Home / Risk Metrics / Maximum Drawdown
Drawdown metricBeginner

Maximum Drawdown

Maximum drawdown is the largest peak-to-trough percentage decline in account equity over a period, measuring the deepest loss an investor would have endured from a high-water mark before a new high was reached.

Quick answer: Maximum drawdown is the largest peak-to-trough percentage decline in account equity over a period, measuring the deepest loss an investor would have endured from a high-water mark before a new high was reached.

In simple words

Maximum drawdown answers one blunt question: from its best point, how far down did the account ever fall before recovering. If your equity peaked at ₹6,00,000 and later sank to ₹4,50,000, that ₹1,50,000 fall is a 25 percent drawdown. It is the single number that best captures the worst pain an equity curve inflicted, and it matters because deep falls are punishingly hard to climb back from.

Purpose

Maximum drawdown exists to quantify worst-case historical pain in one number, because average returns hide the depth of the hole a strategy dug, and depth is what tests both the account and the trader's nerve.

Visual explanation

Maximum Drawdown

An equity curve with its highest peak, the deepest trough below it, and the shaded peak-to-trough span that defines maximum drawdown.

Drawdown CurveEquity0% — high-water markmax drawdowntime →

Professional explanation

What maximum drawdown measures

Maximum drawdown (MDD) tracks equity against its running high-water mark, the highest value reached so far, and records how far below that mark equity ever fell. At each point the current drawdown is the percentage distance below the most recent peak; the maximum drawdown is the most negative of all these figures across the whole period. It is a path statistic, not an endpoint one, so unlike total return it depends on the entire sequence of gains and losses, not just where the curve started and finished. This is why a strategy can end the period up handsomely yet still carry a frightening maximum drawdown somewhere in the middle.

The asymmetry that makes depth deadly

Drawdowns punish asymmetrically: the gain required to recover is always larger than the loss that caused it. A 10 percent drawdown needs about 11 percent to recover, a 25 percent drawdown needs 33 percent, a 50 percent drawdown needs a full 100 percent, and an 80 percent drawdown needs 400 percent merely to break even. The formula is recovery gain = MDD ÷ (1 − MDD). This convexity is the mathematical heart of the whole discipline: keeping the maximum drawdown shallow is not caution for its own sake, it is what keeps recovery inside the realm of the possible rather than the miraculous.

Depth, duration and the underwater period

Maximum drawdown as a single number captures only depth. Two curves can share a 30 percent maximum drawdown while one recovers in three months and the other stays underwater for three years; the second is far harder to live with even though the headline number is identical. Serious analysis therefore reads maximum drawdown alongside drawdown duration, the time from the peak to the recovery of a new high, and the underwater time, the fraction of the period spent below a prior peak. The Ulcer Index and average drawdown exist precisely to capture the dimensions that a single deepest point ignores.

Sensitivity to the sample and the single worst event

Because it is defined by one extreme, maximum drawdown is fragile: it is entirely determined by the single worst peak-to-trough episode in the record, so it can change sharply if the window is extended by one crisis or trimmed to exclude it. A backtest that happens to miss March 2020 will show a flatteringly small maximum drawdown that says little about future tail risk. The metric is a historical fact about what did happen, not a bound on what can happen; the true worst case almost always lies beyond the worst case observed so far.

Why it drives position sizing and stop-trading rules

Maximum drawdown is the metric risk desks translate directly into limits. If a strategy has historically drawn down 25 percent, a desk assumes the future can be worse and sets a maximum tolerable drawdown, say 20 percent, at which trading halts for review, and it sizes positions so that a plausible losing streak stays within that budget. It also feeds risk-adjusted ratios such as Calmar (return divided by maximum drawdown), which reward strategies that earn their return with shallow drawdowns over those that earn the same return through violent ones.

Formula

Maximum drawdown = (Trough − Peak) ÷ Peak (the most negative such value); Recovery gain needed = MDD ÷ (1 − MDD)

Peak = the highest equity (high-water mark) reached before the decline; Trough = the lowest equity reached after that peak and before a new high; the result is a negative fraction, so a fall from 100 to 75 is (75 − 100) ÷ 100 = −0.25, a 25 percent drawdown. Maximum drawdown is the single most negative of all such peak-to-trough figures over the period. Recovery gain needed uses MDD as a positive fraction (0.25), giving 0.25 ÷ 0.75 ≈ 0.333, a 33 percent gain to break even.

Maximum drawdown vs Volatility as a risk measure

AspectMaximum drawdownVolatility (standard deviation)
MeasuresDeepest peak-to-trough loss actually enduredTypical dispersion of returns around the mean
Captures the tailYes, it is the worst realised episodeNo, it treats all deviations symmetrically
Path dependentYes, depends on sequence of returnsNo, order of returns does not change it
Blind spotIgnores how long recovery took and all other drawdownsUnderstates fat-tailed crash risk
Trader relevanceDirectly felt pain and account survivalStatistical, less intuitive to a trader

Practical example

Illustrative example (Indian market)

A Nifty swing trader starts with ₹5,00,000 and grows the account to a high-water mark of ₹6,40,000 over several months. A sharp correction and a losing streak then drag equity down to ₹4,80,000 before a new high is ever made. The maximum drawdown is (4,80,000 − 6,40,000) ÷ 6,40,000 = −1,60,000 ÷ 6,40,000 = −0.25, a 25 percent drawdown. To climb from ₹4,80,000 back to the old peak of ₹6,40,000 requires a gain of 1,60,000 ÷ 4,80,000 ≈ 33 percent, matching MDD ÷ (1 − MDD) = 0.25 ÷ 0.75. The trader who had instead capped the drawdown at 10 percent, a trough of ₹5,76,000, would have needed only about 11 percent to recover, which is why controlling the depth changes the recovery from hard to routine.

An account concentrated in leveraged Bank Nifty positions can register a 40 to 50 percent maximum drawdown in a single volatile expiry week, since SPAN plus exposure margin permits large notional on modest capital. A 50 percent drawdown demands a 100 percent gain to recover, so a week of oversizing can cost a year or more of disciplined compounding.

Advantages

  • Expresses worst-case historical pain in one intuitive, felt number
  • Directly maps to the recovery gain required, exposing the asymmetry of loss
  • Path dependent, so it captures risk that average return and endpoint metrics miss
  • Translates cleanly into a stop-trading limit and a position-sizing budget
  • Feeds risk-adjusted ratios such as Calmar that reward shallow-drawdown returns

Limitations

  • Blind spot: it captures only depth, ignoring how long the account stayed underwater and every drawdown other than the single worst
  • Determined by one extreme episode, so it is fragile to the sample window chosen
  • A historical fact, not a bound; the future worst case is almost always deeper than the worst observed
  • Says nothing about the frequency of painful drawdowns, only the deepest one
  • Not directly comparable across periods of different length or volatility regime

Why it matters in practice

  • It is the risk number traders feel most viscerally and the one that most often triggers abandoning a plan
  • Its recovery asymmetry is the core justification for keeping per-trade risk small

Common mistakes

  • Reading a small backtested maximum drawdown as a ceiling on future losses
  • Trusting a drawdown figure from a window that happened to miss a crash
  • Judging depth alone while ignoring how long the underwater period lasted
  • Assuming a 20 percent drawdown needs only a 20 percent gain to recover
  • Comparing maximum drawdowns of strategies run over different periods as if equal
  • Sizing positions so a normal losing streak already approaches the historical drawdown

Professional usage

Professional risk managers treat maximum drawdown as a planning input, never a guarantee. They assume the future drawdown will exceed the historical one, set a hard maximum tolerable drawdown at which the book is cut or trading pauses, and size positions so an ordinary losing streak consumes only part of that budget. They always read the depth alongside the drawdown duration and the underwater time, because a shallow but endless drawdown and a deep but brief one demand different responses, and they use Calmar to compare how much drawdown a strategy charged for its return.

Key takeaways

  • Maximum drawdown is the largest peak-to-trough fall in equity over a period
  • Recovery is asymmetric: the gain needed is MDD ÷ (1 − MDD), so 50 percent lost needs 100 percent back
  • It captures depth only, not how long the account stayed underwater
  • It is set by one worst episode, so treat it as a historical fact, not a future bound
  • Keeping it shallow is what keeps recovery realistic rather than miraculous

Frequently asked questions

What is maximum drawdown?
Maximum drawdown is the largest percentage fall in account equity from a peak (high-water mark) to a subsequent trough before a new high is reached. It measures the deepest loss an investor would have endured over the period, computed as (Trough − Peak) ÷ Peak and taken as the most negative such value.
How do I calculate maximum drawdown?
Track equity against its running high-water mark, compute the percentage below that mark at every point, and take the most negative figure over the whole period. For a fall from a peak of 100 to a trough of 75, the drawdown is (75 − 100) ÷ 100 = −25 percent.
Why does a drawdown need a bigger gain to recover?
Because the gain is measured on a smaller base after the loss. A 25 percent drawdown leaves 75 percent of capital, and recovering the lost 25 needs 25 ÷ 75 ≈ 33 percent. The formula is recovery gain = MDD ÷ (1 − MDD), and it grows steeply as drawdowns deepen.
What gain do I need to recover a 50 percent drawdown?
A full 100 percent. Losing half leaves half, and doubling what remains is required just to break even: 0.50 ÷ (1 − 0.50) = 1.00. This asymmetry is why deep drawdowns are so dangerous and why capping the maximum drawdown matters more than chasing returns.
Is maximum drawdown the same as volatility?
No. Volatility measures the typical dispersion of returns around their average and is symmetric, while maximum drawdown measures the single deepest peak-to-trough loss actually endured. A strategy can have modest volatility yet a severe maximum drawdown if losses cluster, so the two describe different risks.
What is the blind spot of maximum drawdown?
It captures only the depth of the worst episode. It says nothing about how long the account stayed underwater, how frequently painful drawdowns occurred, or how deep the second and third worst falls were. Two strategies with an identical maximum drawdown can differ enormously in how long recovery took.
Can maximum drawdown predict future losses?
No. It is a historical fact about what did happen, not a bound on what can happen. A future drawdown can easily be deeper, especially if the sample window missed a crisis, so prudent desks assume the future worst case exceeds the observed one.
How is maximum drawdown different from average drawdown?
Maximum drawdown is the single deepest peak-to-trough fall, while average drawdown is the mean depth across all drawdown episodes. The maximum reflects the worst-case tail; the average reflects the typical ongoing pain. Reading both gives a fuller picture than either alone.
What is a good maximum drawdown?
There is no universal figure, and any number presented as safe would mislead. Lower is better for survival, but it must be judged against the return earned and the trader's tolerance. A shallow drawdown with modest return may suit one trader while another accepts deeper drawdowns for higher growth.
How does maximum drawdown affect position sizing?
A desk sizes positions so that a plausible losing streak keeps the drawdown within a pre-set maximum tolerable level. If historical drawdown is 25 percent and the limit is 20 percent, sizing is reduced until stress scenarios stay inside that budget, because the recovery maths makes overshooting expensive.
Why is maximum drawdown sensitive to the test period?
Because it is defined by a single extreme episode, adding or removing one crisis from the window can change it sharply. A backtest that excludes a crash will report a small drawdown that understates real tail risk, so drawdowns should be examined across multiple periods.
What is drawdown duration?
Drawdown duration is the time from a peak to the point where equity recovers a new high. It complements maximum drawdown by measuring how long the account stayed underwater, since a shallow drawdown that persists for years can be harder to endure than a deep one that recovers quickly.
Does maximum drawdown include costs?
Only if the equity curve it is computed from already reflects brokerage, STT, GST, stamp duty and slippage. Drawdowns measured on gross equity understate the real pain, because costs deepen troughs, so always compute maximum drawdown on a net equity curve.
Should I ever look at maximum drawdown alone?
No. On its own it hides duration, frequency and the return that justified the risk. Read it alongside average drawdown, the Ulcer Index, drawdown duration and a risk-adjusted ratio such as Calmar so you see both the depth and the character of the pain.

Voice search & related questions

Natural-language questions people ask about Maximum Drawdown.

What is maximum drawdown in simple terms?
It is how far your account fell from its highest point to its lowest point before it recovered. It is the worst dip your equity ever suffered, shown as a percentage.
Why is a big drawdown so hard to recover?
Because you are gaining on a smaller amount. Lose half your money and you need to double what is left just to get back, which is far harder than the fall was.
What gain do I need after a fifty percent loss?
A hundred percent. Losing half leaves half, so you have to double what remains simply to break even. That is why deep drawdowns are so dangerous.
Does maximum drawdown tell me the whole risk?
No. It only shows the deepest fall, not how long you stayed down or how often painful dips happened. Read it with other numbers like average drawdown.
Can my future drawdown be worse than the past?
Yes, easily. Maximum drawdown is just the worst that has already happened, not a limit. Smart traders assume the future can be deeper and size for it.
Is a smaller maximum drawdown always better?
Usually for survival, yes, but you have to weigh it against the return earned. A shallow drawdown matters because it keeps recovery realistic instead of near impossible.

Sources & references

    Last reviewed 12 July 2026. Educational content only — not investment advice. Markets and rules change; verify current conventions with SEBI, NSE/BSE and your broker.

    ← Previous
    Maximum Favorable Excursion
    Educational content only — not investment advice. Examples use illustrative numbers and simplified models. Risk-management techniques reduce but never remove risk, and trading derivatives involves substantial risk of loss. See our Risk Disclosure and SEBI Disclaimer.