4 / 6 · 7 min
Losing streaks: seven in a row is normal, not broken
A profitable system necessarily goes through long losing runs. Whoever does not know this in advance abandons it at exactly the wrong moment.
Someone with +0.4R expectancy — that is, a good working system — will almost certainly see six to eight consecutive losses within a hundred trades. Not because something broke. Because that is how randomness behaves.
Where long runs come from
If your hit rate is 40%, the chance of losing any given trade is 0.6. The chance of losing six in a row is 0.6 to the sixth ≈ 4.7%. Sounds rare. But a hundred trades contain ninety-five such six-trade windows, and the probability that at least one of them produces the streak exceeds 98%. A rare event becomes near-certain when there are many attempts.
Longest losing streak to expect in 100 trades
| Hit rate | Ordinary streak | Not unusual |
|---|---|---|
| 60% | 4–5 | 7 |
| 50% | 6 | 9 |
| 40% | 7–8 | 11 |
| 30% | 10–11 | 15 |
What a person feels at the seventh loss
At 1% risk, seven losses in a row is minus 6.8% of the deposit. Small in itself. But by the seventh trade the person is already certain that "the system stopped working", "the market changed", "something must be done". They increase size to win it back, or drop the system and take a new one. Both reactions turn an ordinary statistical run into a real disaster. The system meanwhile has not changed one bit.
Changing the system after a losing run
The most expensive mistake in this course, because it destroys not money but the very possibility of building anything. Someone trades a system for forty trades, hits a streak, abandons it and takes another. Forty trades later, the same thing. Within a year they have ten abandoned systems, none of them ever tested on a sample where statistics mean anything. The rule: a system is changed on the RESULT of a hundred trades, not on a feeling after seven.
Drawdown: two different things
Drawdown from a losing streak is simple: streak length × risk per trade. Seven losses at 1% is about 7%. But there is a second drawdown, far more treacherous: the maximum across your whole trading history. It is always larger than it seems, because it is built from several streaks separated by weak recoveries. When planning, take double your computed figure.
The two is not superstition but a correction for losing runs interleaved with partial recoveries. At 2% risk with a 40% hit rate, the computed streak is 7–8, so plan for a drawdown near 30%. If 30% is unacceptable to you, reduce risk per trade — there is no other lever.
What to do during a streak
Nothing. That is the most unpleasant and most correct answer. Do not increase size to win it back — that is a guaranteed way to turn a drawdown into a disaster. Do not change rules mid-flight. Do not hunt for "what broke". The one permissible action is to reduce size if the drawdown approaches a pre-set limit: not to win it back, but to survive until the recovery.
What is decided BEFORE a streak, not during
- What drawdown I consider normal and at which one I stop.
- How many trades must pass before I have the right to doubt the system.
- Whether I reduce size in a drawdown — and by exactly how much.
- What I do on a day with three losses in a row: continue, or close the terminal.
You can test all of this on yourself for free: run fifty paper trades at constant risk. A streak will certainly happen, and you will learn something about your own reaction that no text can teach.
Open paper tradingLive through the streak in advance
Take your risk per trade and compute how much money you lose on seven consecutive losses. Not in percent — in the money you actually feel. Now answer honestly whether you would carry on trading by the same rules on the eighth trade. If the answer is no, your risk per trade is too large, and it is far better to learn that now than at the seventh loss.
Hit rate 50%. What losing streak should not surprise you over a hundred trades?
Six in a row is ordinary; nine is not unusual. The chance of nine in a row for one specific window is 0.5⁹ ≈ 0.2%, but there are ninety-two windows, and at least one such run appears in roughly every fifth hundred trades. That is not the system breaking, that is the system behaving normally.
Drawdown has reached 25%. What do you do?
Whatever you decided BEFORE it started. If you had set a 20% limit, you should already have stopped and started reviewing. If there was no limit, there is no right answer now: any decision taken inside a drawdown is taken by someone whose judgement is impaired. Hence the only lesson for the future: limits are set in advance and written down.