Monte Carlo simulations: thinking about your plan in probabilities
A calculator with a fixed return gives you a date: "you get there in 24 years". The market will not give you that return every year, and the order of good and bad years changes the outcome (that is sequence risk). A Monte Carlo simulation rehearses thousands of possible futures and counts how many turn out well.
How to read a probability of success
In the simulator, a life "succeeds" if it reaches the goal and the money then lasts every year of retirement you chose.
- 85% does not mean "almost certain": in 15 futures out of 100 something fails.
- 100% is not a guarantee either: it only says none of the simulated futures failed under those assumptions.
- More useful than the exact figure is how it changes when you pull a lever: saving more, withdrawing less, retiring a year later.
Many FIRE plans aim for 85-95% and add flexibility: spending a bit less in bad years raises the real probability a lot (see dynamic withdrawal strategies).
Two ways to imagine the future
Random model. Each year a return is drawn with the mean and volatility you set. It is flexible (you can model any portfolio) but treats every year as independent: it does not reproduce streaks like three bad years in a row.
Historical model. Instead of drawing, it chains real 10-year stretches of the US market since 1871 (Robert J. Shiller's data), each starting at a random year. So 1929-1932, the 1973-1974 stagflation or 2008 show up exactly as they happened, streaks included. You choose the share in stocks and the rest goes into 10-year Treasury bonds, rebalanced every year.
With the simulator's default values (€24,000 spending, €20,000 invested, €800/month, 4% withdrawal and 40 years of retirement), the random model with 5% real return and 15% volatility gives around 70% success, and the historical one with a 60/40 portfolio around 90%. The gap is not magic: a 60/40 mix in the US has averaged a little over 6% real, above the random model's conservative 5%.
The historical trap: the US has been one of the best markets in the world over the last century and a half. Others, like Japan since 1990, fared much worse. Use the historical model to see how your plan copes with real crises, not as a promise of returns.
The withdrawal-rate table
Below the fan chart, the simulator shows the probability of success at withdrawal rates from 3% to 5%, with everything else unchanged and the same simulated futures. A higher rate lowers the target (you get there sooner) but asks more of the portfolio in retirement. With the default values, going from 4% to 3% raises success from 70% to 86%: it is the strongest lever for long retirements (see the 4% rule).
On your real portfolio
In the portfolio, the Your financial independence goal block runs the same simulation starting from the real value of your investments: no need to type your net worth. Volatility and retirement years are saved with the goal, and a link opens the simulator with the same data so you can try the historical model or the table.
What no simulation knows
- Everything is in today's money: the return you enter must be after inflation.
- There are no taxes or fees: if you want a margin, lower the return.
- Spending is fixed, with no state pension or extra income.
- Scenarios use a fixed seed: the same inputs always give the same result, which lets you compare changes without noise.
Educational information, not financial advice. Simulations describe assumptions; they do not predict the future.