Risk of Ruin: How It Is Calculated, Why Risk Per Trade Decides It, and the Kelly Criterion

Risk of ruin is the probability that a trading strategy loses enough to end the account before its edge can play out. It depends on three inputs: win rate, payoff ratio, and risk per trade. Of the three, risk per trade is the one you control completely, and it moves the result more than the other two combined.

Every profitable strategy has a losing streak in its future, and the question that decides your career is whether your account can survive it. Risk of ruin puts a number on that question. It is the one figure that tells you whether a strategy with a real edge will make money for you, or whether you will be out of capital before the edge shows up.

Most traders never calculate it. They pick a risk per trade because it feels comfortable, and discover the answer the hard way. This guide covers what risk of ruin is, how it is calculated, what the tables say for realistic win rates and payoffs, why the definition of "ruin" changes on a funded account with a prop firm, and where the Kelly criterion fits in as the sizing formula that sits behind it. It is the third spoke under the guide to forex risk management, which covers the 1% rule and position sizing this page puts numbers on.

What Is Risk of Ruin in Trading?

Risk of ruin is the probability that cumulative losses reduce an account to a point where trading cannot continue, before the strategy's positive expectancy has time to work. It is expressed as a percentage. A strategy with a 2% risk of ruin will, on average, end two accounts in every hundred it is traded on, however good its edge.

The concept comes from the gambler's ruin problem in probability theory, which asks how likely a gambler with finite money is to go broke against an opponent with unlimited money. The market is the opponent with unlimited money. Your account is finite. Even with the odds in your favour, a finite bankroll can be exhausted by a long enough run of bad outcomes, and the smaller each bet is relative to the bankroll, the less likely that becomes.

What makes risk of ruin useful, rather than academic, is that it combines the three numbers traders usually consider separately. Win rate on its own says nothing. Reward-to-risk on its own says nothing. Risk per trade on its own says nothing. Put together, they produce a single probability that answers the only question that matters: will I still be trading when the edge pays off?

How Is Risk of Ruin Calculated?

Risk of ruin is calculated from win rate, payoff ratio, and the risk units between your capital and your ruin point. For even payoffs the formula is RoR = ((1 − A) ÷ (1 + A))^N, where A is win rate minus loss rate and N is the losses it takes to reach ruin. Uneven payoffs need a table.

The key idea is the unit. If you risk 1% of the account per trade and your ruin point is a 12% loss, you are 12 units from ruin. Risk 2% and you are 6 units away. Risk 0.5% and you are 24 units away. The formula raises a fraction to the power of N, so doubling your units does not halve your risk of ruin; it squares the fraction, which reduces ruin risk far more than proportionally.

For an even-money strategy (1:1 payoff) with a 55% win rate, the edge A is 0.55 − 0.45 = 0.10, so the base fraction is 0.90 ÷ 1.10 = 0.818. At 12 units to ruin, RoR = 0.818^12 = 9.0%. At 24 units it is 0.818^24 = 0.8%. At 6 units it is 30%. Same strategy, same edge, and the risk of ruin moves from one in three to less than one in a hundred purely by changing position size.

When wins and losses are different sizes, as they are for almost every real strategy, the closed formula no longer applies and the probability has to be solved numerically under a stated set of assumptions. That is what the tables in the next section do, and the assumptions are listed alongside them. If you want a rule of thumb without the tables: a strategy with positive expectancy at 1% risk and a 12-unit loss budget has a low single-digit risk of ruin; at 2% risk it moves to double digits; at 4% it is a coin flip or worse.

What Does the Risk of Ruin Table Show?

The table below estimates the probability of reaching a 12% loss for a 1:2 payoff strategy at different win rates and risk per trade. At 1% risk and a 45% win rate the estimated risk of ruin is 1.7%. At 2% risk it is 13.2%. At 4% risk, 36.3%. The win rate barely moved the result; position size did.

The figures below are model-based estimates intended to illustrate how risk per trade can affect the probability of reaching a 12% loss threshold. Actual risk of ruin will vary depending on the trader's realised win rate, payoff distribution, trade independence, changing position sizes, transaction costs, and other market factors. For asymmetric strategies such as a 1:2 reward-to-risk profile, risk of ruin depends on the specific assumptions and calculation method used and should therefore be treated as an estimate rather than a universal exact probability.

The model behind them is a fixed-unit random walk. Every trade is independent, the win rate and payoff ratio hold constant, each loss costs one unit and each win pays two, position size does not change as the balance moves, and costs are excluded. Ruin is reaching a 12-unit loss by any path, over an unlimited number of trades. That ruin point is the maximum loss on a Pipcy Classic account, and a reasonable "stop and reassess" line for a personal account.

Three things to read from this.

The 35% column is a warning about thin edges. At 1:2 the break-even win rate is 33.3%, so a 35% strategy is profitable on paper. Its risk of ruin at 1% risk is still 55%, because the edge is too thin to outrun normal variance. A positive expectancy is necessary and nowhere near sufficient.

Moving down a column is the position-size effect. At a 45% win rate, going from 0.5% to 4% risk moves ruin from 0.03% to 36.3%, a factor of more than a thousand. Nothing about the strategy changed.

Moving across a row is the win-rate effect. At 1% risk, going from 40% to 55% win rate moves ruin from 9.6% to 0.05%. That is a large improvement, and it is still smaller than the position-size effect. Win rate is hard to improve and easy to overestimate; risk per trade is a setting.

For an even-money strategy the picture is the same. At a 55% win rate and 1:1 payoff, ruin at a 12% loss is 0.8% at 0.5% risk, 9.0% at 1%, 30.0% at 2%, and 54.8% at 4%. At a 50% win rate and 1:1 the expectancy is zero and the risk of ruin is 100% at any size, because a strategy with no edge eventually loses everything however slowly it is traded.

Why Does Risk Per Trade Decide Risk of Ruin?

Risk per trade dominates risk of ruin because ruin is reached through a sequence of losses, and the number of losses your account can absorb is set by how much each one costs. Halving risk per trade doubles the losses you can survive, and because ruin probability falls exponentially with that number, it can cut ruin risk tenfold.

Traders tend to reason about losing streaks and stop there. The probability of twelve consecutive losses at a 50% win rate is 0.5^12, roughly 0.02%, so twelve units of loss budget feels untouchable. It is not, because ruin does not require the losses to be consecutive. Six losses, two wins, eight losses, one win, five losses also reaches the limit. The probability of reaching a 12-unit loss by any path at a 50% win rate and 1:2 payoff is 0.3%, about twelve times the consecutive-streak figure. Streak probabilities understate the risk; ruin probabilities measure it.

Streaks still matter as a sanity check, because they arrive far more often than intuition suggests. Over 200 trades at a 50% win rate, the probability of at least one run of six straight losses is about 80%. Eight straight: about 32%. Ten straight: about 9%. At a 40% win rate, the probability of at least one eight-loss streak in 200 trades is 75%, and a twelve-loss streak is 15%. A trader risking 2% per trade at a 40% win rate should expect a 16% drawdown from a normal streak, and has a meaningful chance of a 24% one. These are features of a working strategy, and the account has to be sized to survive them as routine.

That is the argument for the 1% rule, and for cutting size during losing streaks, both covered in the guide to reducing drawdown. Both follow from the fact that ruin probability is exponential in the number of units you can lose, which makes them arithmetic rather than temperament.

Where Is the Ruin Point, If Not Zero?

Ruin rarely means a balance of zero. For a personal account it is the loss at which you stop trading, often 30% to 50%. For a funded account it is the firm's maximum loss limit, commonly 8% to 12%. Calculate risk of ruin against the real ruin point; on a funded account it is far closer than most traders assume.

The textbook definition of ruin is total loss, and against that definition risk of ruin looks reassuring. A 45% win rate at 1:2 with 5% risk per trade has only a 0.1% probability of ever reaching zero. But no trader reaches zero; they stop long before. A 50% loss needs a 100% gain to recover, as the guide to drawdown sets out, and most traders quit or change strategy well before that. Against a 50% ruin point, the same strategy at 5% risk has a 3.4% risk of ruin; at 10% risk it is 18.5%.

On a funded account the ruin point is written into the contract. The account ends when losses reach the maximum drawdown limit, whether that is 12% static or 8% trailing. That is the number to use as N, and it changes the picture completely. A strategy that is "safe" at 3% risk against a 50% personal ruin point (roughly 0.4% risk of ruin at 45% win, 1:2) has a 26% risk of ruin against a 12% funded-account limit at the same size. The strategy did not get worse. The ruin point moved closer.

If the limit trails, the effective ruin point moves closer still as you profit, because a trailing limit rises with your equity peak and locks in. A static limit stays where it started. For risk-of-ruin purposes a static limit is a fixed N; a trailing limit is an N that shrinks whenever you have a good week.

What Is the Kelly Criterion, and How Does It Relate to Risk of Ruin?

The Kelly criterion gives the fraction of capital to risk per trade that maximises long-run growth: f* = W − (1 − W) ÷ R, where W is win rate and R is payoff ratio. A 45% win rate at 1:2 gives 17.5%. Full Kelly produces severe drawdowns; most traders use a quarter to a half of it.

Kelly answers the opposite question to risk of ruin. Ruin asks "how likely am I to lose the account at this size?" Kelly asks "what size grows the account fastest?" The two meet in the middle, because the Kelly fraction is also the point beyond which larger size reduces long-run growth and increases ruin risk at the same time. Over-betting Kelly is the one setting where you take more risk to earn less.

Read against the ruin table, these fractions are alarming. Full Kelly at 45% and 1:2 is 17.5% risk per trade, which against a 12% loss limit is ruin on the first losing trade. Even quarter Kelly at 4.4% has roughly a 40% risk of ruin against that limit. The reason is that Kelly assumes you know your true win rate and payoff exactly, that they will not change, and that you can tolerate drawdowns of 50% or more on the way to maximum growth. None of those hold for a discretionary trader, and the third is disqualifying on any funded account.

The practical use of Kelly is as a ceiling. Calculate f* from a journal of at least 100 trades, then trade a small fraction of it, typically a fifth to a quarter, and treat anything above half Kelly as over-betting. For most retail strategies that lands between 0.5% and 2% per trade, which is where the risk-of-ruin table says it should. Ralph Vince's "optimal f" is a related formula that uses the actual distribution of past trade results rather than a single win rate and payoff; it produces similar fractions and carries the same warning.

How Do Expectancy and Sample Size Affect the Calculation?

Risk of ruin is only as reliable as the win rate and payoff you feed into it, and those need at least 100 closed trades to estimate well. Expectancy, (win rate × average win) − (loss rate × average loss), must be positive before risk of ruin means anything. A negative-expectancy strategy has 100% risk of ruin at any size.

The calculation is only as reliable as the win rate you feed it. A win rate measured over 20 trades has a margin of error of roughly plus or minus 20 percentage points, which is the difference between a 1.7% risk of ruin and a 55% one on the table above. Over 100 trades the margin narrows to about 10 points; over 400 to about 5. Until the sample is large, the honest approach is to size for a win rate a few points below what the journal shows, and to recalculate as the sample grows.

Expectancy is the gate. If (W × average win) − (L × average loss) is zero or negative, no position size makes the strategy safe; smaller size only makes the ruin slower. The formula and the break-even table that connects payoff ratio to the win rate you need are in the guide to forex risk management, and a dedicated guide to the risk-reward ratio covers the planned-versus-realised gap that quietly turns a positive expectancy into a negative one.

Two figures from the journal matter more than any others for this calculation: the realised average loss in R (usually slightly more than 1R because of slippage) and the realised win rate. Plug those in, not the planned ones.

How Does Risk of Ruin Work on a Funded Account?

On a funded account, risk of ruin is the probability of hitting the firm's maximum loss; the limit is the ruin point. On Pipcy Classic, with an 18% target and a 12% static maximum, a trader at 1% risk, 1:2 payoff and 45% win rate has an estimated 1.7% ruin risk. At 2% risk, 13.2%; at 3%, 25.9%.

The evaluation makes the ruin point explicit, which is the one advantage it has over a personal account. On Pipcy Classic the ruin point is a 12% loss, static, with no daily limit and no trailing component, so N is fixed for the life of the account. On the pip-based Pips Mastery Challenge the ruin point is 250 pips, so a trader risking 25 pips per trade has 10 units, and one risking 50 pips has 5.

The risk-of-ruin column repeats the ruin-by-any-path figures from the main table, which treat the account as open-ended; an evaluation that ends at the target can only do slightly better than these, so they are a conservative estimate. The trades column is the average number of trades in simulated attempts that reached the target, under the same assumptions.

The last column is where the temptation lives. Doubling risk from 1% to 2% more than halves the number of trades to target, and it multiplies the probability of failing the evaluation by nearly eight. Tripling it to 3% cuts the journey to under a quarter and raises the failure probability fifteen-fold. Because the evaluation fee is small and the funded account is valuable, the asymmetry is entirely against speed. The trader at 1% risk who takes 50 trades to pass almost always passes. The trader at 3% who tries to pass in a dozen loses the account one time in four before getting there, which is a large part of why most traders fail prop challenges.

Once funded, the calculation continues. The same 12% limit applies to the funded account, and a payout does not reset it, so the risk of ruin you accepted in the evaluation is the one you carry into every subsequent month. Choosing it deliberately, from the table rather than from comfort, is the difference between a funded account that lasts and one that ends on an ordinary losing streak.

What Are the Most Common Risk of Ruin Mistakes?

The most common risk of ruin mistakes are calculating it against total loss instead of the real ruin point, trusting a win rate from too few trades, using planned rather than realised payoffs, treating a losing-streak probability as a ruin probability, and sizing at or near full Kelly. Each produces a number far more comforting than the truth.

Using zero as the ruin point. Against total loss almost any sensible size looks safe. Against a 12% limit or a 50% personal threshold, the same size can be a coin flip. Always calculate to the point where you would actually stop.

Trusting a small sample. A 55% win rate over 20 trades may be a 45% strategy having a good month. Size for the lower figure until the journal proves otherwise.

Feeding in planned numbers. Spread, slippage and early exits make the realised average loss larger and the realised average win smaller than planned. Ruin probability rises with both.

Confusing streaks with ruin. Twelve straight losses at 50% is 0.02%; reaching a twelve-unit loss by any path is 0.3%. Ruin does not wait for a tidy sequence.

Sizing near Kelly. Full Kelly assumes perfect knowledge of your edge and tolerance for 50%-plus drawdowns. Quarter Kelly is the aggressive end for a discretionary trader; on a funded account, less.

Frequently Asked Questions

What is risk of ruin in trading?

Risk of ruin is the probability that a trading strategy's losses reduce the account to a point where trading stops, before its positive expectancy has time to produce a profit. It is calculated from win rate, payoff ratio and risk per trade, and it is expressed as a percentage. A 2% risk of ruin means the strategy would end the account roughly two times in every hundred it is traded.

How do you calculate risk of ruin?

For even payoffs, use RoR = ((1 − A) ÷ (1 + A))^N, where A is win rate minus loss rate and N is the number of losing trades between your capital and your ruin point. A 55% win rate with 12 units to ruin gives (0.90 ÷ 1.10)^12, about 9%. For uneven payoffs such as 1:2 there is no such closed formula, so the probability is solved numerically under a stated model, which is what a risk of ruin table or calculator does. Those results are estimates that depend on the assumptions used.

What is an acceptable risk of ruin?

Most professional risk frameworks treat anything above 1% to 2% as too high and aim for well under 1%. On the table above that means risking about 1% per trade or less for a strategy with a 45% win rate and 1:2 payoff against a 12% loss limit. On a funded account, where the ruin point is the firm's maximum loss, aim for the lower end, since a breach ends the account rather than a month.

What is the Kelly criterion in trading?

The Kelly criterion is a formula for the fraction of capital to risk per trade that maximises long-run growth: f* = W − (1 − W) ÷ R, where W is win rate and R is the payoff ratio. A 45% win rate at 1:2 gives 17.5%. Because full Kelly assumes an exactly known edge and tolerates very deep drawdowns, traders use it as a ceiling and size at a quarter to a half of the result, or less on a funded account.

Should you use full Kelly or half Kelly?

Neither, for most traders. Full Kelly maximises growth in theory and produces drawdowns of 50% or more in practice, with a high risk of ruin against any realistic loss limit. Half Kelly gives about three quarters of the growth with far shallower drawdowns. Quarter Kelly or less is the sensible range for discretionary traders whose win rate is an estimate, and on a funded account the firm's maximum loss usually forces sizing well below that.

What is the difference between risk of ruin and gambler's ruin?

Gambler's ruin is the probability problem from which risk of ruin comes: the chance that a gambler with finite money goes broke against an opponent with unlimited money, even with favourable odds. Risk of ruin is the same idea applied to trading, with win rate, payoff ratio and risk per trade as the inputs and the account's loss limit as the ruin point. The maths is shared; the trading version uses realistic, uneven payoffs.

How does risk of ruin apply to a funded account?

On a funded account the ruin point is the firm's maximum loss limit rather than a zero balance, so risk of ruin is the probability of breaching that limit. With a 12% static limit, 1% risk per trade gives 12 units to ruin and, at a 45% win rate and 1:2 payoff, an estimated risk of ruin near 1.7%. At 2% risk the estimate rises to 13.2% and at 3% to 25.9%, which is why sizing up to pass faster is the most common way evaluations are lost.

Risk of ruin is the number that connects your edge to your survival. It takes the win rate and payoff ratio your journal shows, the risk per trade you choose, and the loss at which your account ends, and returns a single probability. The tables above say the same thing in every row: the strategy matters less than the size. A modest edge at 1% risk survives almost everything; a good edge at 4% risk is a coin flip against a 12% limit.

Calculate it against the real ruin point, feed it realised rather than planned figures, treat Kelly as a ceiling rather than a target, and recalculate as the sample grows. On a funded account, where the ruin point is written into the rules and does not move, the calculation is simpler than anywhere else, and the trader who runs it once has an advantage over the majority who never do.

Written by Vladimir Rybakov, Head of PIPCY Academy, CFTe-certified with 19 years of trading experience.

Fact-checked by Snir Ahiel, former co-founder of The5ers and risk management specialist at Pipcy, with 15+ years trading Forex, Stocks, and Options.

Risk disclosure: Trading involves substantial risk. Pipcy provides simulated trading evaluations for educational and assessment purposes. Simulated performance does not represent real trading results, and becoming a funded trader is not guaranteed. Nothing here is financial or investment advice.