What risk actually means
What is a drawdown, and why does it matter?
A drawdown is the fall from a previous high down to the low that follows it, measured as a percentage. Its duration — the time spent below that high before it is regained — is a second measure, and the two are read together to understand what holding the investment involved.
7 min de leitura · Atualizado em 2026-09-07
O memo
A drawdown measures the fall from the highest point an investment has reached down to the lowest point after it. How long it then stays below that high is the second half of the picture.
Para guardar
- A return tells you where you ended. A drawdown tells you what you sat through.
- The fall is one measure. The time spent below the old high is another.
- Getting back to even takes a bigger gain than the fall that caused it.
- 100 → 70
- −30%
- 70 → 100
- +42.9%
A distinção
A drawdown is not volatility. Volatility describes how much something moves in both directions. A drawdown describes one direction only — down from a peak — and it adds the clock.
O erro comum
Reading "unrealised" as "not real". The market value has genuinely fallen; what a sale does is realise that fall on the position — it settles what that holding returned. What happens afterwards depends on what you do next. The duration matters because the decision is taken while the outcome is still open.
Em resumo — Two investments can end at the same place and be nothing alike to hold. The drawdown is the measure that reads the path rather than the finish.
A drawdown answers a question that a return cannot: not where did this end up, but what did it take to get there.
It is the fall from a high point down to the low point that follows, expressed as a percentage. If a portfolio reaches £120,000 and then slides to £84,000, that is a 30% drawdown. The high point is the reference, and it matters that it is the highest point reached so far rather than the starting value — the measure is always taken from the best the investment has ever done, which is also the number the person holding it remembers.
Depth is only half of it
Most explanations stop at depth. That is the visible half, and it is the one that gets quoted.
The second measure is duration: how long the value stays below its previous high before regaining it. There is a name for that stretch — time underwater — and it is the part that has to be held through, which is why it weighs on a decision in a way a percentage does not.
Two falls of the same depth can be entirely different to live with. One that reaches its trough in six weeks and recovers by the end of the year is unpleasant. One that reaches the same trough over two years and stays there for three more is a different kind of experience, and the difference does not show up in the percentage at all.
So the two are read together. The fall says how far; the duration says for how long.
Why recovery takes more than the fall
Here is the arithmetic that the depth figure hides, and it is the single most useful thing to carry away.
A fall and its recovery are not symmetrical, because the gain needed to get back is calculated on what remains — not on what you had before.
| Fall from the high | Gain needed to return to it |
|---|---|
| 10% | 11.1% |
| 20% | 25.0% |
| 30% | 42.9% |
| 40% | 66.7% |
| 50% | 100.0% |
The relationship is not a rule of thumb; it falls straight out of the arithmetic. Losing 30% leaves you with 70% of what you had, and taking 70 back to 100 requires 42.9%. Losing half leaves you needing a double.
This is why a deep drawdown is worse than it looks: the deeper the fall, the larger the return needed to undo it. Note what that does not say. A deeper drawdown requires a bigger recovery return; how long that return takes to arrive depends on the path of future returns, not on the depth. Depth sets the size of the climb, never its duration.
Unrealised is not the same as unreal
While a portfolio is in drawdown the market value has genuinely fallen. That part is not a matter of perspective. What has not happened is a sale, so the loss is unrealised: it can still be undone if the value recovers, and it is a sale that realises it on that position. A recovery may follow, and it may not — and taking part in one, if it comes, depends on what you hold from then on.
That distinction is usually presented as a technicality. It is not — it is the whole reason duration matters. If a drawdown were simply a number that later reversed on its own, its length would be a curiosity. What makes length costly is that the decision to hold or to sell is taken inside it, with no information about how much longer it will run, and with several months of evidence apparently saying the position was a mistake.
A drawdown, in other words, is less a measure of what happened to the money than a measure of what was asked of the person.
Where you will actually meet the number
You will meet drawdown figures in performance reports and fund or strategy factsheets, and in backtests, among other places — and they do not all mean quite the same thing.
On a factsheet, a maximum drawdown is the worst peak-to-trough fall over a stated period. That period is the part to read first. A maximum drawdown measured over a window containing a major market fall, and one measured over a calm stretch, are not comparable numbers — and the smaller of the two is not evidence of a safer approach. Whenever you are shown one, the first question is what window it covers.
In a backtest, the drawdown describes what the rules would have produced over historical data. It records what happened; it does not describe what comes next. That caveat is worth holding onto precisely because drawdown figures are so concrete — a specific percentage over a specific stretch feels like a measurement of the strategy, when it is a measurement of one path through one past.
What a drawdown does not tell you
It does not tell you whether an investment is good. A deeper drawdown is not automatically worse, and a shallow one is not automatically better — some approaches accept larger falls in exchange for a different shape of return, and that trade-off can be entirely reasonable for the right holder.
It does not tell you anything about the future. A maximum drawdown is a historical record, not a floor.
And it does not, on its own, say whether the return was worth the ride. Putting the two together — what was gained, against what had to be endured — is a separate measure, and the subject of its own lesson.
The mistake worth avoiding
The common error is to judge an approach on its final figure alone. A strong ten-year number that required sitting through a 40% fall for two years is not the same proposition as the same number reached quietly, because the first one is only collected by someone who held on through it. A deep or long drawdown puts real pressure on that decision, and abandoning the approach part-way realises the fall rather than remaining invested for a possible recovery.
Reading the drawdown is how you find out which of the two you are being shown.
Confira
Two strategies end a ten-year period at exactly the same value. One spent eighteen months more than 30% below its previous high; the other never fell more than 8%, and never for longer than a quarter. The final numbers are identical. Are the two equivalent for the person holding them?
Ver a resposta
No — and the reason is not a matter of taste. Both paths reach the same value, but only one of them has to be held through a long stretch where the evidence in front of you appears to say the decision was wrong. That is the moment at which a strategy is most likely to be abandoned, and abandoning it realises the fall instead of remaining invested through what comes next. The final figure compresses the whole path into a single endpoint; the drawdown restores the part of that path a holder actually has to live with. The second route asks far less of whoever holds it.
Perguntas frequentes
- What is the difference between a drawdown and a maximum drawdown?
- A drawdown is any fall from a previous high. The maximum drawdown is the deepest of them over a stated period — the worst peak-to-trough decline in that window. Because it depends on the window, a maximum drawdown is only meaningful when you know the period it was measured over.
- If a fund fell 50%, why does it need to gain 100% to recover?
- Because the gain is calculated on what is left, not on what you started with. £100 falling by half leaves £50, and getting £50 back to £100 is a doubling. The deeper the fall, the more lopsided this gets — which is why a drawdown's depth is not a linear measure of the damage.
- Does a drawdown mean I have lost money?
- The market value has genuinely fallen — that part is real. What has not happened yet is a sale, so the loss is unrealised: it can still be reversed if the value recovers, and selling is what realises it on that position. A recovery may follow, and it may not; whether you take part in one depends on what you hold from then on. The distinction is less about arithmetic than about which decisions stay open while you are still holding.
- Is a big drawdown always a bad sign?
- Not on its own. Some approaches accept deeper falls in exchange for a different pattern of returns, and a drawdown taken out of context says nothing about whether that trade-off suited the holder. What it does say is what holding it required — which is information no return figure carries.