Test how long your savings could last across 1,000 simulated market paths. Change your monthly spending, retirement length or stocks, bonds and cash mix, then compare the outcomes.
Start with your portfolio, spending and time horizon. This illustration uses US dollars and reference stocks, bonds and cash assumptions. Run it, then change one input to see what matters.
$24,000/yr (4.8% rate)
Configure your retirement setup above and click Run Simulation.
ILLUSTRATION · NOT YOUR RETIREMENT RESULT
It means 95% of the simulated paths met the model’s success definition. It is not a guarantee that your real retirement has a 95% chance of success. There are two separate uncertainties: sampling only a finite number of paths, and whether the model assumptions describe the future.
950 of 1,000 hypothetical paths succeed.
93.5% to 96.2%
Approximate 95% Wilson interval for the fixed model’s success probability, reflecting sampling error only.
More paths narrow this sampling interval. They do not correct an overly optimistic return forecast, missing expenses, an unsuitable time horizon or a poor model of market falls. Even 100% simulated success is not certainty.
For planning, also change the spending, time horizon and investment assumptions and inspect the failed paths. A small shortfall near the end and an early large shortfall can share the same success score but mean very different things.
This illustration treats paths as independent yes/no trials from one unchanged model. The interval is not suitable without adjustment for dependent paths. Its repeated-sampling coverage concerns that model, not whether your finances will last.
NIST: Wilson intervals for proportions. Fidelity: planning assumptions and limitations.
The tables below come from the same engine as the calculator above, with the same settings: 1,000 lognormal paths with fat-tail adjustment, seed 42, a $500,000 starting portfolio, annual rebalancing, a 0.5% management fee, and withdrawals that escalate with the simulated inflation path (centered on 2.5% a year). Enter matching inputs and the calculator reproduces every cell.
| 30-year survival | 3% | 3.5% | 4% | 4.5% | 5% |
|---|---|---|---|---|---|
| Conservative (30/60/10) | 99.0% | 94.9% | 77.8% | 53.9% | 29.1% |
| Balanced (60/30/10) | 96.7% | 89.7% | 77.6% | 63.4% | 48.7% |
| Aggressive (80/15/5) | 94.0% | 86.8% | 76.1% | 65.1% | 53.6% |
Withdrawal rates as monthly dollars on $500,000: 3% = $1,250, 3.5% = $1,458, 4% = $1,667, 4.5% = $1,875, 5% = $2,083.
At a 4% withdrawal rate the three mixes land within about a point of each other, near 78%. Allocation matters at the edges: at 3% a year the conservative mix survives most often (99.0%), while at 5% the aggressive mix does (53.6% against the conservative mix's 29.1%), because once withdrawals outrun bond yields only equity growth can keep pace.
| 4% rule survival | 20 yrs | 25 yrs | 30 yrs | 35 yrs | 40 yrs |
|---|---|---|---|---|---|
| Conservative (30/60/10) | 99.5% | 94.2% | 77.8% | 55.5% | 40.2% |
| Balanced (60/30/10) | 98.1% | 90.1% | 77.6% | 65.0% | 59.1% |
| Aggressive (80/15/5) | 94.9% | 86.7% | 76.1% | 66.4% | 62.5% |
Longer horizons separate the mixes the same way higher withdrawal rates do: past 30 years the equity-heavy allocations hold up better because they have more growth to outlast the extra years of withdrawals.
Survival means the portfolio never hits zero before the horizon ends; finishing with the starting balance intact is a stricter test. Figures computed August 2026 on the default assumptions with the calculator's stocks, bonds, and cash universe; they move when the assumptions do. Our forward-looking withdrawal analysis, explains the model, its assumptions and its limitations in more detail.
Instead of assuming one fixed return every year, a Monte Carlo simulation generates thousands of possible market futures by drawing random returns from a statistical distribution, then measures how often your retirement plan survives across all of them. The output is the share of paths that meet the model’s success definition, not a guarantee about the real world.
Yes. The simulator runs 1,000 paths entirely in your browser at no cost, with no signup required. A free Portfolio Lab account adds the full platform around it (the 57-asset optimizer, saved portfolios, and PDF reports), but the calculator itself is free to everyone.
A success rate is conditional on the model inputs and success definition. This standalone calculator uses its reference USD stocks, bonds and cash assumptions, with a lognormal model and fat-tail adjustment. That does not prove that market extremes or future returns are accurately represented. The full workstation lets you inspect and change asset assumptions. Test spending and horizon changes as well as investment changes.
More paths reduce random sampling noise, not model error. With 1,000 independent paths, uncertainty depends on the success rate: it is wider near 50% than near 95%. The illustration below shows a Wilson interval for sampling error alone. It does not put a confidence interval around your real retirement outcome.
Between 76.1% and 77.8% over 30 years, whether the mix is conservative, balanced, or aggressive. That is for a $500,000 portfolio with the 4% rule's $1,667 a month escalating with simulated inflation (centered on 2.5% and correlated with returns), a 0.5% management fee, and J.P. Morgan 2026 assumptions. The original Trinity Study, tested on 1926-1995 US returns, put the 4% rule at 95% for a 50/50 portfolio. That historical study used different assets, periods and methods, so the difference cannot be attributed to return forecasts alone.
This tests your spending and horizon using reference investment assumptions. The workstation adds editable assumptions and guardrails spending strategies.
Open the workstation →