What risk of ruin means
Risk of ruin is the probability that your account falls to a level you define as unacceptable, such as losing half of it, before your edge has time to work. It depends on three things: how much you risk per trade, how often you win and how much you win versus lose, and how long you trade.
The key insight is that ruin is driven mostly by position size. A strategy with a real edge can still blow up if each trade risks too much, while the same strategy at a smaller size almost never does. That is why the calculator also searches for the largest risk per trade that keeps ruin at 1% or lower.
How this calculator works
It simulates thousands of accounts trading your strategy and counts how many reach your ruin level within the number of trades you choose. The curve chart repeats this for different risk sizes so you can see where danger starts. The "long-run estimate" is a closed-form upper bound for fixed-size trades with no time limit, based on the same calculation actuaries use for insurance ruin.
Choosing a ruin level
Ruin does not have to mean zero. Many traders treat a 30% to 50% loss as ruin, because recovering requires gains of 43% to 100% (see the drawdown recovery calculator) and because most people stop trading well before that. Use the level that would make you quit or that your broker, prop firm or risk manager would not allow.
The math
For fixed-size trades that win W times the risk with probability p and lose L times the risk otherwise, the probability of ever losing B units of risk is bounded by:
where θ solves p × exp(−θW) + (1 − p) × exp(θL) = 1B is the loss you call ruin divided by the risk per trade. When expectancy is zero or negative, ruin is certain given enough trades.
For an even-payoff strategy (W = L = 1) this reduces to the classic gambler's ruin result, ((1 − p) ÷ p) raised to the power B. The Monte Carlo result on this page also accounts for compounding and a finite number of trades.
Worked examples
Three position sizes for the same strategy
A strategy that wins 45% of trades at 1.5R, so expectancy is positive at +0.125R. Ruin is a 50% loss within 500 trades. Only the risk per trade changes:
Why a good strategy can still blow up
Edge alone does not protect you. At 10% risk per trade, this same winning strategy has more than a one in two chance of losing half the account. The edge is unchanged. Only the size is too big.
Frequently asked questions
What is a good risk of ruin?
Professionals aim for well under 1%. Retail traders often run double-digit ruin risk without knowing. If your chance of losing half your account is above a few percent, reduce your risk per trade.
Why does risk of ruin depend so much on position size?
Ruin needs a long enough streak of losses to reach your threshold. Bigger bets need fewer losses to get there. Halving your risk per trade roughly doubles the streak you can survive and cuts ruin dramatically.
Can I eliminate risk of ruin?
Only by having a positive edge and keeping bets small. With negative expectancy, every size eventually leads to ruin. A smaller size just delays it.
What is the difference between the simulation and the long-run estimate?
The simulation shows your chance within a set number of trades and with compounding. The long-run estimate assumes fixed-size trades and an unlimited horizon, so it is usually higher for strategies with a positive edge.
Are my inputs reliable?
Your win rate and payoff are estimates from a limited sample. Small errors matter a lot at large position sizes. Run the calculator with a slightly lower win rate to check how fragile your plan is.
How does this relate to the Kelly criterion?
Betting more than about twice the Kelly size produces negative growth even with an edge. Staying at a half or a quarter of Kelly is the usual way to keep ruin risk low. See the Kelly criterion calculator.
Lower ruin risk starts with a real edge.
Position size controls how fast you can lose. A tested edge controls whether you should be trading at all.
Last updated October 2026. For education only. Not financial advice. Results depend on your inputs and cannot predict future performance.