Keel's free risk of ruin calculator gives the chance that your account falls to a loss you can't come back from, such as half its starting value. Enter your win rate, reward to risk and the share of the account you risk per trade. It returns the exact odds over the number of trades you choose and a ceiling for the long run.
The share of your current balance you lose when a trade hits its stop.
From your starting balance.
Up to 5,000.
The chance of losing 50% of the account at some point in the next 500 trades. Over any number of trades it is at most 0.22%.
What a typical trade does to the balance once compounding is counted.
The risk per trade that grows the account fastest. Risking about twice this or more shrinks it.
| Risk per trade | Ruin in 500 trades |
|---|---|
| 1% | under 0.01% |
| 2% | 0.20% |
| 5% | 12.0% |
| 10% | 46.8% |
| 20% | 88.7% |
Every number on this page comes from one simple model of trading. Read it before you trust the result, because real trading breaks some of it.
A win multiplies the balance by 1 + risk × reward to risk. A loss multiplies it by 1 − risk. The balance after a run of trades depends only on how many wins and losses it held, but ruin depends on the order, because one bad stretch early can cross the line even if the account would have recovered later.
The calculator follows every possible sequence of wins and losses one trade at a time. At each step it moves the probability of each surviving path forward, and any path that touches the ruin level is counted and stopped. Over the number of trades you enter, the result is exact for the model above, with no random simulation and no sampling error.
win: balance × (1 + risk × R)
loss: balance × (1 − risk)
ruin: balance ≤ start × (1 − ruin level)For the long run it shows a ceiling from ruin theory (Lundberg's inequality). Find the number θ above zero where p × (1 + risk × R)^−θ + (1 − p) × (1 − risk)^−θ = 1. The chance of ever reaching the ruin level is at most (1 − ruin level)^θ. When the account's typical path drifts down, no such θ exists and ruin is certain given enough trades.
The same strategy can be close to safe or close to doomed depending only on how much it risks. Here is one strategy, 40% winners at 2 to 1, at five risk levels:
| Risk per trade | Ruin in 1,000 trades | Losses in a row to ruin |
|---|---|---|
| 1% | under 0.01% | 69 |
| 2% | 0.21% | 35 |
| 5% | 12.1% | 14 |
| 10% | 47.5% | 7 |
| 20% | 91.8% | 4 |
Halving the risk per trade roughly doubles the losing streak you can survive, and the odds of a streak that long fall away quickly. That is why small fixed-percentage risk is the standard advice. It keeps you trading long enough for the edge to show.
Expectancy in R tells you what an average trade makes. It does not tell you what happens to your balance, because losses and gains compound. At 50% risk, a win at 2 to 1 doubles the balance and a loss halves it, so with 45% winners the typical account shrinks even though the average trade makes 0.35R. The median growth row shows this directly. The Kelly fraction is the risk that grows the account fastest, and risking around twice it or more turns growth into decline.
The answer is only as good as the win rate and reward to risk you enter. A handful of trades gives a rough guess at both; use the win rate calculator to see how far a short record can be off. Measured over hundreds of trades, after fees and funding, they give a far more honest number. To see how far real results can stray from the average, reshuffle your trade list with the Monte Carlo simulation.
Keel is a strategy builder for Hyperliquid. Describe an idea, backtest it with fees, slippage and funding, and run it live on Hyperliquid’s crypto perps. Start free, and connect a wallet when you’re ready to trade.
Risk of ruin is the probability that your account falls to a level you can't recover from before your edge has time to work. If you risk a percentage of your balance on each trade, the account never quite reaches zero, so ruin is set as a loss from the starting balance, often 50%. It depends on your win rate, your reward to risk and how much you risk per trade.
With percentage risk, every win multiplies the balance by 1 + risk × reward to risk and every loss by 1 − risk. This calculator follows every possible sequence of wins and losses over the number of trades you choose and adds up the probability of each path that touches the ruin level. That gives the exact answer for the model, with no simulation noise. For the long run it shows an upper bound from ruin theory.
As close to zero as you can get while still growing. A common target is under 1% over the number of trades you plan to take. If the long-run figure says certain, the account is set to shrink at that risk per trade and no amount of patience fixes it.
Little enough that a normal losing streak can't reach your ruin level. With a positive edge, risk of ruin falls very fast as risk per trade falls, which is why 1% to 2% per trade is a common rule. The Kelly fraction is the most you should ever consider: risking about twice the Kelly fraction or more makes the account shrink even with a winning system.
Yes. A strategy that wins 45% of the time at 2 to 1 makes 0.35R a trade on average. Risk 50% a trade on it and a win doubles the account while a loss halves it, so with more losses than wins the typical account goes down. Positive expectancy only turns into growth when the risk per trade is small enough.
Losing streaks are what ruin is made of, and the calculator already counts every streak the win rate allows. It shows how many losses in a row take you to the ruin level from the start. It assumes trades are independent; if your losses bunch together, for example in one kind of market, your real risk is higher than the number here.
Yes. Fees and funding make every win a little smaller and every loss a little bigger, which lowers your edge and raises your risk of ruin. Enter a reward to risk and win rate measured after costs, for example from a backtest that charges fees and funding, to include them.