Last Updated: August 05, 2026
A Monte Carlo simulation is a statistical method that takes your trading strategy's historical results and reshuffles them thousands of times to show the full range of outcomes you might experience going forward. Instead of relying on a single backtest that shows one sequence of wins and losses, Monte Carlo generates thousands of alternative sequences, revealing the probability of ruin, the range of expected drawdowns, and the confidence intervals around your returns. TradeZella offers a free Monte Carlo Simulator that runs this analysis on any strategy in seconds.
Here's why this matters. A backtest tells you what DID happen. Monte Carlo tells you what COULD happen. And the gap between those two things is where most traders get blindsided.
Your EMA crossover strategy produced a 45% win rate and a 1.8:1 reward-to-risk ratio over 200 trades. Solid numbers. But those 200 trades happened in a specific order. What if the losing streaks had clustered differently? What if your worst month had happened right at the start, when your account was smallest? Monte Carlo answers those questions by running your exact win rate and R:R through 10,000 randomized sequences. Some of those sequences will produce beautiful equity curves. Others will hit drawdowns that would make you quit. The distribution across all 10,000 tells you how likely each outcome is.
What Is a Monte Carlo Simulation?
The name comes from the casino district in Monaco. The idea is simple: when you can't calculate the exact probability of a complex outcome, you simulate it thousands of times and count the results.
Think of it like this. You flip a coin 100 times. Mathematically, you expect 50 heads. But run that experiment 10,000 times and you'll see results ranging from 35 heads to 65 heads. Most outcomes cluster near 50, but the tails exist. Some runs produce 10 heads in a row. Others produce 12 tails in a row. None of those outcomes are "wrong." They're all possible with a fair coin.
Trading works the same way. Your strategy has a win rate and an average win-to-loss ratio. Those numbers define the mathematical expectation. But the ORDER of wins and losses is random. And order matters enormously when you're compounding a live account. A losing streak at the beginning of the year hits differently than one in August, because the account size is different.
Monte Carlo simulation takes your trade statistics and randomizes the order of outcomes thousands of times. Each simulation is a "path." Plot all 10,000 paths on one chart and you get an equity curve fan, a visual showing the range from best case to worst case, with the most likely outcomes in the middle.
How Does Monte Carlo Simulation Work in Trading?
Every Monte Carlo simulation needs three inputs from your trading data:
Input 1: Win rate. The percentage of trades that are profitable. A 45% win rate means 45 out of every 100 trades are winners.
Input 2: Average win and average loss size. Usually measured in R-multiple terms. If your average winner is 1.8R and your average loser is 1.0R, your risk-reward ratio is 1.8:1.
Input 3: Number of trades. How many trades the simulation should run per path. More trades gives a clearer picture of long-term behavior. Most simulations use 100 to 500 trades per path.
Some simulators also let you input risk per trade (as a percentage of account), starting account balance, and maximum drawdown threshold.
Here's what happens under the hood:
- The simulator creates a virtual "trade" for each slot. Based on your win rate, each trade is randomly assigned as a winner or loser. A 45% win rate means each trade has a 45% chance of being a winner.
- Winners are assigned your average win size (e.g., +1.8R). Losers are assigned your average loss size (e.g., -1.0R). More advanced simulators sample from your actual distribution of wins and losses rather than using a flat average.
- The simulator runs these trades sequentially against your starting account balance, applying position sizing rules (flat dollar risk or percentage-based) at each step.
- That's one "path." The simulator repeats this 1,000 to 10,000 times, each with a different random order.
- The results are aggregated: median outcome, best case, worst case, and the distribution of maximum drawdowns across all paths.
What Does a Monte Carlo Simulation Tell You?
A well-run simulation produces four outputs that every trader should understand.
1. Probability of Ruin
This is the big one. Probability of ruin is the percentage of simulated paths where your account dropped below a threshold you couldn't recover from. Usually that threshold is 50% drawdown or total account loss.
If 3 out of 10,000 paths hit 50% drawdown, your probability of ruin is 0.03%. That's safe. If 800 out of 10,000 paths hit that level, your probability of ruin is 8%. That should keep you up at night.
Professional traders aim for a probability of ruin below 5%. Below 1% is excellent. Above 10% means something needs to change, usually your position sizing or your stop loss distance. The lever that moves this number fastest is risk per trade. Drop from 2% risk to 1% risk and watch your probability of ruin collapse.
2. Maximum Drawdown Distribution
Your backtest showed a maximum drawdown of 12%. But that was one sequence. Monte Carlo shows the RANGE of possible drawdowns across all paths. You might see that the median max drawdown is 15%, the 95th percentile is 22%, and the worst case is 31%.
That 95th percentile number is the one to plan around. It tells you: "In 95 out of 100 possible futures, your drawdown will stay below 22%." If you can't stomach 22%, you need to reduce your risk per trade. Our drawdown management guide covers the three-tier protocol for handling drawdowns at each severity level.
3. Equity Curve Fan
Plot all simulated paths on one chart and you see a fan shape. The paths spread apart over time because compounding amplifies small differences in win/loss order. The top edge of the fan shows the luckiest sequence. The bottom edge shows the unluckiest. The middle shows where most outcomes cluster.
If the bottom edge of the fan at your trade count still shows a positive account, that's a strategy worth trading. If the bottom edge dips significantly negative, the strategy carries too much ruin risk at your current position size.
4. Confidence Intervals on Returns
Monte Carlo produces confidence bands around your expected return. A 90% confidence interval might say: "After 200 trades, your account will be between $52,000 and $78,000, starting from $50,000, with 90% confidence." That remaining 10% includes the tails, both better and worse than the band.
This is more honest than a single backtest number. Your backtest says "$65,000 after 200 trades." Monte Carlo says "somewhere between $52,000 and $78,000, probably." One of those is actionable. The other is a guess dressed up as a fact.
How Do You Run a Monte Carlo Simulation?
You don't need to code anything. TradeZella's free Monte Carlo Simulator on our tools page runs the full analysis. Here's the step-by-step process:
Step 1: Gather your trade data. You need your win rate, average win size, and average loss size. If you've been tracking trades in TradeZella, these numbers live on your analytics dashboard. Filter by Strategy to get stats for the specific setup you want to simulate. If you don't have live data yet, use the results from your backtest with TradeZella.
Step 2: Enter your inputs. Plug in your win rate, average win (in R or dollars), average loss, risk per trade (as a percentage of account), starting balance, and number of trades to simulate. For beginners: start with 200 trades and 1% risk.
Step 3: Run the simulation. The tool generates thousands of paths in seconds. You'll see the equity curve fan, probability of ruin, drawdown distribution, and confidence intervals.
Step 4: Evaluate the results. Check three numbers first: probability of ruin (below 5%?), 95th percentile max drawdown (can you handle it?), and median ending balance (worth your time?).
Step 5: Adjust and re-run. If your probability of ruin is too high, reduce your risk per trade and run again. If the median return is too low, either the strategy needs more edge or you need more trades. This iterative process is the whole point. You're finding the position size that maximizes returns while keeping ruin probability acceptable.
How Do You Read Monte Carlo Results?
Let's walk through a concrete example. You've backtested a momentum strategy on ES futures and found these stats over 150 trades:
- Win rate: 48%
- Average winner: +2.1R
- Average loser: -1.0R
- Trading expectancy: +0.47R per trade
- Profit factor: 1.94
You run a Monte Carlo simulation with a $50,000 starting balance, 1% risk per trade ($500), and 200 trades per path.
Reading the equity curve fan: The median path ends around $69,000 after 200 trades. That's a 38% return. The 75th percentile path ends at $78,000. The 25th percentile ends at $61,000. The worst 5% of paths end below $56,000, still positive but barely. The best 5% exceed $85,000.
Reading the drawdown distribution: The median max drawdown across all paths is 9.2%. The 95th percentile max drawdown is 16.4%. That means in 95 out of 100 possible futures, your worst drawdown stays under $8,200 (16.4% of $50,000). Can you sit through an $8,200 drawdown without changing your strategy? If yes, this is tradeable. If no, drop to 0.75% risk and re-run.
Reading probability of ruin: With 1% risk and these statistics, the probability of hitting a 25% drawdown is 2.1%. The probability of hitting 50% drawdown is essentially zero. These numbers are solid.
Now watch what happens when you change one input. Same strategy, same stats, but 2% risk per trade instead of 1%. The median path jumps to $91,000. Exciting. But the 95th percentile max drawdown jumps to 28.6%, and the probability of hitting a 25% drawdown rises to 14.8%. Higher returns, but now you have a meaningful chance of a drawdown that makes most traders quit.
That tension between return and ruin risk is what Monte Carlo is built to quantify.
What Is Probability of Ruin and How Do You Lower It?
Probability of ruin measures the likelihood that your account hits a loss level you define as unrecoverable. For most retail traders, that threshold is somewhere between 25% and 50% drawdown. For prop firm traders, it's whatever the firm's maximum drawdown limit is (typically 5-10%).
The math is asymmetric, which is why this matters so much. A 25% drawdown requires a 33% gain to recover. A 50% drawdown requires a 100% gain. Our drawdown recovery guide has the full table, but the pattern is clear: it gets exponentially harder to come back. That's why preventing deep drawdowns is more important than maximizing returns.
Three levers control your probability of ruin:
Lever 1: Risk per trade. This is the single most powerful lever. Cutting risk from 2% to 1% per trade roughly halves your expected return but can reduce your probability of ruin by 80% or more. Use a Position Size Calculator to keep this consistent across different instruments and stop loss strategies.
Lever 2: Win rate. A higher win rate reduces ruin probability because losing streaks are shorter and less frequent. You improve win rate by being more selective about your setups, finding your trading edge, and trading only during optimal conditions (time of day, market regime, volatility).
Lever 3: Win/loss ratio. Larger average wins relative to average losses mean that even a 40% win rate can produce positive expectancy with low ruin probability. The combination of win rate and win/loss ratio determines your risk-reward ratio, which feeds directly into the simulation.
Target benchmarks: if your probability of ruin at 1% risk is below 1%, your strategy is robust. Between 1% and 5% is acceptable for most retail traders. Above 5% means either the edge is thin or the position sizing is too aggressive. Above 10%, don't trade it with real money.
How Does Monte Carlo Compare to Backtesting?
They're complements, not substitutes. One doesn't replace the other.
Backtesting tells you whether a strategy had an edge in the past. You define rules, run them against historical data, and measure the results. Our guide on how to backtest a trading strategy covers this step by step, and automated backtesting lets you write rules in plain English and run them across years of data in seconds.
Monte Carlo tells you how fragile or robust that edge is under different conditions. A strategy can have a positive backtest but a dangerous Monte Carlo profile. That happens when the edge is small and the position sizing is large, meaning the strategy is one bad streak away from a catastrophic drawdown.
The correct workflow is:
- Backtest the strategy to confirm it has a positive expectancy over enough trades
- Run a Monte Carlo simulation on the backtest results to check ruin probability and drawdown distribution
- Adjust position sizing until the Monte Carlo profile shows acceptable ruin probability
- Forward test with real money at the validated position size
This process is part of the broader system-building framework in our how to build a trading system guide, where Monte Carlo validation sits between backtesting and forward testing as the bridge that confirms your strategy can survive real-world randomness.
How Does Position Sizing Affect Your Monte Carlo Results?
Position sizing is the single variable that changes Monte Carlo results the most dramatically. Same strategy, same win rate, same R:R. Only change: the percentage of account risked per trade.
Here's what happens with a 48% win rate and 2.1:1 R:R strategy on a $50,000 account over 200 trades:
- 0.5% risk ($250/trade): Median ending balance $59,000. Max drawdown 95th percentile: 8%. Probability of 25% drawdown: 0.1%. Safe, but slow.
- 1% risk ($500/trade): Median ending balance $69,000. Max drawdown 95th percentile: 16%. Probability of 25% drawdown: 2.1%. Sweet spot for most traders.
- 2% risk ($1,000/trade): Median ending balance $91,000. Max drawdown 95th percentile: 29%. Probability of 25% drawdown: 14.8%. Aggressive. Many traders will hit the drawdown threshold and quit.
- 3% risk ($1,500/trade): Median ending balance $118,000. Max drawdown 95th percentile: 41%. Probability of 25% drawdown: 38%. You're gambling at this point.
The pattern is nonlinear. Doubling risk from 1% to 2% increases median return by about 32% but increases the probability of a severe drawdown by 700%. That's the trap. The return looks tempting. The ruin risk hiding behind it is disproportionate.
For most traders, 0.5% to 1% risk per trade produces the best Monte Carlo profile: positive returns with ruin probability under 5%. This aligns with the risk management framework in our pillar article, where 1% risk is the foundation of the entire four-layer system.
What Mistakes Do Traders Make With Monte Carlo Simulations?
Mistake 1: Using unreliable input data. If your backtest only has 30 trades, the win rate and R:R estimates are noisy. Monte Carlo amplifies that noise across thousands of paths. Garbage in, garbage out. Minimum 50 trades for a usable simulation. 100+ trades gives much more reliable inputs.
Mistake 2: Ignoring the distribution of wins and losses. Basic simulators use flat averages: every winner is exactly +1.8R, every loser is exactly -1.0R. Real trading has fat tails. Your average winner might be 1.8R, but that includes a +5R outlier and many +1R winners. Advanced simulators sample from the actual distribution. If your tool uses flat averages, your results underestimate tail risk.
Mistake 3: Running too few paths. A simulation with 100 paths doesn't capture the tails. At 1,000 paths, you start seeing the rare but dangerous outcomes. At 10,000 paths, the probability estimates stabilize. Always run at least 1,000 paths. 10,000 is better.
Mistake 4: Optimizing position sizing THEN trading with those stats. If you ran Monte Carlo at 1.5% risk and it showed acceptable ruin probability, but then your live results slightly underperform your backtest (which is normal), you're suddenly in the danger zone. Build in a margin of safety. If Monte Carlo says 1.5% is the maximum safe risk, trade at 1%.
Mistake 5: Treating Monte Carlo as a crystal ball. Monte Carlo assumes your future win rate and R:R will match your past performance. Markets change. Strategies degrade. Regimes shift. Run the simulation again every quarter with updated stats. If your trading edge is decaying, the simulation will show it through rising ruin probability.
Where Does Monte Carlo Fit in Your Trading Workflow?
Monte Carlo isn't a daily tool. It's a checkpoint in your system-building process and a periodic health check on live strategies.
When building a new strategy: After your backtest produces at least 50 trades with positive expectancy, run Monte Carlo to validate position sizing. This comes before forward testing. It's the gate between "this strategy has edge" and "this strategy won't blow up my account." Write up the validated risk parameters in your trading plan.
When evaluating live performance: Every 50 to 100 trades, re-run Monte Carlo with your live stats. Compare the simulation output to your earlier validation. If the probability of ruin has increased, something changed. Either the edge degraded, your execution slipped, or market conditions shifted. This is a signal to investigate, not to panic.
When scaling up: Before increasing position size, run Monte Carlo at the new risk level. If going from 1% to 1.5% risk pushes your probability of 25% drawdown from 2% to 8%, that might not be a tradeoff you're willing to make. Let the math decide, not your confidence.
| Risk Per Trade |
Dollar Risk ($50K) |
Median Ending Balance |
95th % Max Drawdown |
Prob. of 25% DD |
Verdict |
| 0.5% |
$250 |
$59,000 |
8% |
0.1% |
Conservative, very safe |
| 1% |
$500 |
$69,000 |
16% |
2.1% |
Sweet spot for most traders |
| 1.5% |
$750 |
$79,000 |
22% |
7.4% |
Moderate risk, experienced only |
| 2% |
$1,000 |
$91,000 |
29% |
14.8% |
Aggressive, drawdown likely |
| 3% |
$1,500 |
$118,000 |
41% |
38% |
Dangerous, near-coin-flip ruin |
| 5% |
$2,500 |
$165,000 |
58% |
67% |
Gambling, not trading |
Based on 48% win rate, 2.1:1 R:R, $50,000 starting balance, 200 trades per path, 10,000 simulations. Results are illustrative. Your actual outcomes will vary based on your strategy's specific statistics.
Key Takeaways
- Monte Carlo simulation reshuffles your trade results thousands of times to show the full range of possible outcomes, not just the single path your backtest produced.
- Probability of ruin measures the chance your account hits a catastrophic drawdown. Target below 5%. Below 1% is excellent.
- Position sizing is the biggest lever. Cutting risk per trade from 2% to 1% can reduce severe drawdown probability by 80% or more while only reducing median returns by about 25%.
- Monte Carlo and backtesting are complements. Backtest first to confirm edge. Monte Carlo second to confirm the position size won't blow you up.
- Plan around the 95th percentile max drawdown, not the median. If you can't handle the bad scenario, reduce size until you can.
- Re-run the simulation every 50 to 100 live trades with updated statistics. Strategy edge changes over time, and your Monte Carlo profile should reflect current reality.
Frequently Asked Questions
What is a Monte Carlo simulation in trading?
A Monte Carlo simulation in trading is a statistical method that takes your strategy's win rate, average win size, and average loss size, then randomizes the order of trades across thousands of simulated paths. Each path represents one possible future for your strategy. The aggregated results show the probability of ruin, the range of expected drawdowns, and confidence intervals around your returns. It answers the question: "Given my strategy's statistics, how bad could things realistically get?"
How many trades do I need before running a Monte Carlo simulation?
You need a minimum of fifty trades to generate reliable input statistics for a Monte Carlo simulation. At fifty trades, your win rate and average win-to-loss ratio estimates are usable but still noisy. One hundred or more trades produces significantly more reliable inputs. Using fewer than fifty trades means the simulation is amplifying uncertainty in your data across thousands of paths, which produces misleading results.
What is a good probability of ruin?
Professional traders target a probability of ruin below five percent. Below one percent is considered excellent and indicates a robust strategy with appropriate position sizing. Between one and five percent is acceptable for most retail traders. Above five percent means either the strategy edge is thin or the position sizing is too aggressive. Above ten percent, the strategy should not be traded with real capital until either the edge improves or the risk per trade decreases.
How does Monte Carlo differ from backtesting?
Backtesting tells you whether a strategy had an edge in the past by running defined rules against historical data. Monte Carlo tells you how fragile or robust that edge is under different conditions by reshuffling the trade results thousands of times. A strategy can have a positive backtest but a dangerous Monte Carlo profile if the edge is small and the position sizing is large. The correct workflow is to backtest first, then run Monte Carlo to validate position sizing before forward testing.
Can Monte Carlo simulation help prop firm traders?
Monte Carlo simulation is especially valuable for prop firm traders because it quantifies the probability of hitting the firm's specific drawdown limits. For example, you can simulate the probability of hitting ten percent total drawdown on an FTMO account given your strategy statistics and position sizing. If the probability is above five percent, you should either improve your statistics, reduce your risk per trade, or find a firm with more generous drawdown limits.
What is the most important Monte Carlo output?
The most important output is the ninety-fifth percentile maximum drawdown. This tells you that in ninety-five out of one hundred possible futures, your worst drawdown will stay below this level. If you cannot psychologically or financially handle that drawdown level, you should reduce your risk per trade until the ninety-fifth percentile drawdown drops to a level you can sustain without changing your strategy or quitting.
How often should I run a Monte Carlo simulation?
Run a Monte Carlo simulation at three points: when first validating a new strategy after backtesting, every fifty to one hundred live trades to check for edge decay, and before increasing your position size. Market conditions and strategy performance change over time, so the simulation should be re-run with updated statistics periodically. If your probability of ruin has increased since the last run, investigate whether your edge has degraded before continuing to trade the strategy at the same risk level.
