Risk of ruin calculator

Risk of ruin is the probability that a run of losses takes an account below a level it cannot come back from. It follows from three things: how often you win, how much you win relative to what you lose, and how much you stake each time. But your win rate is an estimate from a finite sample. No calculator on the first page of results for this term asks how large that sample was; each treats the rate as though it were the truth — so a trader with 40 trades and a trader with 4,000 get handed the same answer. This asks how many trades yours came from, and reports the range that sample actually supports.

Why the number of trades is the field that matters

A win rate is not a property of a strategy you can read off. It is an estimate, and its precision depends entirely on how many trades produced it. Fifty-five percent from sixty trades and fifty-five percent from six thousand are the same number carrying wildly different amounts of evidence — the first is consistent with a true rate anywhere across a broad band, the second pins it down tightly.

This matters here more than almost anywhere, because risk of ruin is violently non-linear in the win rate. A few points of difference can move the answer by an order of magnitude, and those few points are exactly what a small sample cannot resolve. So a calculator that accepts a win rate and returns one number is not answering the question asked; it is answering a different, easier question — what would my risk of ruin be if my estimate happened to be exactly right — and presenting the answer as though the conditional were not there.

The interval used here is the Wilson score interval, which is the standard choice for a proportion and behaves properly near the extremes where the simpler normal approximation breaks down and can hand you a bound below zero. How many trades it takes before an edge is established at all is its own question.

What the simulation does

Each simulated trade is a win with your stated probability and a loss otherwise. On a loss the account gives up the staked fraction; on a win it gains that fraction multiplied by your payoff ratio. A path is ruined the moment it falls below your threshold, and the risk of ruin is the proportion of paths that do. The same is then run at both ends of the interval around your win rate.

It is deterministic: the generator is seeded, so identical inputs always produce an identical answer. That is deliberate. A tool that returned a slightly different figure on every keystroke would look broken, and worse, it would invite re-rolling until the answer felt acceptable — which is the precise habit this site exists to argue against.

Where no simulated path is ruined, the panel says under 1 in 4,000 rather than nought percent. Those are different statements. The first is what was measured; the second is a claim about a probability being zero, which a finite simulation cannot establish and which is not true of any account that trades.

What this does not do

The model assumes your trades are independent and identically distributed, and real trading is neither. Losses cluster, volatility arrives in regimes, and a strategy's edge is not constant across them — see regime shift. Correlated positions make several trades behave as one, which this does not model. Every one of those effects pushes real risk of ruin above the figure here, so treat the output as a floor rather than an estimate.

It also cannot tell you whether your win rate and payoff ratio are honest. If they came from a backtest, they are biased optimistic before anything else is wrong with them, because the configuration was selected partly for having done well in that very data — the domain of backtest overfitting and data snooping. Execution costs also come out of the edge before it reaches the account, and they are not in these inputs unless you netted them out yourself: see execution cost.

And it does not tell you what to risk. There is no acceptable number here that this page could know, because it depends on constraints it cannot see. What it gives you is the consequence of a choice, across the range your own evidence supports.

The term is defined at risk of ruin, and the barrier that converts a decline into an ending is a drawdown limit. Where that barrier moves with your equity instead of staying put, the trailing drawdown calculator computes the distance to it, and if you are unsure which architecture governs you,work out which drawdown rule you have. What actually causes a prop-firm breach is examined here, and the same simulation pointed at an evaluation's rules — pass or breach, across your sample's range — is the prop evaluation survivor. The projection this constrains — what a growth rate compounds to if the account survives to collect it — is the forex compounding calculator. All the calculators are listed under calculators. One caution on the input that decides everything above: this takes your win rate as given, and a win rate selected from many tried variants is not a measurement of anything. Whether yours survives that test is what the Overfit Auditor bounds. Watching the same ruin surface move in real time is the Prop-Evaluee Risk Guardian. And when the inputs cannot support an answer, the panel above refuses and says why rather than rendering a zero — why an instrument refuses to answer is that design, written down.