Monte Carlo Simulation for Retirement, Explained
Learn how a Monte Carlo simulation tests a retirement plan against thousands of market paths, what a success rate means and how to raise your odds.
Published
Free tool: Monte Carlo SimulatorA retirement calculator that assumes 7% every year gives one neat answer. Real markets don't move in a straight line, and the order of good and bad years can make or break a plan. A Monte Carlo simulation plays out thousands of possible futures to show the odds instead. This guide explains how, using the free Monte Carlo Simulator.
Why an average return isn't enough
Two retirees can earn the same average return and end up in very different places. If a crash hits in the first years of retirement, withdrawals come out of a shrunken portfolio, and that money isn't there when the market recovers. The same crash 20 years later does much less damage. This is sequence-of-returns risk, and a single average can't show it.
How a Monte Carlo simulation works
- Set up the plan: a starting balance, a yearly withdrawal or contribution, and a number of years.
- Draw a random return for each year, from an average and a volatility you choose, or from actual years of market history.
- Run the plan year by year: take out the withdrawal, then apply that year's return.
- Repeat thousands of times and count how often the money lasts, along with the range of ending balances.
The result is a success rate, such as "the money lasted in 81% of simulations", and a spread of outcomes from bad to good.
Example
A retiree has $1,000,000 and withdraws $40,000 a year, raised with inflation, for 30 years. The portfolio averages 7% a year with 12% volatility, and inflation is 2.5%. In the simulator's default 1,000 runs:
- Chance the money lasts: 80.8%
- Median ending balance: $707,303 in today's dollars
- With a steady 7% every year: $1,128,533, far more than the median, because swings and withdrawals during slumps drag down growth
- Bad case (10th percentile): the money runs out; when it does, it typically lasts about 24 years
How much can you withdraw?
Success rates over 30 years for withdrawals from $1,000,000, raised with inflation, from 10,000 simulations:
| Yearly withdrawal | Rate | Your assumptions (7%, 12% volatility) | US history, 60% stocks |
|---|---|---|---|
| $30,000 | 3% | 95.9% | 98.0% |
| $40,000 | 4% | 81.5% | 90.3% |
| $50,000 | 5% | 57.5% | 75.0% |
| $60,000 | 6% | 33.5% | 53.8% |
The "4% rule" comes from William Bengen's 1994 study, which found that 4% withdrawals, raised with inflation, would have lasted 30 years through every historical period he tested. A simulation shows the risk around it rather than a guarantee.
Stocks or bonds?
With the history option, each simulated year is a random year from 1928 to 2025, using that year's actual S&P 500 and 10-year Treasury returns and inflation. Success rates for $40,000 a year from $1,000,000 over 30 years:
| In stocks | Chance the money lasts |
|---|---|
| 0% | 51.7% |
| 30% | 86.0% |
| 60% | 90.3% |
| 80% | 89.2% |
| 100% | 86.7% |
All bonds fails often because bonds barely beat inflation over long periods. All stocks has the highest median but the deepest crashes. A mix in between gives the best odds here.
Limits of a simulation
- It's only as good as its inputs. A higher assumed return raises the odds whether or not markets deliver it.
- Random years lose their order. Long stretches of bad years, like the 1970s, don't repeat as they happened.
- Real people adapt. Most retirees spend less after a bad year, which raises the real-world odds.
- Taxes and fees aren't included unless you lower the return to allow for them.
To see the guaranteed math of a single scenario, use the Retirement Calculator; for a deeper look at withdrawal planning, see How Much Do I Need to Retire?.
How to use the Monte Carlo Simulator
- Open the Monte Carlo Simulator.
- Choose to withdraw or save each year, and enter the starting balance, yearly amount and number of years.
- Pick the market model: your own average return, volatility and inflation, or US history with a stock and bond mix.
- Choose 500 to 10,000 simulations, whether to raise the amount with inflation and whether to show today's dollars. "Run a new set" draws fresh random years.
- Read the results: the chance of success, the median, 10th and 90th percentile balances, a fan chart of the range over time, the chance the money lasts each year and a chart of ending balances.
Frequently asked questions
What success rate should I aim for?
Many planners aim for 80% to 90%. A plan near 100% is very safe but may mean spending less than you could.
Why do results change when I run a new set?
Each set uses different random years. With 1,000 simulations the success rate can move a point or two; 10,000 simulations give steadier results.
Should I use today's dollars?
Usually, yes. Balances in today's dollars show what the money would buy, which is easier to compare with your spending.