Risk of ruin
Risk of ruin is the probability that losses drive an account down to a set loss threshold, such as 50% below its starting balance, at any point.
It exists only inside a model: the win rate, the payoff per trade, the sizing method and the ruin level all feed it, and changing any one changes the result. Every figure on this page defines ruin as equity falling to 50% of the starting balance.
This page covers the mechanic; it is not trading advice.
Ruin is a threshold, not a zero balance
In the arithmetic, risking a fixed percentage of current equity never reaches zero. At 5% per trade, each loss takes 5% of a smaller balance, so 14 consecutive losses leave 0.95¹⁴ = 48.77% of the account. Ruin is therefore a chosen Drawdown level. This page uses a 50% fall, which needs a 100% gain to recover.
Positive expectancy is often taken as protection against ruin, because, with a fixed cash risk per trade, ruin is certain only when the edge is zero or negative. That reading is wrong. A strategy with a 55% Win rate and a 1:1 payoff earns an average of +0.10R per trade, where R is the amount risked, yet risking a fixed $500, 5% of a $10,000 starting balance, on every trade it reaches the 50% threshold with a probability of 13.44%, roughly 1 in 7.
Risk of ruin at 1%, 2% and 5% per trade, worked
The model assumes a fixed cash risk per trade, set as a percentage of the starting balance; losses of exactly 1 risk unit and gains of a fixed size set by the payoff ratio; the same probability p of a profitable trade every time, with trades independent; no costs; no limit on the number of trades; and ruin at 50% of the starting balance.
Take a $10,000 account trading EUR/USD long at 1.1253, with a 25-pip Stop-loss at 1.1228, a 25-pip target at 1.1278 and a 55% win rate.
Risk of ruin, 1:1 payoff
Risk of ruin = (q ÷ p)ᴺ, where N = (starting balance − ruin level) ÷ risk per trade- Win rate (p)
- 0.55
- Loss rate (q = 1 − p)
- 0.45
- Risk per trade (5% of $10,000)
- $500
- Risk units to ruin (N = $5,000 ÷ $500)
- 10
Risk of ruin (0.818182¹⁰)
13.44%
At $10 per pip per standard lot, the 25-pip stop costs $250 per lot, so the Position size is the cash risk divided by $250. The $5,000 above the ruin level then holds 50, 25 or 10 risk units.
Doubling the risk from 1% to 2% halves N, which under this model turns the probability into its own square root: √0.0000439 = 0.0066, about 151 times higher.
A higher expectancy can carry a higher risk of ruin
A 40% win rate at a 1:2 Risk-reward ratio, with a 50-pip target at 1.1303, has an Expectancy of 0.40 × 2 − 0.60 = +0.20R per trade, twice that of the 55% strategy. Its risk of ruin, zᴺ where z ≈ 0.8229 is the root of 0.4z³ − z + 0.6 = 0, is higher at every level: 0.0058% at 1%, 0.76% at 2% and 14.23% at 5%. The results swing wider (variance 2.16R² against 0.99R²), and the swings outweigh the higher average.
Fixed-percentage sizing and the model's limits
Re-sizing every trade as a percentage of current equity shrinks each loss as the account falls, so crossing 50% from the start takes 69, 35 or 14 consecutive losses instead of 50, 25 or 10. For the 55% strategy, risk of ruin becomes about 1 in 600,000 at 1%, 0.18% at 2% and 11.59% at 5%: lower, not zero. No closed form exists; these were computed numerically.
Costs, slippage and clustered losses push the real figure up. Costs thin the edge, and a thinner edge moves the figure sharply: at 2% risk, a 52% win rate instead of 55% raises the risk of ruin from 0.66% to 13.52%. Slippage at the stop makes some losses larger than 1 risk unit, and clustered losses make long runs more likely.
Related terms
Related articles
Common questions
Does a positive expectancy mean the risk of ruin is zero?
No. A strategy with a 55% win rate and a 1:1 payoff, trading EUR/USD long from 1.1253 with a 25-pip stop and target, averages a profit of 0.10 times the amount risked per trade. Risking a fixed $500 (5%) of a $10,000 account per trade, it still falls to $5,000 with a probability of (0.45 ÷ 0.55)¹⁰ = 13.44%, assuming independent trades and no costs.
How much does a lower win rate change the risk of ruin?
On a $10,000 account risking a fixed $200 (2%) per trade on a 1:1 payoff, such as a 25-pip stop and target on EUR/USD at 1.1253, the $5,000 ruin level is 25 risk units away. At a 55% win rate the risk of ruin is (0.45 ÷ 0.55)²⁵ = 0.66%. Drop the win rate to 52% and it becomes (0.48 ÷ 0.52)²⁵ = 13.52%, about 20 times higher.
Is risk of ruin the same as maximum drawdown?
No. Maximum drawdown is the deepest fall from a peak an account has already recorded or, in prop trading, a firm's loss limit. Risk of ruin looks forward: the probability of falling to a chosen level. At a fixed 1% risk on a $10,000 account, a 10% firm limit is 10 risk units below the start; the 50% threshold is 50.