Historical backtesting across 150+ years of real market data
Retirement Plan
Personal Details
Leave this blank if you're already retired or retiring immediately — the plan runs exactly as it always has. Fill it in only if retirement is still years away.
Used to determine when Social Security and Pension income begin, if added below.
$
Only used if Current Age is filled in — added each year between now and Retirement Age, growing alongside your balance using the same real historical data as the rest of this tool. Leave blank for none.
Starting Portfolio
$
Asset Allocation
70% / 30%
Backtested using real historical S&P 500 total returns (stocks) and 10-year Treasury total returns (bonds), 1871–2025. 2023–2025 sourced from Aswath Damodaran (NYU Stern) and BLS CPI-U, added to extend past the original dataset's 2022 cutoff.
%
%
50% / 50%
Allocation shifts in a straight line from the starting mix to this mix by the final year of retirement.
Retirement Duration
30 yr
Loading available starting years…
Extra Withdrawals
A one-time large expense — a car, a vacation home, a wedding, or anything else — drawn from the portfolio in addition to your regular withdrawal that year. Specified by year of retirement (e.g. "Year 5"), not a calendar year — since this tool tests the same withdrawal pattern across many different historical starting points at once, an absolute calendar year wouldn't mean the same thing in each one.
Extra Income
A one-time inheritance, home sale, or other windfall that reduces how much you need to draw from the portfolio that year. Same "Year of Retirement" convention as Extra Withdrawals above.
Guaranteed Income
$
Use the "full retirement age" estimate from your Social Security statement, not the age-62 or age-70 figures shown there — this tool applies the early/delayed adjustment itself based on the Start Age below.
Social Security is inflation-indexed — modeled as holding its real purchasing power flat for the entire retirement, consistent with how the rest of this calculator already works in real dollars.
$
Many civilian pensions pay a fixed nominal amount with no cost-of-living adjustment — uncheck this if yours doesn't increase with inflation, and its real purchasing power will erode over time using the same actual historical inflation driving the rest of this simulation.
Retirement Income Tax
%
This is a display-only overlay — it estimates what you'd actually get to spend after federal tax, but does not change the withdrawal amount tested against the portfolio, so Success Rate and every other metric stay based on the actual draw down. Suggested defaults compute your exact 2026 federal bracket tax (IRS Rev. Proc. 2025-32) against your strategy's first-year withdrawal, treated as taxable income after the standard deduction — real returns can differ based on your actual deductions, credits, other income sources, and state tax, which isn't included here.
Medicare Premiums
Most pre-retirees significantly underestimate this cost — Fidelity's 2026 research found 54% incorrectly believe Medicare covers all healthcare expenses, and a Boston College Center for Retirement Research study found roughly two-thirds of pre-retirees underestimate their retirement healthcare spending. Like the tax overlay above, this is display-only and doesn't affect Success Rate. Premiums use the exact 2026 CMS-published IRMAA brackets based on your Filing Status above and each year's approximate income (using total spend as a stand-in, the same simplification the tax estimate uses) — married households are assumed to have both spouses on Medicare. Real Medicare premiums use your actual Modified Adjusted Gross Income from two years prior, which can differ meaningfully from this estimate.
Withdrawal Strategy
4.0%
4.0% of starting balance
4.0%
Recalculated against the current balance every year, not the starting balance. The portfolio never technically reaches exactly $0, but spending falls with it — see Volatile/Small Spending metrics below rather than Success Rate for this strategy.
Withdraws (current balance) ÷ (years remaining) every year. Automatically starts near 1/[duration] and adjusts each year based on actual performance. No manual rate to set.
4.0%
Withdrawal % rises each year as the remaining horizon shortens (an amortization-style calculation), applied to the current balance. This is a simplified version of the Bogleheads VPW method using a single blended assumed return rather than the full published lookup table.
5.0%
%
%
Spending is cut when the current withdrawal rate rises 20% above the initial rate, and raised when it falls 20% below — both disabled in the final 15 years to avoid late-life spending shocks. Based on Guyton & Klinger (2006), Journal of Financial Planning.
%
Withdrawal rate = a + b × (1 ÷ CAPE), recalculated every year using that year's actual market valuation — not fixed at the start of retirement. Ties spending to market valuation: lower when stocks are expensive, higher when cheap. Only usable for historical starting years from 1881 onward (CAPE requires 10 years of trailing earnings data).
%
%
Continuously annuitizes your remaining balance using assumed (not actual historical) return and inflation. These two numbers only affect how much is withdrawn each year — the portfolio itself still grows or shrinks using real historical data, not these assumptions. By design, this fully depletes the portfolio in the final year. Based on a formula published in independent retirement research, backtested across 90 historical cohorts over 60-year retirements.
%
%
The algorithm Yale University uses to determine its endowment's annual spending. Each year's withdrawal blends the prior year's withdrawal with a target rate applied to the current balance, smoothing out market swings. First-year withdrawal is simply the target rate × starting balance.
%
%
Each year, withdraws the larger of (rate × current balance) or (retain % × last year's withdrawal) — so spending only ever drops by at most (100% − retain %) in any single year, even in a sharp downturn, while still fully participating in market upside.
%
%
A fixed base amount (like Constant Dollar) covers essential spending every year regardless of the market. On top of that, whenever the portfolio had a positive real gain last year, an extra amount — the Extras rate times that gain — is added for discretionary spending.
4.0%
Each year's withdrawal blends 75% of last year's withdrawal with 25% of an amortization-style (PMT) calculation against the current balance — smoothing out the swings a pure PMT approach (like VPW) would otherwise produce, at the cost of typically leaving a small balance behind.
%
%
%
Developed by Vanguard's research team as a hybrid of Constant Dollar and Percent of Portfolio. Withdrawal is normally (rate × current balance), but is never allowed to drop more than the floor % or rise more than the ceiling % from last year's withdrawal.
$
$
Monte Carlo
Each simulation resamples a random-length sequence from the same 1871–2025 historical data — actual crash years and calm years, drawn in random order, not an assumed average return.
Scenario Comparison
Save up to 3 full versions of your plan — different strategies, allocations, or spending levels — and compare them side by side using Monte Carlo. Saving captures every setting on this page; switching between scenarios during comparison briefly changes these inputs, then restores exactly what you had before comparing.
Set your portfolio and withdrawal strategy, then run the backtest to see how your plan would have performed across every historical starting year since 1871.