A trader can be right more often than wrong, collect two dollars on every winner for each dollar lost, and still drive the account to zero. It sounds impossible until you’ve sized the positions too large and run into an ordinary losing streak. That’s what risk of ruin measures: the probability an account loses a defined catastrophic share of its starting capital before the edge has enough trades to compound into a gain. It’s a survival question, and it sits underneath every other number on the plan.
Most traders check expectancy and stop there. Positive expectancy tells you the average trade makes money. It says nothing about the path you take to get there, and the path is where accounts die. Risk of ruin fills that gap with a sharper question. Given your win rate, your payoff ratio, and the fraction you put at risk each time, what are the odds a bad run ends the game before the math can work in your favour?
What risk of ruin actually measures
Risk of ruin is a probability, not a level. It answers one thing: how likely is it that your account falls to a point you’ve decided is fatal, before your positive edge has time to show up in the equity curve. Most traders set that fatal point well above zero. A 50% drawdown is a common marker, because an account cut in half needs a 100% gain just to get back to even, and few traders or strategies recover from there with their discipline intact.
You get to choose that threshold, and the choice matters more than it looks. Define ruin as a total wipeout to zero and almost any sane sizing looks safe. Define it instead as the drawdown that ends your ability to trade the plan calmly, often somewhere between 30% and 50%, and the honest numbers get a lot less comfortable. Set the threshold where your discipline actually breaks, not where the balance hits nothing.
Three numbers drive the calculation. Your win rate, the share of trades that end green. Your average payoff ratio, how large the average winner is next to the average loser. And the fraction of capital you risk on each trade. Change any one of them and the ruin probability moves. The first two describe your edge. The third describes your restraint, and it’s the one you set by hand.
Why a real edge still blows up accounts
A positive expectancy doesn’t signal that you’ll survive the drawdown that produces it. Expectancy is an average taken over many trades. Ruin happens on the specific ordering of wins and losses you actually get, and losing streaks aren’t rare events you can plan around. They’re guaranteed to arrive. The only open question is when, and how deep.
Take a trend-following system that wins 40% of the time. Every trade then carries a 60% chance of losing. The probability of eight losses in a row is 0.6 to the eighth power, about 1.7%. That sounds small until you remember a system trades hundreds of times. Across two hundred trades, a run of eight or more straight losers will typically show up more than once. When I count the streaks in my own journal, the clusters are always worse than the ones I pictured before I looked. The edge is real. The variance around it is just as real, and it doesn’t care that the average is positive. This is why expectancy measured in R multiples tells you what to expect on average and nothing about whether you live to collect it.
The three inputs, and the one you actually control
Of the three inputs, two are stubborn. Your win rate and payoff ratio come from your strategy meeting the market, and you can’t simply decide to improve them. Traders who try usually make things worse. Tighten your stops to lift the win rate and you’ll get shaken out of good trades, which drags the payoff ratio down by more than the win rate rises. Those two numbers sit on a seesaw, walked through in win rate versus payoff ratio, and pushing one tends to cost you the other.
The risk fraction is different. You set it directly, before the trade, with no argument from the market. It’s the one lever in the ruin equation that answers to you completely. That’s why sizing discipline, rather than a hunt for a better win rate, is where survival gets won or lost.
Why cutting risk per trade helps far more than it costs
Reducing the fraction you risk lowers your ruin probability much faster than it lowers your growth, because the effect compounds across a long sequence of trades. A small cut to the per-trade risk stacks up over dozens of losses into a very different drawdown, and the size of that difference catches most traders off guard.
I ran the same fourteen-loss streak at two sizes. Risking 5% per trade, fourteen straight losses compound to 0.95 to the fourteenth power, leaving the account near 49% of where it started, roughly a 51% drawdown. Risking 2% per trade, the same fourteen losses give 0.98 to the fourteenth, about 75% of starting capital, a 25% drawdown. The streak is identical. The outcomes aren’t close. The 5% sizer now needs a 105% gain to recover. The 2% sizer needs about 33%. One of those holes is climbable. The other, for most accounts, isn’t.
That asymmetry is the whole argument for smaller size. Position sizing stops being bookkeeping here and becomes the primary control on ruin, which is the through-line of the main position sizing methods. The Kelly criterion gives the fraction that maximises long-run growth, and it’s worth knowing where that ceiling sits, but most traders size well below it. They trade a little growth for a large cut in the odds of a fatal drawdown. Full Kelly is mathematically optimal for growth and psychologically brutal in the drawdowns it’s willing to accept.
Estimating your own ruin curve from your journal
You don’t need to trust a textbook number. You’ve already got the two edge inputs in your own trade history. Count your closed trades, take the share that were winners for your win rate, and average the size of winners against losers for your payoff ratio. With those in hand, you can test candidate risk fractions two ways.
The classical risk-of-ruin formula gives a quick estimate, but it assumes every win and every loss is the same size. Real trading is messier than that, with winners and losers scattered across a range of R multiples. The more honest approach resamples your actual results. Take your real sequence of R multiples, shuffle it ten thousand times at a given risk fraction, and count how many of those simulated equity paths breach your ruin threshold. That share is your estimated risk of ruin at that size. Run the same shuffle at 1%, at 2%, and at 3%, and you get a small table of ruin probabilities that shows exactly what each step up in size costs you. It connects straight to Monte Carlo equity curves, the realistic way to model a system whose trade sizes vary rather than sitting at one fixed win or loss.
One caution. The classical formula tends to understate ruin for a variable-size system, because a handful of oversized losers can do damage the fixed-size assumption never sees. When the simple formula and the simulation disagree, trust the simulation.
Trend following and mean reversion need different sizing
A single fixed rule, risk 2% on everything, isn’t automatically safe. Two systems with very different shapes tolerate similar risk per trade only when the sizing is calibrated to each one’s own ruin curve. A trend-following system wins less than half its trades and leans on a few large winners. A mean-reversion system wins often and banks small, frequent profits. Both can be sound. They fail in different ways.
The low-win, high-payoff system spends long stretches in drawdown between its big winners, so its ruin risk comes from streak length. The high-win, low-payoff system rarely loses, but when it does the losers run large next to the small wins, so its ruin risk comes from the occasional cluster of outsized losses. Ralph Vince built much of his work on this exact problem, and Ralph Vince’s mathematics of money management shows how betting past a strategy’s optimal fraction turns a winning system into a guaranteed loser over enough trades. A percentage copied from a book, without checking it against your own variance, is a guess wearing the costume of a rule.
Sizing is the lever you hold
Risk of ruin is a property of two things multiplied together: your strategy’s statistics and the fraction you choose to risk. The same system can be perfectly safe or quietly doomed depending only on that fraction. Your win rate and payoff ratio drift slowly and resist your effort to move them. The risk fraction shifts the moment you decide it should, and it moves ruin probability more than anything else on the page. Size for the streak you haven’t seen yet, estimate your own curve instead of borrowing one, and keep the fraction low enough that an ordinary bad run stays survivable. Learn the pattern. Ride the trend. Keep the gains.
Educational content only. Not investment advice. Trading involves risk. You are responsible for your decisions.