Robustness
Monte-Carlo simulation in trading
7 min read · by the GetBacktest team
Your backtest produces a single equity curve — one possible path among infinitely many. Monte-Carlo simulation answers a more honest question: “what if the same trades had happened in a different order?” By reshuffling the deck thousands of times, it reveals the range of plausible futures, not just the one that occurred.
The idea
You start from your list of trades (or your statistics: win rate, wins, losses). You replay thousands of sequences by drawing outcomes at random from that distribution, producing thousands of different equity curves.
Each curve is a credible “what if”. Aggregated, they give a distribution of outcomes rather than a single, reassuring number.
What it measures
The range of final returns (P10 to P90 percentiles), typical and extreme maximum drawdown, and above all the risk of ruin: the share of simulations that hit a critical loss threshold.
This reading changes everything: two strategies with the same average return can have radically different risks of ruin. Monte-Carlo makes that risk visible before it hits your account.
Why a single curve misleads
Your backtest curve is the product of one particular order of wins and losses. A simple rearrangement — seven losses clustered earlier — could have ruined you before the edge paid off.
Trusting one trajectory means confusing “what happened” with “what could have happened”. Monte-Carlo corrects that bias by showing the real dispersion of outcomes.
How to read it
Look at the worst decile (P10) and the risk of ruin, not just the median scenario. If the P10 is untenable or the risk of ruin non-negligible, cut risk per trade — the most effective lever.
On GetBacktest, Monte-Carlo runs on your replay trades and shows the P10–P90 band, the median and the ruin threshold. You see at a glance whether your plan survives variance.
Don't believe it — prove it
Backtest this concept on real data, tick by tick, and get a robustness verdict. 7 days of Pro free, no card.
Start for freeFrequently asked questions
What is Monte-Carlo used for in trading?
To estimate the dispersion of a strategy's possible outcomes: drawdown, return range and, above all, risk of ruin — by replaying thousands of trade sequences.
How many simulations are needed?
A few thousand usually stabilize the percentiles. What matters is having enough real trades as input: a distribution drawn from 15 trades stays unreliable.
Does Monte-Carlo replace walk-forward?
No, they're complementary. Walk-forward tests robustness out-of-sample; Monte-Carlo tests robustness to trade order and quantifies risk of ruin.
Read next