Your equity curve only shows one path — the trades you actually took, in the order they happened. But that order was partly luck. Shuffle your wins and losses differently and you'd have ended up somewhere else, sometimes much worse along the way. The Monte Carlo simulation answers the question your single equity curve can't: given my edge, what range of outcomes — and drawdowns — should I realistically expect, and how likely am I to blow up if I size up?
What it does
The tool takes your historical per-trade R-multiples (how much each trade made or lost as a multiple of the risk you took) and replays them thousands of times in random order, sizing each trade at a risk level you choose. Each replay is one possible "future." Stack up all those futures and you get a distribution of outcomes rather than a single line — the spread of where you might end up, and how deep the dips could get along the way.
It resamples with replacement, so it's not just reshuffling your exact history; it's drawing from the same win/loss profile to build genuinely new sequences, including unlucky streaks you haven't hit yet but easily could.
Prerequisite: the simulation runs on trades that carry a defined risk — a stop loss, or a defined-risk options position — because that's what gives each trade an R-multiple. Trades with no risk basis are skipped.
Running a simulation
- Open the Monte Carlo tool and click Run simulation. (It runs on demand rather than automatically, so the page stays fast.)
- Set Risk per trade — the percentage of your account you'd risk on each trade (e.g. 1%). This is the single biggest lever: it scales both your growth and your drawdowns.
- Set the Horizon — how many trades into the future to project (e.g. the next 100 trades).
- Read the fan chart and the probability table (below).
Adjust the two controls and re-read the results. The same filter, risk, and horizon always produce the same projection — the simulation is deterministic, so you can compare two settings fairly and revisit a result later.
Reading the fan chart
The chart plots cumulative return (%) over the horizon. Instead of one line, you get a "fan" of percentile bands:
- Median line — the middle outcome. Half your simulated futures did better, half worse.
- 25–75% band — the typical range; half of all outcomes land inside it.
- 5–95% band — the wider range. Only the luckiest/unluckiest 5% fall outside it.
A fan that climbs steadily and stays narrow means a robust, consistent edge. A fan that's wide, or whose lower band dips below zero, is telling you that variance — not just your average edge — will dominate your experience over this horizon.
Reading the drawdown & risk-of-ruin table
This is the part most traders should look at first. It shows the probability of hitting each maximum drawdown depth somewhere along the way:
| Max drawdown ≥ | Probability |
|---|---|
| 10% | … |
| 20% | … |
| 30% | … |
| 50% (ruin) | … |
Risk of ruin is the chance of a 50%+ drawdown — the kind of hole that's mathematically brutal to climb out of (a 50% loss needs a 100% gain just to get back to even). Even a profitable system can carry an uncomfortable risk of ruin if you size too aggressively.
How to act on it
- If risk of ruin is uncomfortably high → size down. Lower the "Risk per trade" and re-run. You'll usually find the median outcome drops only modestly while the deep-drawdown probabilities fall sharply. That trade-off is the whole point of position sizing.
- Pair it with the Position Sizing tool. Monte Carlo tells you what a given risk level feels like over time; the Kelly / optimal-f suggestions tell you what risk level your edge can mathematically support. Use them together.
- Watch the lower bands, not just the median. Plan for the 5th-percentile path, because eventually you'll live through one.
Important caveats
The simulation is educational, not financial advice, and it rests on assumptions that won't hold perfectly:
- It assumes the future resembles the past. Your projected edge is only as good as the history behind it — a small or unrepresentative sample (a hot streak, a single regime) will mislead. More trades = more trustworthy.
- It assumes trades are independent. Real trading has correlation and clustering (losing streaks in choppy markets, correlated positions) that can make actual drawdowns worse than the model suggests.
- It assumes your edge is stable. If your strategy is decaying (see the Edge Trend view), past R-multiples overstate your future.
Treat the output as a risk-awareness tool — a way to pressure-test your sizing and build respect for variance and drawdown — not a prediction of what your account will do. Past performance is no guarantee of future results.
Comments
0 comments
Article is closed for comments.