The Problem with “Average Returns” #
Here’s something that trips up a lot of people. If someone tells you stocks return 10% on average, you might think: “Great, I’ll just multiply my portfolio by 1.10 each year for 30 years.”
But markets don’t move in straight lines.
Example:
- Year 1: +20% ($100 becomes $120)
- Year 2: -10% ($120 becomes $108)
The arithmetic average is 5%. But your actual compound growth? Only 3.9%. That’s a huge difference over decades.
This is why Monte Carlo simulation matters. Instead of pretending you’ll get steady returns, it generates thousands of random scenarios based on historical volatility. Some years you’re up 25%. Some years you’re down 35%. The simulation captures that reality.
What Monte Carlo Actually Does #
The name comes from the famous casino in Monaco—fitting, since it’s all about probability.
Here’s the basic process:
- Define your inputs: Starting balance, withdrawal rate, time horizon, asset allocation
- Generate random returns: Using historical averages and volatility for each asset class
- Run thousands of scenarios: Typically 10,000 to 100,000 simulations
- Analyze the distribution: What percentage succeeded? Failed? What’s the median outcome?
Each simulation is like one possible future. Maybe you retire right before a crash. Maybe you get lucky with a decade-long bull market. Monte Carlo shows you the full range of possibilities instead of just one made-up path.
Why Sequence of Returns Risk Matters #
This is the part that keeps early retirees up at night.
If you’re still working and contributing, a market crash early in your career is actually good—you’re buying cheap. But if you’re retired and withdrawing, an early crash is devastating. You’re selling low and depleting your portfolio right when it needs to recover.
Two retirees, same average return, wildly different outcomes:
| Scenario | Early Returns | Late Returns | Final Balance |
|---|---|---|---|
| Lucky | +15%, +12%, +8%… | -10%, -5%… | $2.1M |
| Unlucky | -10%, -15%, -8%… | +20%, +15%… | $400K |
Same 7% average. Completely different retirement.
Monte Carlo simulation captures this by randomizing the order of returns across thousands of scenarios. It shows you what happens when bad years hit early versus late.
The 4% Rule and Its Limitations #
You’ve probably heard of the 4% rule—withdraw 4% of your initial portfolio each year, adjusted for inflation, and you’ll probably be fine for 30 years.
The problem? That rule was based on historical US market data from 1926-1992. It assumed:
- A 30-year retirement (not 50+ years for FIRE folks)
- US market conditions continuing indefinitely
- No major structural changes to the economy
For a traditional retirement at 65, the 4% rule might still be reasonable. But if you’re retiring at 40 or 45, you need a longer runway—and Monte Carlo simulation lets you test that.
What the simulations typically show:
| Horizon | Safe Rate | Depletion Risk at 4% |
|---|---|---|
| 30 years | 3.5-4.0% | 5-10% |
| 40 years | 3.0-3.5% | 15-25% |
| 50 years | 2.5-3.0% | 30-45% |
These aren’t guarantees. They’re probability distributions. That’s the point.
Fat Tails and Black Swan Events #
Standard Monte Carlo uses a normal distribution (bell curve) for returns. But real markets have “fat tails”—extreme events happen more often than the bell curve predicts.
Think about it: We’ve had several supposedly “once in a century” crashes just since 2000:
- 2000-2002: Dot-com bust (-78% Nasdaq)
- 2008: Financial crisis (-57% S&P 500)
- 2020: COVID crash (-34% in three weeks)
A normal distribution says these shouldn’t happen that often. Reality disagrees.
Advanced Monte Carlo simulators offer a “fat-tail mode” using Student’s t-distribution instead of normal distribution. This generates more extreme scenarios—both crashes AND booms—giving you a more realistic stress test.
Choosing a Withdrawal Strategy #
Monte Carlo simulation can test different withdrawal approaches:
Constant Dollar (Traditional SWR): You withdraw a fixed inflation-adjusted amount regardless of portfolio value. Simple and predictable, but rigid—you might be taking money from a declining portfolio.
Dynamic Spending (Vanguard Method): You adjust withdrawals based on portfolio performance, with floors and ceilings to prevent wild swings. More sustainable for aggressive withdrawal rates, but requires flexibility in your spending.
The simulations consistently show dynamic spending reduces depletion risk significantly—often cutting it in half compared to constant dollar at the same withdrawal rate.
What the Numbers Can’t Tell You #
Monte Carlo is powerful, but it has blind spots:
Regime changes: The simulation assumes future volatility looks like past volatility. What if we enter a decade-long low-return environment like 2000-2010?
Structural shifts: AI disrupting the economy, demographic changes, climate impacts—none of these are modeled.
Personal factors: Health expenses, family obligations, housing costs—the simulator doesn’t know your life.
Taxes: Most simulators work with nominal returns. Your actual spending power depends on your tax situation.
Use Monte Carlo as one input, not the final answer. Build in safety margins. Stay flexible.
Building Your Safety Margin #
Don’t just trust the 5th percentile result. Add buffers:
- Regime change buffer: -20% (persistent low returns)
- Black swan buffer: -15% (major crisis early in retirement)
- Fee creep buffer: -5% (costs increase over time)
If your 5th percentile outcome is $1.7M, your conservative target might be $1.0M after applying these haircuts.
If you can live comfortably on that conservative number, you’re genuinely safe—not just “statistically probably fine.”
Ready to Run Your Own Simulations? #
Understanding the theory is one thing. Actually testing your numbers is where it gets real.
Try the Calculator
I built a Monte Carlo simulator that runs 10,000+ scenarios with four professionally designed portfolios. It includes fat-tail mode, dynamic spending strategies, and detailed risk metrics.
The calculator lets you test different withdrawal rates, compare portfolios, and see exactly how fees compound over decades. All calculations run in your browser—no data collected.
The Bottom Line #
Simple retirement calculators give you false precision. They show one number that pretends to know the future.
Monte Carlo simulation is honest about uncertainty. It shows you the range of possible outcomes and lets you make informed decisions about risk. You might not like seeing a 15% chance of running out of money, but it’s better to know that now than discover it at 85.
Plan for the 5th percentile. Hope for the median. Stay flexible enough to adjust if reality surprises you.
Additional Resources #
Recommended Reading #
- The Four Pillars of Investing by William Bernstein
- The Bogleheads’ Guide to Investing
- Early Retirement Now SWR Series
Communities #
Other Calculators #
- cFIREsim - Historical data backtesting
- Portfolio Visualizer - Detailed portfolio analysis
- FIRECalc - Historical retirement calculator