Max Drawdown

The largest peak-to-trough decline -- how to measure it and what it tells you about strategy risk.

What is Max Drawdown?

Max drawdown measures the largest peak-to-trough decline in your strategy's equity curve during the backtest. Think of it as the worst dip your strategy experienced -- if you imagine your equity curve as a hiking trail, max drawdown is the deepest valley you had to walk through.

It answers the question: "What was the biggest drop from a high point before the strategy recovered (or the backtest ended)?" This is one of the most important risk metrics because it tells you the worst-case pain you would have experienced if you had been following this strategy in real time.

The max drawdown is the deepest peak-to-trough decline (-30%), even though a shallower -15% drawdown occurred earlier.

How It's Calculated

Simula tracks the highest equity value seen so far (the "running peak") as it moves through the backtest candle by candle. At each candle, it calculates how far the current equity has fallen from that peak:

Drawdown at each candle

drawdown = (current equity - running peak) / running peak * 100

In plain English: take the difference between where the equity is now and the highest point it has reached so far, divide by that highest point, and multiply by 100 to get a percentage. The max drawdown is the most negative value across all candles in the backtest.

Because equity can only be at or below its peak at any given moment, drawdown values are always zero or negative. A drawdown of 0% means the equity is at a new peak. A drawdown of -25% means the equity has fallen 25% from its highest point.

Max drawdown is shown as a negative number

Simula displays max drawdown as a negative percentage (for example, -15.00%). The negative sign indicates a decline. A max drawdown of -15.00% means the equity fell 15% from its peak. A max drawdown of 0% means the equity never dropped from a peak -- it only went up.

Worked Examples

Single drawdown

A strategy's equity curve moves through these values:

Starts at $10,000
Rises to $12,000 (new peak)
Drops to $9,000
Recovers to $13,000

The worst drop was from $12,000 to $9,000. Drawdown = (9,000 - 12,000) / 12,000 * 100 = -25.00%.

Multiple drawdowns -- the deepest one wins

A strategy's equity curve hits several dips:

Rises to $15,000 (first peak)
Drops to $10,500 -- drawdown = (10,500 - 15,000) / 15,000 * 100 = -30.00%
Recovers to $16,000 (new running peak)
Drops to $12,800 -- drawdown = (12,800 - 16,000) / 16,000 * 100 = -20.00%

Two drawdowns occurred: -30.00% and -20.00%. The max drawdown is -30.00% because it was the deepest.

Reading Your Results

Max drawdown of 0%

The equity curve never dipped from a peak -- it only went up (or stayed level). This is extremely rare in practice.

Max drawdown between 0% and -10%

A relatively shallow worst-case decline. The strategy avoided large dips during the backtest.

Max drawdown between -10% and -30%

A moderate worst-case decline. This is common for strategies that hold positions through normal market fluctuations.

Max drawdown deeper than -30%

A significant worst-case decline. The strategy experienced a large drop from its peak at some point during the backtest.

Max drawdown at -100%

The strategy lost all of its equity. This is the floor -- drawdown cannot go below -100%.

Things to Watch Out For

Max drawdown captures a single worst event

Max drawdown only tells you about the single deepest decline. A strategy could have one bad stretch and otherwise be smooth, or it could have many drawdowns nearly as deep as the max. This metric alone does not tell you how often drawdowns happen or how quickly the strategy recovers from them.

Recovery is not shown

A max drawdown of -20% tells you how far the equity fell, but not how long it took to recover -- or whether it recovered at all before the backtest ended. Two strategies can have the same max drawdown, but one might recover in a day while the other takes weeks.

Max drawdown depends on the backtest period

The max drawdown you see is the worst decline that happened during the specific period you backtested. A longer backtest or a different time period might reveal a deeper drawdown that your current backtest did not encounter. A short backtest may not have been tested through enough market conditions to reveal the strategy's true worst case. Monte Carlo simulation can estimate how likely deeper drawdowns are by reshuffling your trades across thousands of scenarios.

Leverage amplifies drawdown

If your strategy uses leverage, drawdowns are amplified proportionally. A 10% price decline becomes a 20% equity decline at 2x leverage, and a 50% equity decline at 5x leverage. Simula shows max drawdown for each leverage level separately so you can see this effect directly.

Calmar Ratio -- divides annualized return by the absolute value of max drawdown. This tells you how much return you earned per unit of drawdown pain.

Total and Annualized Return -- the return figures that the Calmar Ratio pairs with max drawdown to form a risk-reward picture.

Sharpe Ratio -- measures return per unit of overall volatility rather than per unit of worst-case decline. A strategy with a good Sharpe can still have a deep max drawdown if it had one bad stretch in an otherwise smooth equity curve.

Volatility -- measures the overall variation in returns. Max drawdown focuses specifically on the worst decline, while volatility captures the typical size of ups and downs.

Monte Carlo: Interpreting Results -- shows the distribution of max drawdowns across thousands of reshuffled trade sequences, revealing how much worse the drawdown could be with a different ordering of the same trades.