Risk of Ruin Calculator: How Much Risk Ends Accounts

This free risk of ruin calculator estimates the chance that a normal run of losing trades drags your account down to a fixed drawdown limit. Enter your winning-trade percentage, your payoff ratio, and your risk per trade. The tool returns the model probability of hitting your ruin point, plus a comparison table across five risk levels. That table is the real lesson. Small changes in risk per trade move the result by orders of magnitude. Everything runs in your browser, and nothing is stored.

How to use the risk of ruin calculator

The tool needs four numbers. Each one should come from your own trading records, not from hope. Here is the workflow.

1. Enter your winning-trade percentage. Pull it from at least 100 closed trades. If you have fewer, treat every output as a rough sketch.
2. Enter your payoff ratio. Divide your average winning trade by your average losing trade. A system that wins 1.5R and loses 1R has a payoff ratio of 1.5.
3. Enter your risk per trade as a percent of the account. The default is 1. This is the loss a single stopped-out trade costs you.
4. Enter your ruin point. This is the drawdown at which you would stop trading, or at which the account is effectively done. The default is 50. Funded traders should enter their program's max drawdown instead.
5. Press Calculate. Read the headline probability, then study the five-row table below it. The table shows the same system at 0.5, 1, 2, 3 and 5 percent risk.

The output updates as you type. Change one input at a time and watch how the probability responds. That habit teaches more about risk than any lecture.

Worked example: 45% winners, 1.5 payoff, 2% risk

Walk through the default-style case by hand. Your records show 45 percent winners. Your average win is 1.5 times your average loss. You risk 2 percent per trade, and you would stop at a 50 percent drawdown.

First, the edge. Each trade wins 1.5R with probability 0.45 and loses 1R with probability 0.55. Expected value is 0.45 × 1.5 − 0.55 = 0.125R per trade. The edge is positive, so ruin is not certain. The model then normalizes that edge by the average trade size: A = 0.125 ÷ (0.675 + 0.55) ≈ 0.102.

Second, the risk units. Your ruin point sits 50 percent below the start, and each trade risks 2 percent. So U = 50 ÷ 2 = 25 units. Ruin means losing 25 risk units before your edge pulls you away from the cliff.

Third, the probability. The per-unit ratio is (1 − 0.102) ÷ (1 + 0.102) ≈ 0.815. Raise it to the 25th power: 0.815²⁵ ≈ 0.006. The model says roughly 0.6 percent. Now raise risk to 5 percent per trade. U drops to 10, and the same math gives about 12.9 percent. Same system, same edge, more than twenty times the danger. Nothing about the strategy changed. Only the position size did.

The model behind the number

This tool uses the fixed-fractional risk-of-ruin model published by Perry Kaufman in Trading Systems and Methods, in the payoff-aware form used across the risk literature. The formula is: risk of ruin = ((1 − A) ÷ (1 + A))^U. Here A is your per-trade edge normalized for the payoff ratio: A = (p×b − (1−p)) ÷ (p×b + (1−p)), where p is the winning-trade probability and b is the payoff ratio. U is the number of risk units between you and the ruin point: your drawdown limit divided by your risk per trade. With an even payoff (b = 1), A collapses to the classic gambler's-ruin edge of 2p − 1, so the payoff-aware form contains the textbook case.

The algebra hides a neat shortcut. That whole ratio simplifies to (1−p) ÷ (p×b) per risk unit. So each extra unit of buffer multiplies your ruin odds by the same factor, and you can check any output on a pocket calculator. Ralph Vince later refined ruin math further in his work on optimal f, but the fixed-fractional form remains the standard first tool.

Two behaviors are built in. If your expectancy is zero or negative, the tool prints 100 percent. With no positive edge, hitting the limit is a matter of time, and no sizing trick changes that. And the output is always clamped between 0 and 100 percent.

The assumptions, stated honestly

The formula assumes three things. You risk a fixed fraction on every trade. Each trade is independent of the last. And your winning percentage and payoff ratio stay stable forever. Real trading bends all three. Traders size up after wins, markets serve correlated losses in clusters, and edges drift. So treat the output as a comparative tool, not a literal probability. It ranks risk settings against each other with real force. It does not predict your future.

Why halving your risk does more than halving your ruin odds

Here is the core insight of the whole page. Risk per trade sits in the exponent, not in the base. Halve your risk and U doubles. The probability does not halve. It gets squared, as a fraction.

Run the worked example again. At 2 percent risk, U = 25 and the model prints about 0.6 percent. Drop to 1 percent risk. U becomes 50 and the result is roughly 0.006 squared: about 0.0036 percent. That is around 160 times safer, from one halving. Go the other way and the same lever punishes you. At 5 percent risk the number explodes to 12.9 percent. The table under the calculator shows this curve on your own inputs, and it is rarely what people expect.

This is why position sizing carries more practical weight than entry signals. Use the position size calculator to turn your chosen risk percent into an exact lot size for each stop distance. And if you want a ceiling derived from your edge rather than a round number, the Kelly criterion calculator computes it. Most traders should stay well below full Kelly, precisely because of the exponent you just watched work.

Where your inputs should come from

The model is only as honest as its inputs. Winning percentage and payoff ratio must come from real, closed trades. Memory inflates both. A written record does not. If you do not track trades yet, start with the trade journal and log every position, including the embarrassing ones.

Sample size matters. Twenty trades tell you almost nothing; the noise swamps the signal. One hundred trades give a usable first estimate. Recompute your stats every month and feed the fresh numbers back into this page. The expectancy calculator turns the same two inputs into your average R per trade, which is the edge this model needs to stay above zero.

Also know your break-even line. For any payoff ratio there is a minimum winning percentage below which the edge turns negative. The breakeven calculator finds it. If your recorded stats sit near that line, your true ruin number is far closer to 100 percent than any single backtest suggests.

Prop firm accounts: small U, big consequences

Funded accounts change the math in one brutal way. The ruin point is not 50 percent. Most programs terminate the account at an 8 to 10 percent drawdown, and the trailing versions are tighter still. That shrinks U dramatically, and U is the exponent.

Take the same system as before: 45 percent winners, 1.5 payoff. On a personal account with a 50 percent limit and 1 percent risk, U = 50 and the model prints about 0.004 percent. On a funded account with a 10 percent limit, the same 1 percent risk gives U = 10. The probability of breaching the limit jumps to roughly 12.9 percent. One funded account in eight fails on pure variance, with a genuinely profitable system.

Drop the risk to 0.5 percent and U doubles to 20. The number falls to about 1.7 percent. At 0.25 percent it is near 0.03 percent. This is why experienced funded traders risk a quarter to a half percent per trade. It is not timidity; it is the exponent. Before entering any challenge, map your buffer with the drawdown calculator, then set risk per trade so the model number here stays in low single digits at worst.

Honest limitations of any ruin estimate

No closed-form ruin number should be read as a forecast. The model ignores correlation between open positions, so three trades on EUR pairs can behave like one triple-size trade. It ignores slippage and gaps, which push real losses past the planned 1R. It assumes your stats are known exactly, when they are estimates from a noisy sample. And it assumes you follow the plan under pressure, which is the assumption that fails first.

Monte Carlo simulation relaxes some of these constraints, and serious system developers use both approaches side by side. But the fixed-fractional formula earns its place through transparency. You can verify every step by hand, as the worked example shows. Use it to compare risk settings, to sanity-check a challenge plan, and to see why the exponent deserves respect.

A practical habit: treat the output as a gate, not a grade. Pick a ceiling you can accept, say 1 percent, and refuse any risk setting that breaches it on your current stats. When your measured numbers worsen, the gate forces your size down before the market does. All tools on this site are built and checked under the Editorial and Testing Policy.

Want the full toolkit? Get the complete database.

FAQ

What winning percentage and payoff ratio should I enter?

Your own, measured from at least 100 closed trades. If you only have a backtest, use its statistics but shade them down; live results usually come in below tested ones. Never enter aspirational numbers, because the output inherits every ounce of their optimism.

Why does the calculator show 100%?

Your inputs describe a system with zero or negative expectancy. When p × b is less than or equal to 1 − p, each trade loses value on average. With no positive edge, hitting the limit is a matter of time, so the model reports certainty. Fix the edge first; sizing cannot rescue it.

Is the output the real probability my account fails?

No. It is a model estimate built on fixed fractional risk, independent trades, and stable statistics. Real trading violates all three somewhat. The number is most useful for comparison: the same inputs at different risk levels, or two systems at the same risk level.

What ruin point should I enter for a funded account?

Enter the program's maximum drawdown, usually 8 to 12 percent. If the limit trails your equity peak, the effective buffer is even smaller, so treat the model output as a floor rather than a ceiling and size below it.

If my number is very low, can I raise my risk per trade?

Be careful. The exponent works both ways, and raising risk multiplies the danger far faster than intuition expects. A low number also depends on inputs that drift over time. Re-measure your statistics regularly and test any sizing change on a demo account first. Results are not guaranteed; past performance is not indicative of future results.

Related tools: atr position size calculator, risk reward calculator and portfolio heat calculator, plus the full free forex tools directory.