Monte Carlo Simulator for Trading: How It Works & When to Use One (2026)

You backtest a strategy over two years of historical data and the results look strong: 58% win rate, 1.8:1 average reward-to-risk, a maximum drawdown of 12%, and a net return of 47%. Confident in the numbers, you fund the account and start trading live.

Three weeks later, you are down 22% after a cluster of nine losses in the first twelve trades. The strategy's long-term edge hasn't changed, but the sequence of outcomes has been far worse than the backtest suggested was likely.

This is the fundamental limitation of single-path backtesting. One historical sequence of trades produces one equity curve, one maximum drawdown, and one final return. But that sequence is just one possible reality out of thousands. Monte Carlo simulation shows you the rest.

What Is the Problem With Single-Path Backtesting?

A backtest takes your strategy rules, applies them to historical price data, and generates a list of trades in chronological order. The resulting equity curve is treated as representative of how the strategy performs.

But that equity curve is an artifact of the specific order in which trades occurred. If the same 200 trades had occurred in a different sequence — the same wins and losses, just shuffled — the drawdown profile, peak equity, and recovery periods would all change.

Consider a strategy that produced 120 winners and 80 losers over a year. In the historical backtest, the largest losing streak was 5 trades. But a different arrangement of those same 200 trades could easily produce a losing streak of 8, 9, or even 12. The strategy's edge is identical in both cases. The only difference is luck in sequencing.

Single-path backtesting tells you what did happen. Monte Carlo simulation tells you what could happen.

What Does Monte Carlo Simulation Actually Do?

Monte Carlo simulation for trading takes your historical trade results — the list of individual wins and losses with their magnitudes — and reshuffles them into random sequences. Each reshuffle creates a new synthetic equity curve. Repeat this thousands of times and you get a distribution of possible outcomes.

The process works in three steps:

1. Extract trade results: Take the P&L of each individual trade from your backtest
2. Reshuffle randomly: Create a new random ordering of those same trades
3. Simulate the equity curve: Apply the reshuffled trades to the starting account balance, tracking the equity curve, maximum drawdown, and final balance

Repeat steps 2 and 3 for 5,000 to 10,000 iterations, and you have a statistical distribution of outcomes that all share the same underlying edge but differ in their sequencing.

The result is not a single number. It is a range: the best case, the worst case, and everything in between.

What Key Concepts Matter in Monte Carlo Simulation?

Trade Clustering and Streak Probability

Losses are not evenly distributed. In real trading and in Monte Carlo simulations, losses cluster into streaks. A strategy with a 55% win rate will, over 100 trades, almost certainly experience a losing streak of 6 or more. Over 500 trades, a streak of 9-10 becomes probable.

Monte Carlo makes these clusters visible. Instead of relying on the single historical losing streak from your backtest, you see the distribution of maximum losing streaks across thousands of simulations. If the 95th percentile maximum losing streak is 11 trades, you need to size your account to survive 11 consecutive losses — not the 5-trade streak from the backtest.

This has direct implications for position sizing. If your risk per trade would cause account failure after 8 consecutive losses, but Monte Carlo shows a realistic probability of 11-trade streaks, you are undersized for survival.

Equity Curve Confidence Bands

The most visually informative Monte Carlo output is the confidence band chart. This plots three or more equity curves from the simulation distribution:

• 5th percentile curve: The near-worst-case outcome. Only 5% of simulations performed worse than this.
• Median curve (50th percentile): The middle outcome. Half of simulations did better, half did worse.
• 95th percentile curve: The near-best-case outcome. Only 5% of simulations performed better.

The gap between the 5th and 95th percentile curves represents the range of outcomes you should prepare for. If the 5th percentile curve still ends in profit after your planned trading horizon, your strategy's edge is robust enough to survive bad sequencing. If the 5th percentile curve goes negative, your strategy may be profitable on average but carries meaningful risk of loss over any single evaluation period.

For prop firm challenges, the 5th percentile curve is the one that matters. Challenges fail on the worst outcomes, not the average ones.

Maximum Drawdown Distribution

The maximum drawdown from your backtest is a single data point. Monte Carlo produces a distribution of maximum drawdowns across all simulations. This distribution answers a question that single-path backtesting cannot: what is the realistic worst-case drawdown for my strategy?

If your backtest showed a 12% maximum drawdown, but the 95th percentile Monte Carlo drawdown is 28%, your strategy is roughly twice as risky as the backtest suggests. This is the number you should use when setting risk of ruin thresholds and determining whether your account can withstand the strategy's true risk profile.

How Do You Interpret Monte Carlo Results?

Rule 1: If the 95th Percentile Drawdown Exceeds Your Tolerance, Reduce Size

Your tolerance is the maximum drawdown you can absorb without abandoning the strategy or breaching an account limit. For personal accounts, this might be 20-30%. For prop firm challenges, it is whatever the firm's maximum drawdown rule specifies.

If the 95th percentile Monte Carlo drawdown exceeds that number, the path forward is to reduce risk per trade until it doesn't. The risk of ruin calculator can help you find the position size that aligns your Monte Carlo worst case with your actual tolerance.

Rule 2: If the 5th Percentile Equity Is Still Profitable, Your Edge Is Robust

The 5th percentile represents near-worst-case sequencing. If even this scenario ends in profit, your strategy has enough edge to overcome bad luck. This is the strongest signal of a viable trading system: profitability even when the cards fall poorly.

If the 5th percentile is flat or negative, you have a strategy that is profitable on average but unreliable over any single evaluation period. This matters enormously for prop firm challenges and for traders who need consistent monthly returns.

Rule 3: The Gap Between Median and Worst Case Measures Fragility

A wide gap between the median and 5th percentile outcomes indicates a fragile strategy — one that is highly sensitive to trade sequencing. This typically occurs when:

• The win rate is near 50% with a moderate R:R (small edge, high variance)
• The strategy takes few trades (small sample size magnifies sequencing effects)
• Individual trade outcomes are highly variable (some big winners, many small losers)

A narrow gap indicates a robust strategy — the outcome is similar regardless of how trades are ordered. This usually correlates with higher win rates, consistent trade sizing, and a large number of trades.

How Do You Use Monte Carlo for Prop Firm Challenges?

Prop firm challenges introduce hard constraints that make Monte Carlo simulation particularly valuable. You are not just asking "will my strategy be profitable?" but "will my strategy hit a 10% profit target before breaching a 10% drawdown limit within 30 days?"

To answer this, configure the Monte Carlo simulation with:

• Starting balance: The challenge account size
• Profit target: The challenge's required gain (e.g., 8% or 10%)
• Maximum drawdown: The challenge's hard limit (e.g., 10% or 12%)
• Daily loss limit: The single-day maximum loss (e.g., 5%)
• Trade frequency: How many trades you take per day
• Challenge duration: Maximum number of trading days

The simulation outputs a pass rate: the percentage of 10,000 simulated challenges that reached the profit target without breaching either drawdown limit. If the pass rate is 70%, roughly 7 out of 10 challenge attempts should succeed — meaning on average you need 1.4 challenge fees to pass.

The prop firm simulator is built specifically for this analysis, and the Monte Carlo simulator handles the general-purpose distribution calculations.

What Are the Limitations of Monte Carlo Simulation?

Monte Carlo simulation is powerful, but it operates under assumptions that traders should understand.

Trade Independence

The standard Monte Carlo approach assumes each trade is independent — the outcome of trade N has no effect on trade N+1. In practice, this is violated when:

• You trade correlated instruments (multiple forex pairs sharing a common currency)
• Market regime shifts cause extended periods of poor performance across all strategies
• Your execution quality degrades during drawdowns (wider stops, premature exits)

These dependencies mean the real worst case may be worse than Monte Carlo predicts, because losses tend to cluster more in reality than in randomized simulations.

Slippage and Execution Degradation

Monte Carlo reshuffles historical trade P&L values, but those values were calculated under specific market conditions. If your strategy scales up, slippage may increase. If you trade during volatile news events, fills may worsen. These factors are not captured in the simulation.

Garbage In, Garbage Out

Monte Carlo cannot improve a bad backtest. If the underlying trade data is curve-fitted, overly optimized, or derived from unrealistically clean execution assumptions, the simulation will produce optimistic results. The quality of Monte Carlo output is directly proportional to the quality of the input data.

Before running Monte Carlo, ensure your backtest methodology is sound. The how to backtest on TradingView guide covers the fundamentals of producing reliable backtest data.

What Is the Practical Workflow: Backtest, Simulate, and Size?

The complete process for validating a strategy before deploying capital:

Step 1: Backtest thoroughly. Generate at least 100 trades (ideally 200+) across multiple market conditions. Record each trade's P&L.

Step 2: Run Monte Carlo simulation. Use the Monte Carlo simulator to reshuffle those trades across 5,000-10,000 iterations. Review the confidence bands, drawdown distribution, and pass rates.

Step 3: Size based on the 95th percentile drawdown. Not the backtest drawdown, not the median Monte Carlo drawdown, but the near-worst-case. Use the position size calculator to translate this into concrete lot sizes.

Step 4: Validate with risk of ruin. Confirm that your position size produces a risk of ruin below 1%. If it doesn't, reduce size until it does. This is the final check before live capital is at risk. The risk of ruin guide walks through the complete formula and interpretation.

Step 5: Re-evaluate periodically. As you accumulate live trades, rerun Monte Carlo with updated data. Your live performance may differ from the backtest, and the simulation should reflect reality, not historical assumptions.

This workflow replaces the common mistake of sizing based on a single backtest's maximum drawdown — a number that represents one sequence among thousands of possibilities. Monte Carlo doesn't eliminate risk. It quantifies it, so your position sizing decisions are grounded in the full distribution of outcomes rather than a single fortunate (or unfortunate) historical path.

Consistent application of this process is what separates strategies that survive 10 consecutive losses from those that look profitable right up until they blow up.

What Questions Do Traders Ask About Monte Carlo Simulation?

How many simulations should I run for reliable Monte Carlo results?

A minimum of 1,000 iterations produces usable results. For more statistically stable output, 5,000 to 10,000 iterations is standard. Beyond 10,000, the marginal improvement in precision is negligible for most trading applications. The Monte Carlo simulator defaults to an appropriate iteration count for this reason.

Can Monte Carlo simulation predict whether my strategy will be profitable?

Not exactly. Monte Carlo simulation does not predict the future — it reveals the range of outcomes that are consistent with your historical trade data. If your backtest shows a genuine edge, Monte Carlo will confirm that the edge persists across most random sequences. But it cannot account for future market regime changes, strategy degradation, or execution differences between backtesting and live trading.

How is Monte Carlo different from risk of ruin calculation?

The risk of ruin formula provides a single probability based on your strategy's average edge and position size. It assumes trade independence and a fixed win rate. Monte Carlo simulation uses your actual trade distribution (including variable win/loss sizes) and produces an entire range of outcomes. Monte Carlo is more granular — it shows drawdown distributions, confidence bands, and pass probabilities — while risk of ruin gives a quick, closed-form probability. The two are complementary: use risk of ruin for quick sizing checks and Monte Carlo for thorough strategy validation.

Should I use Monte Carlo on live trades or backtest results?

Both, but at different stages. Before deploying a strategy live, run Monte Carlo on backtest results to set initial position sizes. Once you have accumulated 50-100 live trades, rerun the simulation using live data. Live results typically show wider drawdowns and lower win rates than backtests due to slippage, emotional interference, and market changes. Updating your Monte Carlo inputs with live data ensures your sizing remains appropriate for your actual performance, not your idealized backtest performance.