Monte Carlo Simulation in Backtesting, Explained
A Monte Carlo backtest takes the trades your strategy made and reshuffles or resamples them thousands of times to show the range of outcomes that could have happened, not just the single path history handed you. Real history is only one roll of the dice. If your winners and losers had shown up in a different order, your equity curve and your worst losing streak would look different. Monte Carlo builds thousands of these alternate curves so you can see a realistic spread: a typical result, a lucky one, and an ugly one. That helps you answer a key question, which is how much of your backtest was skill in the rules versus plain luck in the ordering. It gives you context about worst-case drawdowns and consistency. It can't predict the future, and it only reshuffles what already happened.
A single backtest is one path through history. Monte Carlo reshuffles the trades thousands of times to show the realistic range of drawdowns and outcomes.
Key points
- Monte Carlo takes your strategy's list of trades and reorders or resamples them many times to build thousands of alternate equity curves.
- History gives you only one ordering of wins and losses, and a different order could have produced a very different worst losing streak.
- Looking at the full spread of outcomes shows you a typical result, a lucky one, and an ugly one, instead of a single number.
- It's especially useful for estimating how deep a drawdown you might realistically face, which one backtest can badly understate.
- The method rests on assumptions, like treating trades as independent, so the output is only as sound as those assumptions.
- Monte Carlo gives you context about luck and consistency, but it can't predict the future and only reshuffles what already happened.
Frequently asked questions
What does a Monte Carlo simulation actually do to my backtest?
It takes the trades you already have and shuffles their order (or resamples them) thousands of times, drawing a new equity curve each time. Stacking all those curves together shows you the range of results that could plausibly have happened, not just the one path history dealt.
Why isn't a single backtest result enough?
Because that one result depends heavily on the exact order your wins and losses arrived. Move a big losing streak to the start and your account might not have survived it. Monte Carlo reveals how fragile or sturdy your result is by testing many possible orderings.
What do the percentiles or ranges mean in plain terms?
They describe outcomes from good to bad. A middle (or median) line is a typical result. A bottom slice, like the worst 5%, shows a rough bad-case scenario you should be mentally and financially ready for. Planning around the ugly end is safer than hoping for the average.
Can Monte Carlo predict what my strategy will do next?
No. It only rearranges trades that already happened, so it can't see new market conditions coming. It's a tool for understanding risk and luck in your existing results, and it should be read as context, never as a forecast.
Where can I see a Monte Carlo view of my strategy in Agenticks?
When the AlgoAgent backtests a strategy, you can ask it to run a Monte Carlo pass and show you the spread of possible outcomes and drawdowns, explained in plain words inside the chat. It's part of how the agent stress-tests an idea rather than a separate tool you configure.
Related on Agenticks
This content is for educational purposes only and does not constitute financial advice. Trading involves risk, including possible loss of capital.