Monte-Carlo simulation for trading strategies
Your backtest is just one history out of many that could have happened. Monte-Carlo simulation replays thousands of alternative histories so you can see how much of the result was skill — and how bad the drawdown could plausibly get.
What is a Monte-Carlo simulation?
A Monte-Carlo simulation takes your strategy's trades and reshuffles or resamples their order hundreds or thousands of times, building a whole distribution of equity curves instead of a single one. The exact sequence of wins and losses in your backtest was partly luck; Monte-Carlo maps the range of outcomes that same edge could realistically have produced.
What it tells you
- Probability of profit — in what fraction of simulated histories you ended up green.
- Drawdown distribution — not just the one drawdown you saw, but the plausible worst cases (e.g. the 95th-percentile drawdown). Measure a drawdown here.
- Risk of ruin — the chance of a loss deep enough to blow past your pain threshold.
- How lucky the ordering was — if your real curve sits at the lucky edge of the fan, be cautious.
How to read the fan chart
The result is usually a "fan" of percentile bands (5th, 25th, median, 75th, 95th). If your actual equity curve hugs the top of the fan, the backtest flattered you. If the median is still clearly profitable and the worst-case drawdown is survivable at your position size, that's a good sign.
How many simulations?
About 1,000 resamples is a stable, standard choice — enough to estimate percentile drawdowns and the probability of profit reliably. QUANTHEON Lab runs 1,000 by default and folds the result, with Walk-Forward and a Deflated-Sharpe haircut, into a single honesty verdict.
FAQ
What is a Monte-Carlo simulation in trading?
It reshuffles or resamples the order of a strategy's trades many times to build a distribution of possible equity curves, revealing how much of the result was lucky ordering.
How many Monte-Carlo simulations are enough?
Around 1,000 resamples is a common, stable choice — enough for reliable percentile drawdowns and probability of profit.
Does Monte-Carlo fix overfitting?
No — it measures luck in the trade sequence, not whether the parameters were curve-fit. Pair it with Walk-Forward and a Deflated-Sharpe check.
Related: What is overfitting? · Walk-Forward analysis · Max drawdown calculator