Monte Carlo Simulation
Instead of predicting one outcome, Monte Carlo runs thousands of random scenarios to show the full range of what could happen to your portfolio.
Why one forecast isn't enough
Traditional forecasts give you a single number: "your portfolio will be worth $X in 20 years." But markets are uncertain. Monte Carlo simulation generates thousands of possible futures by randomly sampling from historical return distributions.
The process
Geometric Brownian Motion
The mathematical model behind each simulated price path.
Stock price
Expected return
Volatility
Random normal
Each simulation step takes the current price, applies the expected drift (μ), adjusts for volatility (σ), and adds a random shock. Repeating this across thousands of paths creates a probability distribution of outcomes.
Key metrics from simulation
| Metric | What it tells you | Good range |
|---|---|---|
| Median outcome | The most likely portfolio value | — |
| 5th percentile | Worst-case scenario (95% confidence) | Higher = safer |
| CAGR | Compound annual growth rate | 7–12% |
| Sharpe Ratio | Return per unit of risk | > 1.0 |
| Max Drawdown | Worst peak-to-trough decline | < 30% |
Reading the fan chart
The fan chart shows percentile bands from the simulation. The dark center band (25th–75th percentile) represents the most likely outcomes. The outer bands (5th–95th) show tail risks and best cases.
More paths = more precision
10,000 simulations give a smooth, reliable probability distribution.
Shape matters
A wide fan means high uncertainty. A narrow fan means predictable outcomes.
Risk management
Focus on the 5th percentile — that's your 'bad luck' scenario to plan for.
Not a prediction
Monte Carlo shows probabilities, not certainties. It's a tool for thinking, not fortune-telling.