Drawdown and recovery: the maths of losing streaks and climbing back

A drawdown is the fall in an account's value from its highest point to a subsequent low, usually expressed as a percentage of that peak. Recovery is the gain needed to get back to the peak, and it is always larger than the drawdown: a 10% fall needs about an 11% gain, a 25% fall needs 33%, and a 50% fall needs 100%. That asymmetry is why controlling the depth of drawdowns matters more to long-term survival than almost anything else in risk management.

This guide covers how drawdown is measured, the recovery maths, how losing streaks turn into drawdowns at different risk levels, how long recovery takes, and how to write drawdown rules before you need them. It assumes you already size trades by percentage risk, as described in the position sizing guide.

How drawdown is measured

Peak to trough

Drawdown is always measured from a peak. Take this illustrative sequence of account values:

• Start: $10,000
• After a good month: $11,000 (new peak)
• After a bad run: $9,350
• Recovering: $10,200
• New high: $11,400

The maximum drawdown in this sequence is from the $11,000 peak to the $9,350 trough: a fall of $1,650, which is $1,650 ÷ $11,000 = 15%. Notice it is measured from the peak, not from the starting balance. Against the starting $10,000, the trough looks like a 6.5% loss, which badly understates how the account behaved. Traders who measure from their deposit rather than their high-water mark routinely underestimate their drawdowns.

The drawdown ends when the account makes a new high above $11,000, which happens at $11,400.

Current drawdown and maximum drawdown

• Current drawdown is how far the account is below its peak right now. At $10,200, it is ($11,000 − $10,200) ÷ $11,000 = 7.3%.
• Maximum drawdown is the deepest peak-to-trough fall over the whole history. Here, 15%.

Both are useful. Current drawdown drives your rules in real time. Maximum drawdown tells you what the approach has actually put you through.

Balance or equity

Balance only changes when trades close. Equity includes open profit and loss. Measuring drawdown on balance flatters an account that holds losing positions open, because the losses are not counted until they are realised. Equity drawdown is the honest measure. The margin call and stop-out guide explains the difference between balance, equity and margin on your account screen.

Duration

Depth is only half the picture. Drawdown duration is the time from a peak to the point where the account surpasses it. An account can have a modest maximum drawdown but spend months below its high, which is psychologically draining and has a real opportunity cost. Record both the depth and the length of each drawdown in your trading journal.

The recovery asymmetry

If an account falls by a fraction d, the remaining balance is (1 − d) of the peak. To get back, it has to grow by d ÷ (1 − d). The numbers:

• Down 5% → needs 5.3% to recover
• Down 10% → needs 11.1%
• Down 15% → needs 17.6%
• Down 20% → needs 25%
• Down 25% → needs 33.3%
• Down 30% → needs 42.9%
• Down 40% → needs 66.7%
• Down 50% → needs 100%
• Down 60% → needs 150%
• Down 75% → needs 300%
• Down 90% → needs 900%

The curve is gentle at first and then steep. Small drawdowns are close to symmetrical: lose 5%, make 5.3%, and you are back. Beyond about 30% the required gain starts to run away from the loss, and beyond 50% recovery becomes a very large task for any approach.

A related effect is sometimes called volatility drag. A gain of 10% followed by a loss of 10% leaves you at 99% of where you started, not 100%. A gain of 20% followed by a loss of 20% leaves 96%. A gain of 50% followed by a loss of 50% leaves 75%. The bigger the swings, the more the compounding works against you, even when the percentage ups and downs look balanced. Large position sizes do not just risk large losses; they make it harder to keep gains.

Losing streaks become drawdowns

Most drawdowns come from runs of losing trades. With percentage risk sizing, each loss is a fixed fraction of the current balance, so a streak of n losses at risk r leaves (1 − r) to the power n of the starting balance. The resulting drawdown:

• 1% per trade: 10 losses ≈ 9.6% drawdown; 20 losses ≈ 18.2%.
• 2% per trade: 10 losses ≈ 18.3%; 20 losses ≈ 33.2%.
• 5% per trade: 10 losses ≈ 40.1%; 20 losses ≈ 64.2%.

Twenty losses in a row sounds extreme, but it is the wrong way to think about it. Drawdowns rarely come from one unbroken streak. They come from sequences like lose, lose, win, lose, lose, lose, win, lose, in which the occasional win is not enough to stop the slide. A strategy with a 40% win rate and a 1:2 risk-reward ratio has a positive expectancy and will still produce long stretches that look like this. The risk-reward ratio guide shows how likely long losing streaks are at different win rates; the short version is that they are a normal part of any approach and get longer as the win rate falls.

What this means in practice: your risk per trade determines how deep a normal bad run will go. At 1%, a bad run is uncomfortable. At 5%, an ordinary bad run can take out most of the account, and the recovery maths above turns that into a hole you may never climb out of.

Recovery takes time, not just gains

The asymmetry table tells you how large the gain needs to be. It does not tell you how long it will take, and that is often the bigger problem.

Here is an illustrative example. Suppose an approach has a positive expectancy of +0.2R per trade, meaning that on average each trade makes 0.2 times the amount risked. At 1% risk, that averages about 0.2% of the account per trade. (These are invented numbers for the example; real expectancy is uncertain, varies over time and is reduced by costs.)

• To recover a 10% drawdown (an 11.1% gain), it needs about 53 trades on average.
• To recover a 15% drawdown (17.6%), about 81 trades.
• To recover a 25% drawdown (33.3%), about 112 trades.
• To recover a 50% drawdown (100%), about 347 trades.

If you take around 20 trades a month, that is a few months for a 10% drawdown and well over a year for a 50% one, and that assumes the edge is real and holds up throughout. Averages also hide variation: some recoveries are faster and some much slower.

This is the practical case for small risk. Deep drawdowns do not just cost money; they cost time, and the long recovery period is when traders are most tempted to abandon their rules.

The trap of trying to win it back

The instinctive response to a drawdown is to increase size so that the next winner repairs more of the damage. The maths says this is backwards.

Consider the extreme version, doubling the risk after every loss, which is the logic of a martingale. Starting at 1% of the starting balance and doubling after each loss:

• Loss 1: 1%. Cumulative: 1%.
• Loss 2: 2%. Cumulative: 3%.
• Loss 3: 4%. Cumulative: 7%.
• Loss 4: 8%. Cumulative: 15%.
• Loss 5: 16%. Cumulative: 31%.
• Trade 6 would risk 32%.

Five losses in a row is an ordinary event for many strategies. After it, the next trade risks almost a third of the original account to win back the previous losses. One more loss and the account is down by more than 60%. Less extreme versions, "I'll just go 1.5% until I'm back", follow the same path more slowly.

Fixed percentage sizing does the opposite automatically: as the balance falls, 1% of it is a smaller amount, so each loss is smaller in money terms. That self-correcting feature is one of the main reasons the method is widely used. The trading psychology guide covers the emotional side of revenge trading and loss chasing.

Setting drawdown rules in advance

A drawdown rule is a decision you make while calm, to apply when you are not. The common structures:

Daily and weekly loss limits

Stop trading for the day once losses reach a set amount, such as 2% or three losing trades, and for the week at a larger figure. These limits stop one bad session from turning into a large drawdown, and they interrupt the cycle of loss and over-trading. Many proprietary trading evaluations use daily loss limits and maximum overall loss limits for the same reason; some measure the maximum from the starting balance and others from the highest equity reached (a trailing drawdown), which behaves very differently.

Stepped risk reduction

Reduce your risk per trade as the drawdown deepens, and restore it only as the account recovers. An illustrative set of rules:

• Normal: 1% risk per trade.
• 5% below peak: cut to 0.5% per trade.
• 10% below peak: stop trading live and review.
• Back to within 5% of peak: return to 1%.

Here is the effect. From a $10,000 peak, the account falls to $9,500, a 5% drawdown. Ten more losses at 1% would take it to about $8,592, a 14.1% drawdown. Ten more losses at 0.5% take it to about $9,036, a 9.6% drawdown, which also keeps it just short of the review trigger. Smaller size in a drawdown buys time to find out whether something has gone wrong.

The trade-off is that reduced size slows recovery if the drawdown was just bad luck. That is an acceptable price for most traders, because it avoids the much worse outcome if the drawdown was not bad luck.

A hard stop and review

At some level, stop trading live and go back to analysis. The review should ask specific questions: has the market behaved in a way the approach was not designed for? Am I following the rules, or have my entries, exits or sizes drifted? Are my results within the range the approach's history or backtest suggested? Trading on a demo account or at minimum size during the review lets you keep collecting data without adding to the damage.

Is it a normal drawdown or a broken strategy?

This is the hardest question in trading, and there is no formula that answers it cleanly. Some useful checks:

• Compare to expectations. If your records suggest the approach has had drawdowns of 10% to 12% before, a 9% drawdown is within its normal range. A 20% drawdown is not, and is a signal to stop and look.
• Compare the statistics, not just the balance. Has the win rate fallen? Has the average loss grown relative to the average win? Are costs higher than they used to be? Your journal should hold these numbers.
• Check your own execution. Many drawdowns that look like a strategy failure turn out to be rule-breaking: oversized trades, skipped stops, trades outside the plan.
• Respect sample size. Twenty trades is too few to judge anything. A strategy with a genuine edge can lose over that many trades, and a strategy with no edge can win.
• Assume the next maximum will be worse. A maximum drawdown from a backtest, or from your history, is the deepest one so far. A longer future gives more chances for a deeper one, and backtests usually understate real costs, slippage and gaps. Plan for a worse drawdown than you have already seen.

Gaps in particular can turn a controlled drawdown into a sudden one. The gap risk and weekend risk guide explains how to size for them, and the currency correlation guide explains how several correlated positions can produce one large loss on a single day.

Comparing approaches by drawdown

When you compare two ways of trading, total return on its own is misleading. An approach that made 30% with a 40% maximum drawdown along the way is very different from one that made 15% with a 6% maximum drawdown. Professionals often look at return relative to maximum drawdown for this reason; one well-known version, the Calmar ratio, divides annual return by maximum drawdown. You do not need the formal ratio, but you should always ask "what did it cost to get there?" alongside "what did it make?".

This matters when you see marketing for signals, copy-trading or trading bots. A track record without a drawdown figure, or with a short history that has not yet met a bad period, tells you very little. Our guides to copy trading and forex scam red flags cover what to look for.

Common mistakes

• Measuring from the deposit instead of the peak. It hides how deep drawdowns really are.
• Measuring on balance instead of equity. Open losses are still losses.
• Risking too much per trade. Risk per trade sets the depth of a normal bad run. Above 2% or so, ordinary streaks start to produce drawdowns that are very hard to recover from.
• Increasing size to recover. It deepens the hole faster than it fills it.
• No written rules. Rules invented in the middle of a drawdown are usually emotional and inconsistent.
• Abandoning a sound approach too early, or a broken one too late. Both come from not knowing what a normal drawdown looks like for your approach.
• Ignoring duration. A long, shallow drawdown can be as damaging to discipline as a short, deep one.

Drawdowns are part of trading for everyone. Most retail CFD accounts lose money, as our risk warning explains, and the guide to why most retail CFD traders lose shows that oversized positions are a recurring reason. Leverage magnifies both the depth of drawdowns and the speed at which they happen; the leverage and margin guide explains how.

Frequently asked questions

What is drawdown in trading?

Drawdown is the fall in an account's value from its highest point to a later low, expressed as a percentage of the peak. If an account rises to $11,000 and then falls to $9,900, the drawdown is $1,100 ÷ $11,000 = 10%. See the drawdown glossary entry for a short definition.

What is maximum drawdown?

Maximum drawdown is the largest peak-to-trough fall in an account's history, or over a test period. It shows the worst decline the account has experienced before making a new high, and is one of the main measures of how risky a trading approach has been in practice.

Why does a 50% loss need a 100% gain to recover?

Because the gain is calculated on a smaller balance. If $10,000 falls 50% to $5,000, you need to make $5,000 to get back, and $5,000 is 100% of the $5,000 you have left. In general, the gain needed is the drawdown divided by one minus the drawdown.

What is an acceptable drawdown?

There is no universal figure. It depends on your approach, your risk per trade and how much decline you can sit through without changing your behaviour. Many traders set a maximum they are willing to accept, such as 10% to 20% from the peak, and use reduced risk and review rules to try to stay within it. The important thing is to decide in advance.

How long does it take to recover from a drawdown?

It depends on the depth of the drawdown, your risk per trade, how often you trade and your real edge after costs. As an illustration, an approach averaging 0.2% of the account per trade would need roughly 53 trades to recover a 10% drawdown and roughly 112 to recover a 25% drawdown. Deeper drawdowns take disproportionately longer.

Should I reduce my position size during a drawdown?

Fixed percentage sizing already reduces your size in money terms as the balance falls. Many traders go further and cut their percentage risk once a drawdown passes a set level, restoring it as the account recovers. This slows recovery a little if the drawdown was bad luck, and protects the account if it was not.