How to Calculate Risk of Ruin From Your Trading Journal Data

Risk of ruin is the probability that a run of losses, given your current edge and position sizing, wipes out your account (or drops it below a level you consider unrecoverable) before your edge has a chance to play out. It does not measure a single bad trade. It measures what happens over the next 50 or 200 trades if your current habits continue unchanged. Wikipedia's overview of the concept defines it as the likelihood of "losing all one's investment capital or extinguishing one's bankroll below the minimum for further play," and its key point is one worth sitting with: a strategy can have a positive win rate and still carry a high risk of ruin if position sizing is too aggressive relative to that edge (Wikipedia).

You need three inputs, and all three should come from your journal rather than assumption.

The reason this has to come from logged trades and not from your trading plan is simple: plans describe intent, journals describe behavior. A trader whose plan says "risk 1% per trade" but who journals shows an average risk of 2.3% because of size creep on "high conviction" setups is working with a materially different risk of ruin than the plan implies.

A commonly used approximation for risk of ruin, once you have those three inputs, restates the problem in terms of a per-trade edge and the number of risk units your account holds. One widely cited simplified form is:

RoR = ((1 - A) / (1 + A))^N

Where A is your per-trade edge (a function of win rate and risk-reward) and N is the number of "risk units" in your account, which is 1 divided by your risk-per-trade percentage. A trader risking 2% per trade has 50 risk units; a trader risking 5% per trade has only 20. That difference alone can move risk of ruin from low single digits to well over 40%, even with the same win rate, because N shrinks so much faster than A changes.

This is why position sizing dominates the calculation more than most traders expect. Two traders with an identical win rate and risk-reward ratio can have wildly different risk of ruin purely because one risks 1% per trade and the other risks 5%.

Say your journal shows, over your last 80 trades: a 45% win rate, an average risk-reward ratio of 1.8 to 1, and an average risk of 1.5% of account equity per trade. Plug those into the framework above and the drop-in matters far less than the shape of the relationship: at a 45% win rate with a 1.8R average reward, the strategy has a real edge, but at 1.5% risk per trade, that edge has roughly 67 risk units of runway to survive drawdowns before the strategy has time to express itself. Drop the risk per trade to 0.5% (200 risk units) and the same edge tolerates a much longer losing streak. Raise it to 4% (25 risk units) and a normal losing streak, one your win rate guarantees will happen periodically, has a real chance of doing lasting damage.

This is also where a related concept, expectancy, is worth checking alongside risk of ruin rather than instead of it. Expectancy (win rate multiplied by average win, minus loss rate multiplied by average loss) tells you whether a strategy is profitable on average. Risk of ruin tells you whether you can survive long enough, at your current sizing, to collect on that average. A strategy can have positive expectancy and still be oversized to the point of high ruin risk.

Traders systematically misjudge their own win rate and average risk-reward when working from memory. Recency bias inflates recent losing streaks or recent winning streaks depending on what just happened, and it is easy to remember the trades that "should have worked" as wins in spirit. A journal with real entries, exits, and position sizes removes that distortion. It is also the only way to catch risk-per-trade creep, since almost no trader risks a perfectly consistent percentage on every trade; the plan says one number and the fills say another.

Pulling win rate, average risk-reward, and average risk per trade from an actual trade log, broken down by symbol, setup, or time period, is the kind of analytics Astro Trading Journal is built to surface without a separate spreadsheet. Astro calculates net and gross P&L, win rate, and average win versus average loss automatically from logged or synced trades, with breakdowns by symbol, asset class, direction, and setup, so the three risk of ruin inputs are already sitting in the account's performance view rather than needing to be reconstructed by hand.

Risk of ruin is not a number you calculate once. Win rate and average risk-reward drift as market conditions change, as a strategy gets refined, or as a trader's discipline around position sizing improves or slips. A trader whose win rate degrades from 50% to 40% over a rough quarter, while risk per trade quietly creeps from 1% to 2.5% because of a few oversized "revenge" trades, can move from a low-risk setup to a genuinely dangerous one without ever changing their stated trading plan.

The practical habit is to recompute the three inputs on a rolling basis, monthly or after every 30 to 50 trades, rather than relying on a figure calculated once at the start of a strategy. Reviewing trades on a set cadence, weekly or monthly, is also how most of the drift in average risk per trade actually gets caught before it compounds into a real drawdown.