If you have ever used a retirement calculator that told you a single FIRE number and a single retirement age, you have been given a point estimate. The calculator picks one set of assumptions, runs the maths once, and prints a result. It tells you nothing about the odds your plan actually works. That is where Monte Carlo simulation comes in.
This post explains the difference between the 4% rule and Monte Carlo, why probability-based planning matters more the earlier you retire, and how to read a Monte Carlo chart without getting overwhelmed.
The short version
Quick refresher: what the 4% rule actually says
The 4% rule comes from William Bengen's 1994 paper “Determining Withdrawal Rates Using Historical Data” in the Journal of Financial Planning, later reinforced by the 1998 Trinity Study (Cooley, Hubbard & Walz). It asks a very specific question: if I withdraw a fixed percentage of my starting portfolio in year one, then adjust that dollar amount for inflation every year after, what is the highest withdrawal rate that would have survived 30 years across historical US market data?
The answer, roughly, was 4%. Withdraw 4% in year one, adjust for inflation each year, and your portfolio had a very high chance of lasting 30 years. Flip it and you get the 25× rule: multiply your desired annual spending by 25 to get your target portfolio.
The 4% rule has three important quiet assumptions:
- A 30-year retirement horizon
- US market returns, mostly 20th-century data
- A fixed withdrawal schedule that ignores market conditions
Every one of those assumptions is fragile for an Australian early retiree. FIRE retirements often last 40–60 years. Future returns may not look like the past. And real people change their spending when markets tank.
What Monte Carlo actually does
Monte Carlo is a fancy name for a simple idea: instead of running your plan once with one set of assumed returns, run it thousands of times with different random returns drawn from a realistic distribution. Count how many versions of your plan succeeded. That ratio is your probability of success.
ProjectFi runs 5,000 simulations by default. Each simulation:
- Generates a unique sequence of annual returns, using your expected return and an assumed volatility
- Applies inflation, taxes, super rules, and any life events in your plan
- Tracks your portfolio year by year until you die (default: age 90)
- Records whether the portfolio survived the whole retirement
The output is not a single number but a distribution. A typical result might look like:
Here's what one actually looks like. A typical Australian FIRE couple: both 42, combined household income of $180k, $400k in super between them, planning to retire at 60. The chart below shows how 1,000 simulated futures for that plan land over the next 50 years.
The wider emerald band covers the bottom 10% of outcomes through to the top 10%. The darker band inside is the middle half. The teal line running through it is the typical outcome, the median. Notice how the band widens as age advances. That is sequence-of-returns risk you can actually see. Two retirees who both average 7% long-run can end up in very different places depending on when those good and bad years fall.
The headline number, plan success rate, is the proportion of simulated futures where the plan both reaches FIRE by the target age and still has positive net worth at 90. It is the strictest honest definition of "did the plan work", because it fails on either a missed target date or portfolio ruin.
The big reason Monte Carlo matters: sequence-of-returns risk
Two retirees can have the exact same average return over 30 years and land in wildly different places. The difference is when the good and bad years happened.
Imagine two identical $1M portfolios, each withdrawing $40k per year inflation- adjusted. Each averages 7% returns over 30 years. But retiree A gets -15%, -10%, -5% in years 1–3. Retiree B gets +20%, +15%, +10% in years 1–3. Both average out to 7% eventually, but:
- Retiree A is withdrawing from a shrinking portfolio during the worst years. By year 4 their balance is well below $800k, so every future withdrawal is a bigger slice of what remains.
- Retiree B built a buffer in the good years. Their portfolio can absorb the same eventual downturn without threatening the plan.
Sequence-of-returns risk is the single biggest hidden risk in early retirement, and it is exactly what a point estimate hides. A 4% rule calculation gives you an average path that probably never happens. Monte Carlo gives you the full spread of paths, including the ugly ones.
How to read a Monte Carlo result without panic
When you first look at a fan chart with its cloud of possible futures, it is easy to feel like your plan is hopeless because some simulations end in failure. Don't. Most plans fail in some simulations. What matters is:
- Success rate (%): the headline number. 85–95% is generally considered a healthy plan. Above 95% may indicate you are over-saving and could retire earlier or spend more. Below 80% is a warning.
- Median terminal balance: the 50th percentile ending balance. If this is still in the millions, your plan is very likely to succeed even though the worst cases might fail.
- 10th percentile terminal balance: the bottom 10% of outcomes. This is the "unlucky but not impossibly unlucky" path. If this is zero, you are relying on a lucky draw of returns.
- Shape of the fan: a narrow fan means your plan is robust to variance. A wide fan means small changes in returns swing the outcome a lot.
A single 92% success rate is not a licence to stop thinking. It is a signal that, under the assumptions you gave the model, your plan holds up in most plausible futures. Play with the inputs: drop returns to 5% real, push inflation to 3.5%, see what happens. A plan that still hits 85% under pessimistic inputs is a plan you can actually trust.
Try it yourself
Run 5,000 simulations on your plan
ProjectFi's Monte Carlo simulator runs 5,000 trials on your actual profile, including Australian tax, super preservation, and Age Pension. Get a success probability, a fan chart, and a histogram of FIRE ages.
Run the numbers4% rule vs Monte Carlo: side by side
| 4% rule | Monte Carlo | |
|---|---|---|
| What it tells you | One number (FIRE target) | A probability and a range of outcomes |
| Horizon | 30 years | Whatever you set (often 40–60 for FIRE) |
| Handles sequence risk | No | Yes |
| Handles one-off events | No | Yes (inheritances, property sales, career breaks) |
| Typical use | Quick sanity check | Stress-testing the plan before you commit |
How many simulations are enough?
Most retirement planning tools that do offer Monte Carlo use 100–500 runs, because running a full simulation for each trial is computationally expensive. ProjectFi runs 5,000 trials per plan by default. This matters because:
- Fewer runs produce noisier success rates. 100 simulations might tell you your plan is 83% successful when the true number is closer to 90%, just because one or two bad runs swung the sample.
- 1,000 runs give stable success rates to within a percentage point or two.
- For FIRE plans where the decision is binary (retire now or keep working), that stability is the difference between acting on signal and acting on noise.
So what should you actually do?
- Start with the 4% rule as a back-of-envelope target. It is fast and approximately right.
- Run Monte Carlo before you quit. If you are within a few years of FIRE, a probability-based check is worth the 10 seconds it takes.
- Look for a success rate of 85%+ under realistic assumptions. Above 95% you may be over-saving.
- Stress-test. Drop returns 1–2% and raise inflation 0.5%, and see whether the plan still holds.
- Re-run annually. Your plan is a living thing. Markets, income, expenses, and life goals change. A 92% plan today can be an 80% plan in two years if you are not paying attention.
The bottom line