Financial Market Tools for Traders, Investors & Finance Professionals
Learn / Backtest Evaluation — BT05
Maximum drawdown is the deepest peak-to-trough decline within the observed period. The currency and percentage formulas differ, and the gain required to recover is not symmetric with the loss that caused it. This guide walks through the maximum drawdown calculation, the peak, trough and recovery timeline, the underwater curve, and how to read the Calmar ratio, Ulcer Index and Recovery Factor across three lenses: depth, duration and frequency.
An equity curve that trends upward can still hide how deep the account sank along the way and how long it stayed underwater. Until you look at that dip, you cannot judge whether a strategy is survivable in practice. The central measure of that sinking is maximum drawdown. If you want the full picture first, the complete TradingView backtesting workflow sets out where this lesson fits in the wider process.
01 — Definition
Drawdown is the distance an equity curve falls from a previous high, or peak, down to its current value or trough. Maximum drawdown is the largest peak-to-trough decline within the observed period. In other words, it captures the worst fall measured from the best prior moment the account had reached.
The key point is that the peak is not a fixed value: it is updated every time equity sets a new high. At each point you keep a record of the running high, measure the fall from it, and take the deepest of those falls as the maximum drawdown. Even if price temporarily bounces, a deeper trough that appears later will update the maximum drawdown to that new low. For how each metric fits together, the guide on how to read backtest results is a useful companion.
02 — Formula
Maximum drawdown can be viewed in currency terms or in percentage terms. Every figure below is fictional educational data. Suppose the peak at one point is 1,200,000 and the subsequent trough is 900,000.
The currency view makes the scale of the loss easy to feel, but it is hard to compare strategies that run different amounts of capital. The percentage view does not depend on capital size and is easier to compare, which is why percentage is used most often when weighing strategies against each other.
A second important choice is which series you measure the drawdown on. Balance reflects only profit and loss from closed trades. Equity, by contrast, is a mark-to-market value that also includes the unrealized profit and loss of open positions. Strategies that hold through open losses tend to show a deeper trough on an equity basis, which is the more conservative and often more realistic view. Even for the same strategy, the balance-based and equity-based maximum drawdowns will not match, so always state which one you measured. This choice of series also shapes results in Monte Carlo testing of risk of ruin.
03 — Timeline
Maximum drawdown is not only a measure of depth. The figure below uses the same fictional dataset to mark the peak, trough and recovery point, along with two distinct spans of time.
Drawdown duration is the time taken to fall from the peak to the trough, and recovery duration is the time taken to climb back to the prior peak. Their sum, the time spent underwater (10 months in this example), is as important as depth when gauging how long capital is tied up and how heavy the psychological load becomes.
04 — Asymmetry
“Fall 20% and a 20% rise brings you back” is simply wrong. A loss shrinks your base, and the recovery is calculated on that smaller base, so the gain required is always larger than the loss. This is the asymmetry of drawdown.
| Loss | Remaining capital | Required recovery |
|---|---|---|
| 5% | 95% | about 5.3% |
| 10% | 90% | about 11.1% |
| 20% | 80% | 25.0% |
| 30% | 70% | about 42.9% |
| 50% | 50% | 100.0% |
| 60% | 40% | 150.0% |
While the fall is shallow, the gap between loss and required recovery is small, but it widens sharply as the fall deepens. A 50% loss needs the account to double, and a 60% loss needs a 2.5x rise. That is why avoiding even one deep drawdown carries such weight for the long-run equity curve. If you want to translate a loss into position size, the FX & CFD Lot Size Calculator converts a risk amount into concrete units.
Import your Strategy Tester CSV, XLSX or report and the Lab visualizes the equity curve, the maximum drawdown, the depth of the trough and how recovery unfolds. Start free and see exactly where the peak, trough and recovery point sit in your own data.
Visualize your drawdown free05 — Underwater
The underwater curve plots the percentage decline from the running peak at each point, always in the region at or below 0%. While the equity curve is setting new highs the line sits flush against the 0% surface, and when it falls the line dips below by exactly that amount. At a glance it shows not just how deep the account went but how long and how often it stayed underwater.
Two strategies can share the same 25% maximum drawdown, yet if one recovers quickly while the other stays underwater for a long stretch, the risk they feel like is entirely different. The underwater curve is valuable because it shows this time underwater and this frequency alongside depth, all at once.
06 — Ratios
Several metrics tie maximum drawdown back to return and risk. Each one is comparable only when the calculation definition and assumptions match. The figures below are fictional educational data.
It expresses how much annualized return was earned per unit of downside depth. For example, an annualized return of 18% and a maximum drawdown of 25% gives 18 ÷ 25 = 0.72. A higher value is read as greater efficiency relative to the fall, but because it shifts with the annualization method and the length of the observation window, compare only figures built on the same definition.
It shows how many times the net profit earned over the whole period covers the largest fall. For example, a net profit of 500,000 against a maximum drawdown of 300,000 gives 500,000 ÷ 300,000 ≒ 1.67. It captures how deep the intervening trough was relative to the profit generated.
It aggregates only the percentage decline from the peak as a root mean square. Where standard deviation measures dispersion symmetrically in both directions, the Ulcer Index focuses solely on the depth and duration of drawdowns. Because it does not count upside variation as risk, it is often described as closer to the felt discomfort of a sustained fall. The deeper and longer a decline runs, the larger the value grows.
07 — Shapes
The same maximum drawdown can take many shapes. Whether it is a single enormous loss, a long shallow decline, or a run of small repeated dips changes what you need to do about it. Splitting the view into these three lenses makes it easier to organize.
Axis 1 — Depth
How deep the trough goes. Deep drawdowns, where the required recovery rises steeply, weigh especially heavily on the equity curve.
Axis 2 — Duration
Time spent underwater. Even a shallow fall ties up capital and creates opportunity cost when recovery is slow.
Axis 3 — Frequency
How often declines above a set threshold occur. A high-frequency strategy carries a heavy load to keep running even when each fall is moderate.
The table below compares three fictional strategies across those three axes. Judged on the depth of the maximum drawdown alone, Strategy A looks the worst, but the picture changes once slow recovery and frequency are included.
| Strategy | Max DD | Avg. recovery | DD count (>5%) | Calmar | Ulcer |
|---|---|---|---|---|---|
| Strategy A (one huge DD) | 30% | 9 mo | 3 | 0.60 | 9.2 |
| Strategy B (repeated small DDs) | 15% | 4 mo | 8 | 0.85 | 6.1 |
| Strategy C (long shallow loss) | 18% | 14 mo | 5 | 0.70 | 11.5 |
Strategy C has only a moderate maximum drawdown, yet it takes the longest to recover and carries the highest Ulcer Index, revealing a heavy underwater load. Separating depth, duration and frequency surfaces qualities that a single number hides. When you handle several strategies together, the guide to multi-strategy portfolio backtesting covering correlation, allocation and combined drawdown is a useful next step.
08 — Tolerance
“What maximum drawdown is safe to accept” has no universal answer. The fall you can tolerate depends on capital size, objective, the horizon you plan to run, leverage, and your psychological capacity to keep going while holding the loss. Even the same 20% means different things depending on whether it is likely to recover quickly or to sit underwater for a long time.
So rather than judging by a single threshold, the practical approach is to weigh the required recovery, the recovery duration and the frequency and shape of drawdowns together. There is also a risk of overfitting making the drawdown look artificially small, so testing across separated periods is essential. The guide on walk-forward and out-of-sample analysis explains how to confirm behavior outside the fitted window. Costs often deepen the trough as well, so reading it alongside how to model slippage, spread and fees brings you closer to a realistic maximum drawdown.
09 — Tool
Enter a peak value and a trough value to calculate the drawdown amount, the drawdown percentage and the gain required to return to the prior peak. Values are processed only in your browser and are never sent anywhere. The static worked example below lets you follow the same logic even where the calculator does not run.
| Item | Value | Formula |
|---|---|---|
| Drawdown amount | 300,000 | 1,200,000 − 900,000 |
| Drawdown percentage | 25.0% | 300,000 ÷ 1,200,000 |
| Required recovery | 33.3% | 300,000 ÷ 900,000 |
With a peak of 1,200,000 and a trough of 900,000, the drawdown is 25.0%. Returning to the prior peak requires a 33.3% gain.
Even when the required recovery grows large with your inputs, that alone does not make a strategy good or bad. Use the tool to grasp, neutrally, the plain fact that a deeper fall demands a larger gain to undo it. When you want to go further into drawdown deep dives, rolling statistics and period-by-period robustness, higher plans that support saving, comparison and advanced validation cover it. Compare features and plans
FAQ
Next
Walk-Forward Analysis and Out-of-Sample Testing Guide →
The next step for confirming, across separated periods, whether your maximum drawdown recurs.
Features, free-plan scope and specifications may change. Confirm the latest details and pricing on the plan page.
