Risk of Ruin
Risk of ruin is the estimated probability that a sequence of losses drives trading capital below a defined survival threshold before any edge can compound, given the win rate, the payoff and the fraction of capital risked per trade.
Quick answer: Risk of ruin is the estimated probability that a sequence of losses drives trading capital below a defined survival threshold before any edge can compound, given the win rate, the payoff and the fraction of capital risked per trade.
In simple words
Risk of ruin is the chance that a bad run wipes you out before your edge has time to work. It depends on three things: how reliable your edge is, how much you risk on each trade, and how much capital you have to absorb losses. Bet a large fraction of your account and even a real edge can be destroyed by an ordinary losing streak; bet small and survival becomes almost certain. It is a model estimate of the probability of blowing up, not a precise forecast.
Purpose
Risk of ruin exists to make the survival consequence of position sizing explicit, showing that the same edge can be near-certain to survive or near-certain to fail depending only on how much is risked per trade.
Visual explanation
Risk of Ruin
How the probability of ruin falls steeply as risk per trade shrinks and as the number of capital units grows, for a fixed edge.
Professional explanation
What ruin means and why it is a threshold, not zero
Ruin does not necessarily mean reaching a zero balance. In practice it is a threshold below which a trader can no longer continue: a margin level that forces closure, a maximum drawdown mandate that stops the desk, or simply the psychological point at which the trader quits. Risk of ruin is the probability that the equity path touches that threshold at any point, given the statistical properties of the trades. Framing it as a threshold matters because a trader is finished long before the account literally reaches nothing, and because the relevant question is survival of the ability to keep trading, not the arithmetic of the last rupee.
A simplified model for fixed-fractional betting
For a simplified model where each bet risks a fixed fraction of capital and wins and losses are of comparable size, the risk of ruin can be approximated as ((1 − Edge) ÷ (1 + Edge)) raised to the power of the number of capital units at risk. Edge here is the fractional advantage per bet, roughly (win probability − loss probability) for even-money outcomes, and the number of units is capital divided by the amount risked per trade. The formula shows two things vividly: with no edge (Edge = 0) the base is 1 and ruin is certain over enough trades, while any positive edge makes the base less than 1 so that ruin probability shrinks geometrically as the number of units grows. It is a teaching model, not an exact figure for real, variable payoffs.
The three levers: edge, bet size and units of capital
Risk of ruin is governed by three inputs and their interaction. A larger edge lowers ruin, but its effect is bounded, real edges are small and uncertain. Risk per trade is the most powerful and most controllable lever: halving the fraction risked roughly doubles the number of capital units, which drives the ruin probability down by a power, not a proportion. The number of units, capital divided by risk per trade, is therefore the exponent that does the heavy lifting. This is the quantitative core of the whole discipline: survival is bought far more cheaply by cutting bet size than by improving the edge, because bet size enters as an exponent.
Why full Kelly is near the edge of ruin
The Kelly criterion maximises the long-run growth rate of capital, but it does so by betting an aggressive fraction that produces large drawdowns and a meaningful path-dependent risk of ruin if the edge is even slightly overestimated. Because a real edge is never known exactly and is unstable across regimes, betting full Kelly on an estimated edge courts ruin whenever the estimate is too high. This is why practitioners bet a fraction of Kelly, often half or less: the growth rate falls only modestly while the drawdowns and the risk of ruin fall sharply. Risk of ruin is the lens that exposes why fractional Kelly is prudent rather than timid.
Model assumptions and how they break
Every risk-of-ruin formula rests on assumptions that real trading violates. It typically assumes independent trades with a stable win rate and fixed payoff, whereas real returns are serially correlated, edges drift, and payoffs are asymmetric and fat-tailed. It usually ignores that a single gap or tail event can breach the threshold in one move rather than through a gradual streak, and it assumes the edge is known when it is only estimated. The output is therefore a model-based estimate whose main value is comparative and directional, showing how ruin responds to bet size, rather than a literal probability to be trusted to two decimal places.
Formula
Risk of ruin ≈ ((1 − Edge) ÷ (1 + Edge))^U, where U = Capital ÷ Risk per trade
Edge = the fractional statistical advantage per trade, approximately (win probability − loss probability) for even-money bets, so a 55 percent win rate on even bets gives Edge ≈ 0.10; Risk per trade = the amount of capital risked on each trade; Capital = total trading capital; U = the number of units of capital at risk, i.e. Capital ÷ Risk per trade. With Edge = 0 the base is 1 and ruin is certain; any positive edge makes the base below 1 so ruin falls as U rises. This is a simplified model estimate assuming independent, comparable-size bets, not an exact figure for real payoffs.
Small bet size vs Large bet size, same edge
| Aspect | Risk 1 percent per trade | Risk 10 percent per trade |
|---|---|---|
| Units of capital (U) | 100 | 10 |
| Effect on ruin | Ruin probability driven very low | Ruin probability materially high |
| Losing-streak tolerance | Absorbs long streaks | A modest streak approaches the threshold |
| Growth per winning trade | Slower | Faster |
| Survival of the edge | Near-certain if edge is real | Fragile even if edge is real |
Practical example
Illustrative example (Indian market)
A trader with ₹5,00,000 has a genuine edge on a Nifty setup: a 55 percent win rate at roughly even payoff, so Edge ≈ 0.55 − 0.45 = 0.10, giving a base of (1 − 0.10) ÷ (1 + 0.10) = 0.90 ÷ 1.10 ≈ 0.818. If they risk 1 percent, ₹5,000, per trade, the number of units is 5,00,000 ÷ 5,000 = 100, and the risk of ruin is about 0.818 raised to the 100th power, a vanishingly small number: survival is near-certain. If instead they risk 10 percent, ₹50,000, per trade, U falls to just 10, and 0.818 to the 10th power is roughly 0.13, a 13 percent chance of ruin despite the identical edge. The edge did not change; only the bet size did, and it moved survival from a near-certainty to a one-in-eight chance of blowing up.
In Indian F&O the margin available on ₹5,00,000 can tempt a trader to risk far more than 1 to 2 percent per trade, collapsing the number of capital units and pushing risk of ruin up sharply. The broker's permitted leverage sets no ceiling on ruin; only the self-imposed risk-per-trade fraction does.
Advantages
- Makes the survival cost of bet size explicit, exposing why sizing dominates edge
- Shows quantitatively that risk per trade enters as an exponent, so cutting it helps disproportionately
- Explains why fractional Kelly is prudent when the edge is uncertain
- Useful comparatively to rank the survival impact of different sizing rules
- Reinforces that survival, not maximising growth, is the first objective
Limitations
- Blind spot: it is a model estimate assuming independent, stable-edge, comparable-size trades, none of which strictly hold in real markets
- It ignores fat tails and gaps that can breach the threshold in a single move, not a gradual streak
- It assumes the edge is known, whereas a real edge is only estimated and drifts with regime
- Serial correlation and clustered losses make real ruin higher than the independent-trade model implies
- The exact probability should not be trusted to precise decimals; its value is directional
Why it matters in practice
- It reframes position sizing as a survival decision rather than a return decision
- It shows why an overestimated edge bet at full Kelly is a fast route to ruin
Common mistakes
- Sizing off available margin instead of a risk-per-trade fraction that keeps ruin low
- Assuming a positive edge alone guarantees survival regardless of bet size
- Betting full Kelly on an estimated edge and ignoring the ruin it courts
- Trusting the model's exact probability while ignoring fat tails and gaps
- Treating trades as independent when losing streaks cluster in trending markets
- Believing a small backtested drawdown implies a small risk of ruin going forward
Professional usage
Risk desks use risk of ruin as the conceptual bridge between an edge and a sizing rule: they choose the fraction risked per trade so that, under conservative assumptions about the edge and its uncertainty, the modelled probability of hitting the drawdown threshold is negligible. They deliberately bet a fraction of Kelly because they distrust their own edge estimates, and they stress the model with fatter tails and correlated losses to see how quickly ruin rises. Above all they treat the formula as a directional guide to sizing, not a precise forecast, because survival must hold even when the model's assumptions fail.
Key takeaways
- Risk of ruin is the modelled probability that losses breach a survival threshold before the edge compounds
- It depends on edge, risk per trade and the number of capital units, with bet size the dominant lever
- Risk per trade enters as an exponent, so cutting it lowers ruin disproportionately
- Full Kelly courts ruin when the edge is overestimated, which is why fractional Kelly is prudent
- It is a model estimate, not a precise forecast, and real ruin is higher because of fat tails and clustering
Frequently asked questions
What is risk of ruin?
How is risk of ruin calculated?
What does ruin actually mean?
What is the biggest lever on risk of ruin?
Can I have an edge and still go broke?
Why is full Kelly risky for risk of ruin?
What is the blind spot of risk-of-ruin formulas?
How many capital units should I risk down to?
Is risk of ruin a precise probability?
How does risk of ruin relate to the Kelly criterion?
Does risk of ruin account for a single catastrophic move?
How does leverage affect risk of ruin?
Why do losing streaks matter more than the loss rate?
Can I reduce risk of ruin without a better strategy?
Voice search & related questions
Natural-language questions people ask about Risk of Ruin.
What is risk of ruin?
Can a good trader still blow up?
How do I lower my risk of ruin?
Why is betting your whole edge dangerous?
Is the risk of ruin formula exact?
Does leverage raise my risk of ruin?
Sources & references
Last reviewed 12 July 2026. Educational content only — not investment advice. Markets and rules change; verify current conventions with SEBI, NSE/BSE and your broker.