Loss-streak probability.
A streak may look rare. A whole trading horizon gives it more chances. Compare one fixed block with the full horizon.
Chance across your whole horizon
54.61%
Chance of 6 consecutive losses within 100 future trades. Starting with no existing streak.
The model’s boundaryIndependent trades · fixed win probability · wins or losses only. A streak is not the same as account ruin.
Compare horizons, loss scenarios & assumptions
Reference probabilities at a 50% win rate. Every table cell starts with no existing streak, even when your main calculation carries one.
| Streak length | 25 trades | 50 trades | 100 trades | 200 trades |
|---|---|---|---|---|
| 3 losses | 85.99% | 98.27% | 99.97% | >99.99% |
| 5 losses | 31.16% | 55.19% | 81.01% | 96.59% |
| 6 losses | 15.78% | 31.46% | 54.61% | 80.09% |
| 8 losses | 3.69% | 8.36% | 17.02% | 31.97% |
| 10 losses | 0.83% | 2.04% | 4.41% | 8.99% |
Translate a streak into a loss scenario
Six new losses of 1% of the balance remaining before each loss reduce that block's starting balance by 5.85%.
A fixed 1% of the block's initial balance on every loss instead totals 6.00%. Neither includes fees, slippage, wins or a moving drawdown floor.
Challenge the win-rate assumption
Five percentage points on either side of your input, clipped to 0–100%. These are scenarios, not confidence bounds from a trade sample.
| Win rate | Probability |
|---|---|
| 45.00% | 72.90% |
| 50.00% | 54.61% |
| 55.00% | 36.29% |
Use the probability to understand the model, not to label a strategy broken or safe. Dependence, breakevens, changing position size and uncertainty about win rate are outside this two-outcome calculation.
Method, scope & reproducibility
Track the probability mass of paths that have not yet reached the target, grouped by the number of losses currently in progress. A win resets a path's current run to zero. A loss advances it by one; reaching the target absorbs that path exactly once.
S[j] ← (1 − winRate) × previous S[j−1]
hit ← previous hit + (1 − winRate) × previous S[k−1]
The initial mass is placed at your existing run length. The main result is the absorbed mass after the future-trade horizon. The separate fixed-block result remains (1 − winRate)k. Capital examples use 1 − (1 − risk)k or k × risk, and are not ruin probabilities.
Method: ps-losing-streak-1.0.0. JSON and Markdown exports include the actual inputs, result, method, assumptions and unavailable evidence. No simulation seed is used. Full product methodology.
Optional: simulate strategy survival
To test how wins, losses and position size interact with a drawdown floor, use this separate strategy simulator. Its inputs are independent of the losing-streak calculation above.
Max Drawdown in this widget is a static floor from starting balance (v1) — a fixed floor measured from your starting balance, not a floor that ratchets upward with a new equity peak. Streak probability is Bernoulli; floor hits in the widget still use this static Max DD.
The primary calculator above reports whole-horizon streak probability. The fixed-block table below answers only whether the next k trades all lose. This optional strategy simulator uses separate inputs to check a static drawdown floor; none of these results can be substituted for another.
Same engine at the same free accuracy — save a rule set, add a stop, or import a history. The stamped file is written after that run, not from this typed-average result.
The mechanism
At a 50% win rate, P(k losses) = 50%^k. The 6-loss row is 1.56%.
| Consecutive losses k | P(k losses) at 50% win rate |
|---|---|
| 3 | 12.5% |
| 4 | 6.25% |
| 5 | 3.13% |
| 6 | 1.56% |
| 8 | 0.39% |
| 10 | 0.1% |
A 6-loss streak at 1% risk is 6 lives consumed. Whether that reaches the floor depends on the allowance, not on this probability alone.
Independence is an assumption
The horizon tool and fixed-block table assume independent trades. Dependence or a changing win probability can change streak probabilities; this page does not estimate those effects from a trade sample. The separate strategy simulator also draws independent sequences from its entered win rate. The free personal-rule engine currently models a stated Max Drawdown allowance as a fixed floor measured from your starting balance, not a floor that ratchets upward with a new equity peak.
Risk of ruin calculator · methodology.
FAQ
- How is losing-streak probability calculated across a trade horizon?
- A finite-state recursion tracks paths by their current run length and counts each path once when it first reaches the target. The separate fixed-block probability remains (1 − win rate)^k; it answers a different question.
- Does this change the Monte Carlo simulator?
- No. The horizon tool computes a two-outcome losing-run probability without random sampling. The separate, unchanged simulator draws strategy sequences and checks a static drawdown floor. A losing streak is not itself account ruin.
- Does this say a streak will happen?
- No. It reports a probability under an independence assumption for numbers you enter. It does not predict a next trade.
Run it against your own numbers
The same engine, unlimited runs, at the same free accuracy as this page's calculator — open the full personal rule editor to save a rule set, add a profit goal or daily loss stop, or import a trade history.
Open the personal rule editor — free, full accuracy →Rules and outcomes vary by firm and by trader; nothing on this page is investment or trading advice, and PropSurvival is not affiliated with, endorsed by, or sponsored by any prop trading firm. This calculator models a stated rule set — it does not predict market outcomes.