What Is Maximum Drawdown? Formula, Examples & How to Calculate It

Why maximum drawdown matters more than return

Two strategies can both return 40% in a year. One dips 15% along the way; the other plunges 45% before ripping back — and the second is far riskier, even though the endpoint is identical. Maximum drawdown (MDD) measures exactly that: the largest peak-to-trough decline in a price or equity curve, and it is the single most common risk statistic in fund reviews, prop firm evaluations, and strategy backtests.

The reason it dominates return in risk conversations is loss asymmetry. Drawdowns are multiplicative, so the way back up is always longer than the way down:

• −10% requires +11.1% to break even
• −20% requires +25%
• −30% requires +42.9%
• −50% requires +100%
• −70% requires +233.3%

A strategy that halves its value needs to double to get back — the return has to do twice the work. That is why a fund with a 40% return and a 50% drawdown is usually a worse bet than a fund with a 20% return and a 15% drawdown.

This guide explains the formula step by step, walks through a complete numeric example, covers drawdown duration and recovery, and gives context on what counts as "acceptable" MDD for different kinds of accounts.

The maximum drawdown formula

MDD is computed from a chronological price (or portfolio value) series. The idea: at every point in time, measure how far the current value sits below the highest value seen so far (the running peak). The maximum of those measurements is the maximum drawdown.

1. List the values in time order — one closing price per period, oldest first.
2. Track the running peak — for each value, the peak is the largest value from the start of the series up to and including that value.
3. Measure the drawdown at each point — drawdown = (peak − value) / peak, expressed as a percentage.
4. Take the maximum — the largest drawdown across all points is the MDD. The peak and the value that produced it are the MDD peak and trough.

So the core formula is:

MDD = (peak − trough) / peak × 100%

where the trough is the lowest value reached after the relevant peak. Note that the percentage is relative to the peak, not the starting value of the series — a portfolio that rises from 100 to 200 and then falls to 120 has a 40% drawdown (from the 200 peak), not the 20% "loss" against the original 100.

Worked example

Take a price series: 100, 110, 105, 90, 85, 95, 120, 115.

• The running peak moves 100 → 110, holds at 110 through the decline, then jumps to 120.
• Drawdowns point by point: 0%, 0%, 4.5%, 18.2%, 22.7%, 13.6%, 0%, 4.2%.
• The maximum is 22.7% = (110 − 85) / 110: peak 110, trough 85.
• Duration: the peak-to-trough leg took 3 periods (110 → 105 → 90 → 85).
• Recovery: the first value back at or above 110 is 120, two periods after the trough — the full episode lasted 5 periods.

For a quick direct calculation without a full series, the same formula applies to any peak and trough pair: a fund that peaks at $10,000 and falls to $7,000 has a 30% drawdown, and a current value of $8,500 is still 15% below the peak.

Drawdown duration, recovery, and current drawdown

Three related numbers complete the picture:

• Duration (peak to trough) — how long the decline itself took. A 30% drawdown over two weeks and a 30% drawdown over two years are very different risks: the first is a sharp event, the second suggests the strategy's thesis may be broken.
• Recovery time (trough to new peak) — how long it took the value to climb back to the original peak. This is where loss asymmetry bites; recovery is usually much longer than the decline, which is why "time in a drawdown" is itself a well-known psychological and operational risk.
• Current drawdown — the distance of today's value below the running peak. It tells you whether the account is in a drawdown right now and how deep, independent of history.

Two 20% drawdowns can have very different durations; the MDD percentage alone hides that, which is why serious risk reports pair MDD with duration and recovery time (and why the table under the headline number on the calculator matters as much as the number itself).

What counts as a "good" maximum drawdown?

There is no universal threshold — it depends on the strategy, the assets, and the account owner. Common reference points:

• Below 10% — low. Typical of diversified, low-leverage long-only portfolios.
• 10–25% — moderate. Normal for concentrated equity strategies and many long/short funds through a bear market.
• 25–50% — high. Common in leveraged strategies, crypto, and trend-following through adverse regimes.
• Above 50% — extreme. Rarely acceptable outside high-risk crypto or venture-style allocations.

Prop firms are the strictest context: most funding programs cap total drawdown around 5–10% of the account and daily loss around 3–5%, so a strategy with a historical MDD of 20% will never be funded there regardless of its returns. When evaluating a strategy for a specific account, compare its historical MDD against that account's hard limit — not against the market average.

A useful companion metric is the Calmar ratio = annualized return ÷ MDD. It normalizes return against the pain of the worst drawdown; practitioners often look for roughly 1 or above, and 3+ is considered strong. A 30% return with a 15% MDD (Calmar 2) usually beats a 50% return with a 50% MDD (Calmar 1) for most risk budgets.

Calculate maximum drawdown from your own data

The Maximum Drawdown Calculator runs the full analysis in the browser. Two modes:

• Series mode — paste a price or equity series (one value per line, or comma/semicolon separated; up to 10,000 points, oldest first). It returns the MDD percentage with the exact peak and trough values, the peak-to-trough duration, recovery status and recovery time, the current drawdown, a color-coded drawdown curve (yellow under 10%, orange 10–20%, red above 20%), a table of every individual peak-to-trough episode ranked by depth, and a risk assessment band.
• Direct mode — enter just the peak, trough, and current value to get the drawdown percentage and recovery state without a full series.

Everything runs client-side: the data you paste is never uploaded. If you want the day-by-day drawdown series itself rather than episode-level analysis, the simpler Drawdown Calculator produces the full drawdown table directly.

Mistakes that make drawdown numbers misleading

• Measuring against the starting value. Drawdown is always relative to the running peak, never the first value in the series. A curve that doubles and then halves is still up 100% from day one, but it "lost" 40% in drawdown terms.
• Mismatched marking. If the account or fund is marked to the close, use closing values; for crypto or intraday-traded accounts, intraday highs and lows show deeper drawdowns than closes do. Be consistent, and say which you used.
• Too short a history. MDD is a worst-case statistic; a three-year window simply may not contain the worst drawdown. Evaluate any strategy over at least one full market cycle.
• Ignoring duration. Two strategies with identical MDD% but very different recovery times are not the same risk — a 20% drawdown that takes nine years to recover is, in practice, a different strategy.
• Confusing MDD with VaR. Value at Risk bounds a loss over a fixed horizon at a confidence level; MDD is the realized worst peak-to-trough move in the data you actually have. They answer different questions and should be reported together, not instead of each other.