Monte Carlo Simulation: Stress-Testing Your Portfolio With Thousands of Scenarios
Why Single-Point Estimates Fail
Traditional financial planning uses averages: “If stocks return 10% and bonds return 5%, your portfolio will be worth X in 30 years.” The problem? Markets don’t deliver average returns in any given year. A 20% loss in year one followed by 20% gains creates a very different outcome than steady 10% returns, even if the average is similar.
Monte Carlo fixes this by modeling the randomness. It uses historical return distributions — including the fat tails, crashes, and booms — to generate realistic scenarios. The output isn’t a single number but a probability distribution: “You have an 85% chance of maintaining your spending through age 95.”
How Monte Carlo Simulation Works
| Step | What Happens | Example |
|---|---|---|
| 1. Define Inputs | Set portfolio allocation, expected returns, volatility, correlations | 60% stocks (10% return, 15% vol), 40% bonds (4% return, 5% vol) |
| 2. Generate Random Returns | Draw random annual returns from the defined distributions | Year 1: +18%, Year 2: -12%, Year 3: +7%… |
| 3. Simulate Portfolio Path | Calculate portfolio value year by year, including contributions and withdrawals | $500K → $590K → $519K → $555K… |
| 4. Repeat 10,000+ Times | Run the same process with different random draws | 10,000 different 30-year portfolio paths |
| 5. Analyze Results | Calculate success rate, percentile outcomes, worst cases | 87% success rate, median ending value $2.1M |
Reading Monte Carlo Results
A Monte Carlo output typically shows percentile bands — the 10th percentile (bad luck), 25th, 50th (median), 75th, and 90th percentile (good luck) outcomes. Here’s how to interpret them:
| Percentile | Meaning | Use Case |
|---|---|---|
| 10th percentile | Only 10% of scenarios were worse | Your realistic worst case — plan for this |
| 25th percentile | Below-average but not catastrophic | Conservative planning baseline |
| 50th percentile (median) | The middle outcome — half better, half worse | Your “expected” outcome |
| 75th percentile | Better than average markets | Optimistic scenario |
| 90th percentile | Exceptional market conditions | Don’t count on this |
Monte Carlo for Retirement Planning
The most common use of Monte Carlo in personal finance is answering: “Will I run out of money in retirement?” You input your current savings, annual contributions until retirement, expected spending in retirement, asset allocation, and time horizon. The simulation tells you what percentage of scenarios sustain your spending throughout retirement.
Financial planners generally consider an 80–90% success rate acceptable. Below 80%, you should adjust — save more, spend less, work longer, or take more investment risk. Above 95% suggests you could spend more or invest more conservatively.
Monte Carlo vs. Historical Backtesting
| Feature | Monte Carlo Simulation | Historical Backtesting |
|---|---|---|
| Data Source | Random draws from statistical distributions | Actual historical returns |
| Number of Scenarios | Unlimited (typically 10,000+) | Limited by history length |
| Can Model Events Never Seen | Yes — generates novel combinations | No — limited to what actually happened |
| Sequence Risk | Fully captured — different random orderings | Only historical sequences available |
| Weakness | Results depend on input assumptions | Past may not represent the future |
Limitations of Monte Carlo
Monte Carlo is powerful but imperfect. The biggest limitation: garbage in, garbage out. If your assumptions about expected returns, volatility, or correlations are wrong, the simulation results will mislead you. Most Monte Carlo models also assume returns follow normal distributions — but real markets have fatter tails (more extreme events) than normal distributions predict.
Other limitations: most simulations don’t model changing behavior (you’ll probably spend less if markets crash), they assume constant correlations (which break down in crises), and they can create false precision — an “87.3% success rate” looks scientific but depends entirely on input assumptions that are themselves uncertain.
Free Monte Carlo Tools
You don’t need expensive software. Several free tools run solid Monte Carlo simulations: Portfolio Visualizer (portfoliovisualizer.com), FIRECalc (firecalc.com), and cFIREsim (cfiresim.com) all offer retirement-focused Monte Carlo analysis. Many robo-advisors and brokerage platforms also include Monte Carlo tools in their planning suites.
Key Takeaways
- Monte Carlo simulation runs thousands of random scenarios to show the probability distribution of portfolio outcomes — not just a single expected return.
- It captures sequence risk (the order of returns matters) and market randomness that single-point estimates miss.
- For retirement planning, aim for an 80–90% success rate across scenarios. Below 80% means your plan needs adjustments.
- Results are only as good as the inputs — run simulations with multiple assumption sets to test robustness.
- Free tools like Portfolio Visualizer and FIRECalc make Monte Carlo accessible to any investor.
Frequently Asked Questions
How many simulations should a Monte Carlo run?
At least 10,000 scenarios for reliable results. Most tools default to 10,000, and increasing beyond that rarely changes the results meaningfully. The law of large numbers ensures that 10,000 simulations produce stable probability estimates.
What is a good Monte Carlo success rate for retirement?
Most financial planners target 80–90% success rates. Below 80% suggests meaningful risk of running out of money. Above 95% may mean you’re being too conservative and could afford to spend more or take less risk. An 85% success rate is a common sweet spot — high enough for comfort, low enough that you’re not over-saving.
Is Monte Carlo better than the 4% rule?
Monte Carlo provides more nuanced guidance. The 4% rule is a useful starting point but assumes a fixed withdrawal rate regardless of market conditions. Monte Carlo shows the probability of success for any withdrawal rate and lets you model dynamic spending — spending more after good years and less after bad years.
Can Monte Carlo predict market crashes?
No. Monte Carlo doesn’t predict specific events — it models the probability of various outcomes including crashes. Some simulated scenarios will include market crashes; others won’t. The value is seeing how your portfolio performs across the full range of possibilities, not in predicting which scenario will actually occur.
How often should I rerun Monte Carlo simulations?
At least annually, or whenever your circumstances change significantly (job change, inheritance, major expense, change in goals). Markets don’t change the fundamentals of your simulation — your inputs do. Update your portfolio value, contribution rate, and time horizon each year and rerun to see if you’re still on track.