearly stated window — default to trailing 36 months.
2. Divide CAGR by the absolute value of maximum drawdown to produce the Calmar Ratio.
3. Publish the window alongside the ratio whenever it is shared or compared to another strategy.
4. Recompute Calmar over at least one alternate window (such as 12 months and since-inception) to check sensitivity.
5. Cross-check Calmar against Sharpe and Sortino for the same period, and investigate any strong divergence.
6. Review the drawdown's duration, not just its depth, using the strategy's underlying equity curve.
7. For leveraged or crypto strategies, confirm the reported drawdown would not have triggered a margin call or forced liquidation in practice.
8. Treat an unusually high Calmar Ratio as a prompt to check whether the window happened to exclude the strategy's worst historical stretch.

Tool Opportunity

A dedicated Swoopr performance metrics calculator should compute Calmar, Sharpe, and Sortino together from the same equity curve, so the three ratios can be compared side by side rather than sourced from different reports with different windows.

Recommended inputs: a full equity curve or periodic return series, the measurement window (with a 36-month default), the risk-free rate for Sharpe, and a minimum acceptable return for Sortino.

Expected outputs: CAGR, maximum drawdown, maximum drawdown duration, and the Calmar, Sharpe, and Sortino ratios for the selected window, plus the same figures recalculated over at least one alternate window for sensitivity comparison.

Validation requirements: reject return series too short to support the selected window, flag when a ratio is being driven by a single outlier period, clearly label backtested versus live-traded data, and never present any single ratio as a complete measure of strategy risk.

Frequently Asked Questions

Sources and Methodology

This guide prioritizes the metric's original published methodology and widely used institutional and educational references on risk-adjusted performance measurement. Principal source categories:

• Terry W. Young, "Calmar Ratio: A Smoother Tool," Futures magazine, 1991 — the original publication introducing the Calmar Ratio and its trailing-36-month convention for evaluating commodity trading advisors.
• CFA Institute curriculum materials on risk-adjusted performance measurement, covering drawdown-based and volatility-based ratios used in investment performance evaluation.
• Investopedia's reference entry on the Calmar Ratio, used as a general cross-check on the standard formula and common usage conventions.

The worked figures in this guide are hypothetical illustrations built to demonstrate the calculation, not historical performance claims for any specific strategy or product.

Conclusion

Calmar Ratio = CAGR ÷ |Maximum Drawdown|, typically over a trailing 36-month window, and it exists to answer a question CAGR alone cannot: how much annualized return did a strategy produce relative to the worst peak-to-trough pain it actually put a trader through. A higher raw return does not automatically win once that pain is priced in, and a single strong or weak Calmar figure is never the whole picture — read it alongside Sharpe, Sortino, and drawdown duration, and always alongside the window it was computed over.

Use this page as part of the larger Swoopr learning architecture. Move to the parent hub for broader orientation on performance metrics, or to a specific companion guide when a particular calculation or comparison is needed.

Related Reading

• Portfolio performance metrics hub — the parent guide this page belongs to.
• Compound Annual Growth Rate (CAGR) — the numerator of the Calmar Ratio, explained in depth.
• Maximum drawdown duration — how long a strategy stays underwater, a measure Calmar's denominator does not capture on its own.
• Sharpe Ratio — the classic volatility-based risk-adjusted return measure Calmar is most often compared against.
• Sortino Ratio — a downside-only refinement of Sharpe, and a third lens for the same return series.
• Portfolio risk: correlation, concentration, drawdowns, and portfolio heat — the broader risk-management framework these performance metrics feed into.