Monte Carlo retirement calculator
By Nicolas StrautUse the Monte Carlo retirement calculator to see how often your savings last through retirement when markets rise and fall at random. It runs 10,000 simulated retirements from your balance, spending, and stock and bond mix, and shows the return and volatility behind every one.
It's free, needs no account, and your inputs stay in your browser.
Every invested retirement and taxable balance, combined
Leave at 0 if you've already retired
In today's dollars, raised with inflation every year
Social Security or a pension, in today's dollars
Bengen's study tested at least 30 years
The rest is 10-year Treasury bonds, rebalanced every year. This is an input to test, not a suggestion for your portfolio.
Before inflation. The 60% stock average for 1928 to 2025 is 9.04%
The standard deviation of yearly returns. Historical for this mix is 12.12%
The 1928 to 2025 average is 3.02%
Moves each simulated retirement's 2 worst years to retirement years 1 and 2, with the same returns
Savings that lasted
92%
| Median ending balance | $1,587,791 |
| 10th percentile ending balance | $101,677 |
| 90th percentile ending balance | $5,111,727 |
| First-year withdrawal rate | 4% |
92% of 10,000 simulated retirements still had money after 30 years. Ending balances are in today's dollars. This is an estimate, not financial advice.
In the trials that ran out, the money usually lasted 25 years. Your first-year withdrawal rate is 4% of your balance at retirement, vs a 4% start under the 4% rule.
These results come from 10,000 random sets of yearly returns. Each year's return is drawn around an average of 9.04% with a typical swing of 12.12%, the average and swing of a 60% stock, 40% bond mix from 1928 to 2025. Prices rise 3.02% a year, so the average return after inflation is 5.84%. Real returns can be lower than the past, so read this as a range, not a forecast.
- 90th percentile
- Median
- 10th percentile
| Year | 10th percentile | Median | 90th percentile |
|---|---|---|---|
| 5 | $730,379 | $1,061,574 | $1,492,760 |
| 10 | $623,713 | $1,126,621 | $1,893,049 |
| 15 | $516,304 | $1,198,782 | $2,398,415 |
| 20 | $398,583 | $1,297,484 | $3,029,713 |
| 25 | $264,262 | $1,428,212 | $3,912,997 |
| 30 | $101,677 | $1,587,791 | $5,111,727 |
Your numbers never leave your browser.
Figures from NYU Stern historical returns, the Minneapolis Fed CPI table, and the 2026 Social Security Trustees Report.
What this result doesn't show
- The simulation spends from one combined balance and leaves out taxes, so a pre-tax 401(k) or IRA balance looks larger than what you can spend.
- It doesn't show which account to draw from each year, or the required distributions that start in your 70s.
- It doesn't show when to claim Social Security, which changes how much the portfolio has to cover.
The retirement withdrawal plan works out the account order, the Roth conversions and each person's filing age year by year, for $79 once.
30-day money-back guarantee. If the plan isn't useful to you, reply to your receipt within 30 days for a full refund. You don't need to give a reason.
Key takeaways
- 92% of 10,000 simulated retirements lasted 30 years when $1,000,000 paid out $40,000 a year, rising with inflation, in a 60% stock mix.
- With each retirement's two worst years moved to the start, the same returns lasted in 74% of the trials.
- The defaults are the 1928 to 2025 average and swing of S&P 500 and 10-year Treasury returns, and every figure on the page can be changed.
How this is calculated
Each of 10,000 trials draws a return for every year from a normal distribution around your expected return and volatility. Spending, less guaranteed income, comes out at the start of each year and rises with inflation. The success rate is the share of trials that never run out.
In this article
- How to use the Monte Carlo retirement calculator
- What your Monte Carlo results mean
- Understanding Monte Carlo simulations
- Where the return and volatility come from
- What is a good Monte Carlo success rate?
- How to test sequence of returns risk
- Example of retiring with $1,000,000 and spending $40,000 a year
- How accurate is a Monte Carlo retirement calculator?
- Related calculators
- Frequently asked questions about Monte Carlo retirement calculators
Key takeaways
- 92% of 10,000 simulated retirements lasted 30 years when $1,000,000 paid out $40,000 a year, rising with inflation, in a 60% stock mix.
- With each retirement's two worst years moved to the start, the same returns lasted in 74% of the trials.
- The defaults are the 1928 to 2025 average and swing of S&P 500 and 10-year Treasury returns, and every figure on the page can be changed.
How this is calculated
Each of 10,000 trials draws a return for every year from a normal distribution around your expected return and volatility. Spending, less guaranteed income, comes out at the start of each year and rises with inflation. The success rate is the share of trials that never run out.
How to use the Monte Carlo retirement calculator
Work through the inputs in this order, then change one at a time to see what moves the result.
1. Enter your savings and when you retire
Combine every invested balance: workplace plans, IRAs, and taxable brokerage. Leave out the emergency fund and your home. If you're still working, enter the years until you retire and what you'll save each year until then. Leave both at 0 if you've already retired.
2. Enter your spending and guaranteed income
Spending is a yearly figure in today's dollars, and the calculator raises it with inflation every year. Guaranteed income is Social Security or a pension. The preset fills in the 2026 average retirement benefit, $24,852 a year,1 so swap in your own estimate once you have it. The Tweed Social Security claiming plan works out a claiming age.
3. Choose a stock share
The rest of the portfolio is 10-year Treasury bonds, rebalanced every year. Changing the share refills the return and volatility with that mix's history. The 60% default is a starting input, not a suggestion for your portfolio.
4. Check the return, volatility, and inflation
Volatility is how far a typical year lands from the average. Higher volatility means a wider range of outcomes at the same average. Lower the return to test a weaker future than the past, and the reset button restores the historical figures.
What your Monte Carlo results mean
Success rate
The large figure is the share of the 10,000 trials that still had money at the end of the retirement years you entered.
With the two worst years first
Shown when Bad years first is on. It runs the same trials with each one's two lowest years moved to the start, and the gap from the success rate is the cost of a bad start.
Ending balance
The 10th percentile is the result 9 in 10 trials beat, the median is the middle trial, and the 90th percentile is the result only 1 in 10 beat. All three are in today's dollars.
When the money runs out
In the trials that failed, the typical number of years the money lasted.
First-year withdrawal rate
Spending minus guaranteed income, divided by your balance at retirement. The calculator shows it next to the 4% starting rate of the 4% rule.
Understanding Monte Carlo simulations
What is a Monte Carlo simulation for retirement?
A Monte Carlo simulation for retirement runs your plan through thousands of random market paths and counts how many of them leave you with money at the end. Every path uses the same average return and the same volatility, and only the luck of the draw changes. Some paths start with a crash and some with a boom, which is the range a single average return hides.
Monte Carlo vs a regular retirement calculator
A regular retirement calculator grows your savings at one steady return, so it gives one answer. A Monte Carlo calculator varies the return every year, so it gives a range and a probability. The Tweed retirement calculator answers how much you'll have and whether it meets a 25x target, while this one answers how often a balance lasts. Use both.
Where the return and volatility come from
The default return and volatility are the average and the typical swing of yearly returns for each stock and bond mix from 1928 to 2025, from the S&P 500 with dividends and 10-year Treasury bonds.2
| Stock share | Average yearly return | Typical yearly swing | Worst calendar year |
|---|---|---|---|
| 100% | 11.86% | 19.40% | -43.84% (1931) |
| 80% | 10.45% | 15.63% | -35.58% (1931) |
| 60% | 9.04% | 12.12% | -27.33% (1931) |
| 40% | 7.64% | 9.17% | -19.07% (1931) |
| 20% | 6.23% | 7.48% | -17.87% (2022) |
| 0% | 4.82% | 7.90% | -17.83% (2022) |
The all-bond row swings more than the 20% stock row because stocks and bonds often moved in different directions in the same year.
The default inflation rate, 3.02%, is the average yearly rise in consumer prices over the same 98 years.3 The Social Security Trustees assume 2.4% a year over the long run in their 2026 report, and the preset button uses that figure.4
These are past results, and no agency publishes a forecast of stock returns, so treat the defaults as a starting point and test lower ones.
What is a good Monte Carlo success rate?
A good Monte Carlo success rate is one you're comfortable with given how much you could cut spending in a bad stretch, and there's no official threshold. The table runs $1,000,000 in a 60% stock mix for 30 years at the historical defaults.
| First-year withdrawal | Success rate | With the two worst years first |
|---|---|---|
| 3% ($30,000) | 98.7% | 94.5% |
| 3.5% ($35,000) | 96.4% | 86.2% |
| 4% ($40,000) | 92.3% | 73.6% |
| 4.5% ($45,000) | 85.3% | 58.7% |
| 5% ($50,000) | 76.7% | 42.9% |
Bengen's 1994 study found that a 4% first-year withdrawal, raised for inflation, lasted at least 33 years in every US starting year he tested.5 A simulation can produce worse paths than history did, which is why a 4% rate shows less than 100% here. The guide to the 4% rule covers where that rate comes from.
A lower success rate isn't a prediction you'll run out. It's the share of random paths where spending stayed fixed no matter what markets did.
How to test sequence of returns risk
To test sequence of returns risk, turn on Bad years first. The calculator moves each simulated retirement's two worst years to the start and runs it again with the same returns.
A short stretch at the start does the most damage, because you're selling into it. Two years matches the 1973 to 1974 decline that Bengen found the most damaging.
The drop in success rate is the cost of a bad start at your spending level. In the table above it's 4.2 points at a 3% withdrawal and 33.8 points at 5%.
The guide to sequence of returns risk shows the same effect in real market history, with two retirees and one set of returns.
Example of retiring with $1,000,000 and spending $40,000 a year
These are the calculator's default inputs, so the result matches what you see above before you change anything.
| Input | Value |
|---|---|
| Current savings | $1,000,000 |
| Years until you retire | 0 |
| Yearly spending | $40,000 |
| Guaranteed income | $0 |
| Years in retirement | 30 |
| Stock share | 60% |
| Expected return, volatility, and inflation | 9.04%, 12.12%, 3.02% |
| Result | Value |
|---|---|
| Success rate | 92.3% |
| With the two worst years first | 73.6% |
| 10th percentile ending balance | $101,677 |
| Median ending balance | $1,587,791 |
| 90th percentile ending balance | $5,111,727 |
The median trial ends with more than it started, in today's dollars, and about 1 in 13 trials runs out. The two-worst-years-first run shows how much of that risk sits in the first two years.
How accurate is a Monte Carlo retirement calculator?
A Monte Carlo retirement calculator is as accurate as its assumptions, so it measures how fragile a plan is under those assumptions, not what will happen. Five assumptions shape every result:
- Returns follow a bell curve, so very bad years are a little rarer than in real history.
- Each year is drawn on its own, so the simulation has no long bear markets or booms built in, beyond what chance produces.
- Inflation is one steady rate, so it leaves out a 1970s-style inflation spike.
- Spending stays fixed in today's dollars, though most retirees cut back after a bad year.
- Taxes, fees, and required minimum distributions are left out, so results run optimistic for a pre-tax balance.
So run it at a return a point or two lower than the default, and see whether the plan still holds.
Taxes are the piece this calculator leaves out. The Tweed retirement withdrawal plan adds them, because it picks which account you spend from each year and how much you convert to Roth before RMDs start. It also sets your Social Security filing age.
Year one free
See the tax your withdrawal order costs
Frequently asked questions about Monte Carlo retirement calculators
What is the best Monte Carlo retirement calculator?
The best Monte Carlo retirement calculator depends on how much of your plan you want to model. A free calculator like this one tests a balance and a spending level. Paid planners such as Boldin and ProjectionLab add taxes, Social Security timing, and Roth conversions, and Tweed reviews both.
How many simulations does a Monte Carlo retirement calculator need?
A Monte Carlo retirement calculator needs enough simulations for the success rate to stop moving when you run it again. Run 20 times on this page's default inputs, 1,000 trials gave success rates from 91.2% to 94.4%, while 10,000 trials stayed between 91.9% and 92.9%. That is why this calculator runs 10,000 every time.
Does a Monte Carlo simulation include Social Security?
A Monte Carlo simulation includes Social Security only if you enter it. In this calculator, the guaranteed income field is taken off your spending each year before anything is drawn from savings. The benefit rises with inflation and does not move with markets, so a larger benefit lowers how much of each bad year your savings have to absorb.
What is the difference between Monte Carlo and historical backtesting?
The difference between Monte Carlo and historical backtesting is where the returns come from. Monte Carlo draws random returns around an average, so it can produce paths that never happened. Historical backtesting replays real past years in order, so it is limited to about a century of starting points. Some paid planners offer both.
What should I do if my Monte Carlo success rate is low?
If your Monte Carlo success rate is low, change one input at a time to see which one moves it most. Spending less, retiring a few years later, or adding guaranteed income usually moves it further than changing the stock share. Then check the gap with Bad years first turned on.
Personal finance writer, former Forbes contributor and This Week in Fintech writer. He isn't a financial advisor or a CPA.
Sources
- https://www.ssa.gov/news/en/cola/factsheets/2026.html
- https://pages.stern.nyu.edu/~adamodar/New_Home_Page/datafile/histretSP.html
- https://www.minneapolisfed.org/about-us/monetary-policy/inflation-calculator/consumer-price-index-1913-
- https://www.ssa.gov/oact/TR/2026/V_B_econ.html
- https://www.financialplanningassociation.org/sites/default/files/2021-04/MAR04%20Determining%20Withdrawal%20Rates%20Using%20Historical%20Data.pdf