Understanding Risk of Ruin

Risk of ruin is the probability of losing a chosen share of the account - often half - before an edge has had the chance to accumulate. It combines the win rate, the payoff ratio and the fraction of the account risked per trade.

It answers a question that expectancy cannot. Expectancy describes what an average trade is worth over a long run; ruin is about the sequence, and a long enough losing streak ends an account regardless of what the average would eventually have been.

Doubling the risk per trade does considerably more than double the probability of ruin. Losses compound against a shrinking base, so each successive loss is a larger proportional step toward the threshold.

That asymmetry is why professional risk limits look so conservative to newer traders. It is not caution as a virtue - it is the arithmetic of remaining in the game long enough for a small edge to work.

At a forty percent win rate, a run of six consecutive losses is unremarkable and will happen repeatedly across a few hundred trades. At a fifty percent win rate, runs of five and six still occur regularly.

The practical question is not whether such a streak will happen but whether the account survives it and whether you will still be following the strategy afterwards. Both are position-size questions.

The model is exquisitely sensitive to the win rate and payoff you feed it, and both are estimates from a finite sample. A win rate estimated from thirty trades, or from memory, produces a number with no information in it.

The useful way to use it is comparatively. Rather than asking "what is my risk of ruin", ask "how much does it change if I halve my size" - a question the model answers robustly even when the inputs are rough.

The threshold is whatever you set. Fifty percent is common because a fifty percent loss requires a hundred percent gain to recover, which is a practical point of no return for most accounts.

For a funded or prop account, the relevant threshold is the programme maximum drawdown, which is frequently far tighter than fifty percent and reached far sooner than participants expect.

It assumes trades are independent, that the win rate and payoff are stable, and that risk stays a constant fraction. Real trading violates all three - strategies degrade, market regimes change, and correlated positions cluster losses.

Every one of those violations makes real ruin more likely than the model suggests, not less. Treat the output as an optimistic floor.