Published 2026-10-14
Monte Carlo simulation
Monte Carlo simulation re-runs a strategy over many randomised paths to see the range of outcomes, not just the one that happened. It turns a single backtest number into a distribution.
On this page
What it is
Monte Carlo simulation re-runs a strategy across many randomised paths to see the range of possible outcomes.
The backtest you ran is one path. Monte Carlo asks: what were the other ones?
Why one path is not enough
A single backtest gives a single drawdown — the one that happened.
| Question | Single backtest | Monte Carlo |
|---|---|---|
| Worst case | Whatever occurred | A distribution of worst cases |
| Risk of ruin | Unseen if it didn’t happen | Estimated directly |
| Confidence | None | A range |
The number that ends a strategy is usually not the one in the backtest — it is a worse path that was possible.
How to check
Ask what was randomised (order of returns, slippage, parameters) and how many paths. Reshuffling the same returns tests sequence risk; it does not test whether the edge is real. Those are two different questions.