Maximum drawdown explained

Maximum drawdown is the largest fall from a peak to a later low before a new peak was reached. It is given as a percentage. If an investment climbed to £10,000, fell to £6,000, and only later passed £10,000 again, its maximum drawdown over that period was 40%.

It answers a practical question that return figures do not: how bad did it get on the way?

How to calculate it

Walk through the values in date order. Keep track of the highest value seen so far. At each point, work out how far the current value is below that high, as a percentage. The largest of those percentages is the maximum drawdown.

In a spreadsheet, put the values in column A. In column B, keep a running maximum: in B2 type =MAX(A$2:A2) and fill down. In column C, type =A2/B2-1 and fill down. The smallest number in column C is the maximum drawdown, shown as a negative percentage.

Why it matters

Losses and gains are not symmetrical. As a hypothetical illustration: after a 20% fall you need a 25% gain to get back to where you started. After a 50% fall you need 100%. After a 75% fall you need 300%. The deeper the drawdown, the longer the climb back.

It also tests behaviour. Many people say they can sit through a large fall, and then sell near the bottom when it happens. A strategy's drawdown is a fair warning of what following it would have felt like.

What it does not tell you

How long the fall lasted. A drawdown that recovered in three months and one that took ten years can have the same depth. Look at recovery time too, if it is shown.

How often it happened. Maximum drawdown is the single worst case in the period tested. A different period could have a worse one.

The future. A ticker that has never fallen more than 30% can fall further.

Drawdown on our ticker pages

We show the average maximum drawdown across each ticker's backtests. On SPY it is minus 28.7%. On QQQ it is minus 38.6%. On NVDA it is minus 52.5%, and on AAPL minus 55.6%. GLD, the gold fund, comes in at minus 23.9%.

Figures are from the StrategyLens published cards, updated 29 October 2026. They are hypothetical backtests with no trading costs, before tax, and they will change as the cards are refreshed.

These are the falls inside the strategy backtests, not the falls of the stocks themselves. A strategy that spends time in cash can fall less than the stock, and that is one of the few things a timing rule can genuinely offer. The cost is usually a lower return, which is why we always show the drawdown beside the CAGR and the comparison with holding.

Using it well

Read maximum drawdown beside return per year. A higher return with a much deeper drawdown may not be worth it to you, and a lower drawdown with a much lower return may not be either. Only you can decide which trade-off suits you, and it depends on your goals, your time horizon and how much loss you can take.

Drawdown in a strategy versus in the stock

A timing strategy can have a smaller drawdown than the stock it trades, simply because it is in cash for part of the time. If a rule is out of the market during part of a crash, it misses part of the fall. That is a genuine benefit, and it is the main case for many timing rules.

But a smaller drawdown is not free. Being out of the market during part of a fall usually means being out during part of the recovery too, and recoveries often start sharply. Over a full cycle the strategy can fall less and still end with less money. That is the pattern in many of our backtests: a shallower worst fall, and a lower return per year than holding.

So when you compare drawdowns, compare the returns that came with them over the same dates. A strategy that fell 25% and grew at 2% a year has not clearly done better than holding something that fell 35% and grew at 10% a year. Which you would prefer depends on your goals and how much loss you can take.

Calmar ratio: return per unit of drawdown

One way to combine the two figures is the Calmar ratio: CAGR divided by the maximum drawdown, ignoring the minus sign. A strategy with a 6% CAGR and a 30% maximum drawdown has a Calmar ratio of 0.2. It is a rough guide, and like every single number it hides the path, but it makes a quick comparison between two results easier.

This is research, not financial advice. Past results, real or simulated, do not tell you what will happen next.

What is a good maximum drawdown?

There is no single good figure. It depends on the return that came with it and on how much loss you can take. Compare drawdowns only between things tested over the same period.

How is maximum drawdown different from volatility?

Volatility measures how much returns swing in both directions. Maximum drawdown measures only the worst fall from a peak, which is closer to the loss someone would actually have felt.

How much do you need to gain to recover from a drawdown?

After a fall of x%, the gain needed is x ÷ (100 − x), as a percentage. A 50% fall needs a 100% gain to get back to the start.