With a 40% win rate, the most likely losing streak over 100 trades is 7 in a row, and it reaches 11 or more about one time in eight. That does not mean your method is broken: it is what probability predicts, before your edge even enters the picture.
On a Thursday at 3 p.m., you take your fourth consecutive loss. Your setup was clean, your stop respected, your size correct. Yet a question settles in: what if the system is the problem? After the sixth loss, you change a rule. After the seventh, you skip the next trade, which would have been a winner.
Most abandoned methods share this pattern: the streak was within the norm, but nobody knew the norm.
How many losses in a row should you expect?
It depends on your win rate and the number of trades. Over 100 independent trades, here is what a simulation of 40,000 runs of 100 trades per win rate gives (how to read it: at 50%, the most likely losing streak is 6):
| Win rate | Most likely streak | Exceeded about 1 time in 10 | Probability of at least 7 losses in a row |
|---|---|---|---|
| 60% | 4 | 6 | about 9% |
| 50% | 6 | 8 | about 32% |
| 40% | 7 | 11 | about 69% |
| 33% | 9 | 14 | about 90% |
"Most likely streak" is the median of the 40,000 simulations. "Exceeded about 1 time in 10" is the value that roughly 10% of runs go beyond. Read the row that matches your actual win rate, measured on your recent trades, not the one you hope for.
Two immediate consequences.
A 50% win rate protects you from nothing. Six losses in a row are more likely than four or five. The intuition ("a fair coin alternates") is wrong: chance produces streaks, and a trader who mistakes them for an anomaly experiences them as a betrayal.
A low win rate is not a flaw. At 33%, nine losses in a row are the norm, not the exception. A system that wins 33% of the time with gains worth more than twice the average loss is perfectly viable, but it requires enduring streaks the table calls long. This is why win rate alone is a trap: it says nothing about the gain-to-loss ratio, nor about what a streak costs to live through.
Why are streaks longer than we think?
Because intuition reasons about the average, not the worst stretch. At a 40% win rate, you lose 6 trades out of 10: over 100 trades, 60 losses are scattered at random. With 60 losses to place and only 40 wins to separate them, it is almost impossible for all of them to stay isolated.
The calculation also counts opportunities. A 7-loss streak can start at any trade: the larger the sample, the more chances a long streak has to appear. The table applies to 100 trades, which is two to four months of activity for many traders. Over 500 trades, streaks get longer still.
What does a streak cost your account?
It depends on your risk per trade. Example on a 10,000 account, with a fixed percentage risk on the remaining capital:
| Losses in a row | At 1% per trade | At 2% per trade |
|---|---|---|
| 3 | −3.0% (9,703) | −5.9% (9,412) |
| 7 | −6.8% (9,321) | −13.2% (8,681) |
| 11 | −10.5% (8,953) | −19.9% (8,007) |
| 14 | −13.1% (8,687) | −24.6% (7,536) |
These figures are a calculation example (0.99 then 0.98 raised to the number of losses), not real results.
Two useful readings:
- The cost compounds, it does not add up. Losing 1% seven times is not −7% but −6.8%, because each loss applies to an already reduced balance.
- Recovering takes more than you lost. After −10.5%, you need +11.7% to get back to your starting capital. After −19.9%, you need +24.9%. The gap widens fast, which is why streaks are so dangerous at 2% and bearable at 1%.
Take a maximum loss limit of 10%, like the one some challenges set (example value). At 1% per trade, 11 losses in a row hit it. With a 40% win rate, 11 or more consecutive losses happen about one time in eight over 100 trades. A streak within the norm can therefore be enough to fail a challenge if risk per trade is poorly calibrated. The table also helps you choose that risk.
Compare your current streak to this table →
How do you tell a normal streak from a worrying one?
Compare it to the row for your win rate. As long as the streak stays below the "exceeded about 1 time in 10" column, it says nothing about your setup. Beyond it, that is not a verdict, it is a reason to review: a healthy system exceeds that threshold in fewer than one case out of ten, which is still frequent enough that you should not conclude from a single stretch.
The rule that prevents emotional decisions comes down to three points, to be written down before the next streak:
- The review threshold. For example: if my streak exceeds the "1 time in 10" value for my win rate, I review my latest trades.
- The time of the review. Outside the session, with a cool head, never between two trades.
- What gets measured. Were the setups consistent with the plan, the stops placed as intended, the size respected? If so, the streak is noise. If not, the problem is execution, not the system.
This distinction is essential: a streak of losses that followed the plan is a statistical fact, a streak of losses that broke the plan is an execution error. Only the first belongs in the table.
The limits of this table
A clean calculation can mislead if its assumptions are false. These ones partly are:
- Trades are assumed to be independent. In reality, two trades taken the same day on the same market are correlated: an unfavorable regime produces clusters of losses. Real streaks are therefore usually more irregular than those in the table.
- The win rate is assumed constant. A system performs differently depending on market conditions. The table describes a stable system.
- The estimated win rate is itself uncertain. With 30 trades, your 40% may be 30% or 50%. On a small sample, read the two neighboring rows.
- The table says nothing about the future. It describes what pure chance produces at a given win rate, nothing more.
This content is educational and is not investment advice. Trading involves a risk of capital loss.
Measuring your streak on Trade Stack
The Overview in Trade Stack shows your win rate, your winning and losing trades, and your current streak. You compare it with the row of the table that matches your win rate to see whether it is within the norm. For a reliable reading, log every trade, including the ones you would rather forget: a log with no missing trades gives you an honest win rate.
See my current streak on Trade Stack
Key takeaways
- Over 100 trades, the most likely losing streak is 4, 6, 7 and 9 for win rates of 60, 50, 40 and 33%.
- A streak beyond the "1 time in 10" column justifies a cool-headed review, not an immediate change of method.
- The cost of a streak compounds: 11 losses in a row is −10.5% at 1% risk per trade, −19.9% at 2%.
- Separate the streak that followed the plan (data) from the streak that broke it (an execution error).
- Write your review threshold before the streak, not during it.
Frequently asked questions
How many losses in a row are normal with a 50% win rate?
Over 100 trades, 6 in a row is the most likely, and a streak of 9 or more shows up about one time in ten. Over 500 trades, expect more.
Should you stop trading after several losses in a row?
Not because of the streak itself. Set a daily loss limit or a maximum number of trades in advance, and stop if it is reached. The rule protects your capital and your state of mind without claiming the system is broken.
Does a high win rate prevent long streaks?
It shortens them but does not remove them. At 60%, 4 losses in a row is the most likely value, and 7 or more remain possible about one time in eleven.
Method: simulation of 40,000 runs of 100 independent trades per win rate (fixed seed), longest losing streak recorded in each run. "Most likely streak": median. "Exceeded about 1 time in 10": 90th percentile. Capital cost: (1 − risk)^n, example on 10,000.
