Monte Carlo Simulation Trading: When It's Actually Useful
3 minutes
Monte Carlo Simulation Trading: When It's Actually Useful
Picture a new investor with a two-stock portfolio being told to run 10,000 simulations on it. The advice sounds sophisticated, but with only two holdings, the main risk is already obvious: everything rides on two companies.
That tension sits at the heart of Monte Carlo simulation trading. Used well, it's a powerful stress test for a rules-based strategy. Used badly, it's complex math that creates a false sense of precision. This guide covers what the tool does, where it earns its place, and where it's overkill.
What Is Monte Carlo Simulation?
A Monte Carlo simulation runs a random process thousands of times to map out the range of possible results. Instead of producing one forecast, it produces a spread of outcomes and shows how likely each part of that spread is under the model's assumptions.
Think of rolling two dice. You can't predict the next roll, but roll them 10,000 times and a clear pattern emerges: sevens show up most often, while twos and twelves are rare. Monte Carlo applies the same idea to uncertain processes, including trading. The method was first conceived at Los Alamos in 1946 by mathematician Stanislaw Ulam and later developed with John von Neumann and others.
The illustration below applies the same principle to geometry.
Source: Gironi.it
How Monte Carlo Simulation Trading Works
In trading, the random process is usually built from one of two sources:
Past trades. You take the list of trades from a strategy's backtest and shuffle their order.
Assumed distributions. You set assumptions about how prices move, such as average change, volatility, and the shape of the distribution, then generate thousands of artificial price paths.
The first stays close to real data; the second depends on whether its assumptions are right.
Monte Carlo Backtesting: Reshuffling the Trade Sequence
A backtest shows one historical path, with trades arriving in a specific order. That order was partly luck. Monte Carlo backtesting asks what happens if the same trades had arrived in a different sequence.
Each shuffle keeps every trade but changes the order. In the simplest setup, the final tally is unchanged, but the journey from start to finish can look very different. Back-to-back losses hit very differently than the same losses spread out.
Before reshuffling anything, though, the backtest itself needs to be sound, because the usual backtesting pitfalls carry straight through.
When It's Actually Useful
Robustness Testing a Rules-Based Strategy
Robustness testing asks whether a strategy's results depend on a lucky arrangement of events. If the historical path looks smooth but most reshuffled paths look far rougher, the smooth history may have been the exception. Monte Carlo fits rules-based strategies best because they produce a clear, repeatable list of trades to resample.
Estimating a Monte Carlo Drawdown Range
A drawdown is the drop from a peak to a later low. A single backtest shows one worst drawdown. A Monte Carlo drawdown analysis shows a range of possible worst drawdowns, because reordering trades changes how losses bunch together. The point isn't a number; it's that history's worst case may not be yours.
Comparing Position Sizing Rules
Position sizing, meaning how much capital goes into each trade, shapes the path as much as the trades themselves. Monte Carlo simulation trading lets you run the same trade list under different sizing rules and compare how widely the possible paths spread. A wider spread signals more path risk, even when the trade list is identical.
When It's Overkill
Monte Carlo isn't always the right tool. It adds little value when:
The portfolio is tiny or concentrated. With one or two holdings, the risk is visible without any modeling.
There are too few trades. Reshuffling a handful of trades just recycles the same small sample.
Decisions aren't rule-based. Discretionary calls can't be written down consistently, so there's nothing clean to resample.
The goal is to "prove" a forecast. Monte Carlo simulation trading explores uncertainty. It can't confirm that a strategy will work.
Monte Carlo Simulation Limitations
Every simulation is a model, and models inherit their builders' blind spots. Two Monte Carlo simulation limitations matter most.
Fat Tails and the Normal-Distribution Trap
Simulations that generate price paths often assume returns follow a normal distribution, or bell curve, which treats extreme moves as vanishingly rare. Benoit Mandelbrot's 1963 paper in The Journal of Business argued that speculative price changes are better described by fat-tailed distributions than by the bell curve, meaning extreme moves occur more often than predicted. The comparison below shows the difference.
Image courtesy of David Spratt via ResearchGate.
If the model's tails are thinner than reality's, the simulation will understate how severe the worst paths can be.
Garbage In, Garbage Out
Monte Carlo can't fix a flawed input. Overfitting is a well-studied problem: a 2015 paper by Bailey, Borwein, López de Prado and Zhu proposes a framework for estimating the probability that a backtest is overfit. If a backtest is overfit, ignores trading costs, or uses data a trader wouldn't have had at the time, reshuffling those trades just produces thousands of versions of the same mistake.
A Quick Checklist Before You Run One
Clear rules. Can the strategy be written down so that anyone would place the same trades?
Enough trades. Is the sample large enough that reshuffling reveals something new?
Realistic costs. Do the trades include commissions, spreads, and slippage?
Fat-tail awareness. If you're generating price paths, does the model allow for extreme moves?
Range, not promise. Will you read the output as a spread of possibilities rather than a prediction?
If any of the first three look shaky, step back and validate the strategy itself before simulating anything.
The Bottom Line
Monte Carlo simulation trading is a stress test, not a crystal ball. It's most useful for rules-based strategies with plenty of trades, where reshuffling reveals path risk and drawdown ranges that a single backtest hides. For small, concentrated, or discretionary portfolios, simpler questions usually tell you more.
Frequently Asked Questions
What is Monte Carlo simulation in trading?
It's a method that reruns a strategy's trades, or simulated price paths, thousands of times at random. Monte Carlo simulation trading maps the range of paths a strategy could take instead of producing a single forecast.
How does Monte Carlo backtesting work?
It takes the trades from a backtest and shuffles their order many times. In the simplest setup the final tally stays the same, but the path from start to finish changes with each shuffle.
Why does robustness testing matter for a trading strategy?
It checks whether a strategy's results depend on a lucky order of events. If most reshuffled paths look much rougher than the historical one, the smooth history may have been the exception.
What does a Monte Carlo drawdown analysis show?
It shows a range of possible worst peak-to-trough declines, created by reordering the same trades. It's a reminder that the worst drop in history may not be the worst one you experience, not a prediction.
What are the main Monte Carlo simulation limitations?
Results are only as good as the inputs: reshuffling can't fix an overfit or cost-free backtest. Models that assume a bell curve can also understate fat tails, meaning extreme moves that occur more often than predicted.
This content is educational and general. It is not investment, legal, or tax advice, is not a recommendation to buy or sell any security, and does not consider your individual circumstances. Any securities or strategies mentioned are illustrative only. Consult a qualified professional about your situation.
Other blogs