How to use the withdrawal plan explorer
The explorer lets you try a retirement spending plan against every 30-year retirement in U.S. history since 1928, and against 5,000 more built from that history, and see how spending would have gone year by year.
Quick start
- Start with the recommended plan. The page opens on it. The tiles at the top summarize how it did.
- Change one setting at a time. Move a slider in the Settings panel; everything recalculates instantly. Values you have changed from the recommended plan turn blue.
- Read the difference. Every tile shows how your plan compares with the recommended one, in green when it is better and red when it is worse. A red banner appears if more than 2% of reshuffled retirements run out of money.
- Look at the charts. They show spending for the typical and unlucky retiree, how deep the cuts were, which start years were hardest, and any single retiree you pick.
- Unsure what a setting does? Hover over its (?) icon for a moment to peek, or tap or click it to pin the explanation open. Close it with ×, the Escape key or a click elsewhere.
- Share or start over. "Copy a link to these settings" makes a link that reopens your exact scenario. "Reset" goes back to the recommended plan. Presets load other complete plans for comparison.
All amounts are in today's dollars per $100 saved. Multiply by 10,000 for a $1 million portfolio: $6.50 means $65,000 a year.
How the spending rule works
In year one you spend the starting withdrawal rate times your portfolio. Every January after that, four steps set the year's spending, all after inflation:
1. PlanMove last year's spending along the planned path (by default, retirees' observed pattern of
about −1% a year).
2. CheckDivide planned spending by today's portfolio. If that rate is far above the starting rate,
cut; far below, raise. In between, change nothing.
3. CapNever spend more than the cap above the planned path.
4. FloorNever spend less than the floor, a fixed share of year-one spending.
To run the recommended plan yourself each January, with the planned change for every age and two worked retirements, see Following the plan, year by year.
The portfolio is rebalanced to your stock and bond mix each year, and the year's spending is withdrawn at year end. A retirement runs out if the portfolio cannot pay the year's spending; after that, spending from the portfolio is zero.
Reading the results
Tiles
| Tile | What it tells you |
|---|
Charts
| Chart | What it shows |
|---|---|
| Spending over 30 years | Each year's spending across the 69 historical retirements: the middle retiree (thick blue line), the range from the 10th to the 90th percentile (shaded), the unluckiest retiree (dashed red), and the plan, cap and floor. The grey dashed line is the recommended plan's middle retiree, for comparison. |
| Deepest spending cut | Each retiree is placed by the deepest cut they lived through, measured two ways: below the plan, and below their own best year so far. Bars for your plan sit above the recommended plan's. |
| Lowest yearly spending by start year | The leanest year for each retirement start from 1928 to 1996. Retirements that ran out are marked ×. Starts in 1929 and the mid-1960s are usually the hardest. |
| One retiree | Pick a start year to see that retiree's spending each year, with the plan, cap and floor, and their portfolio balance underneath. |
Hover over any chart for exact values. Use the chart's toolbar to zoom or download it as an image.
Every setting explained
The same explanations appear when you hover over or tap a setting's (?) icon.
Key ideas
- Real (today's) dollars
- All amounts are adjusted for inflation, so $65,000 in year 20 buys what $65,000 buys today.
- Withdrawal rate
- Spending divided by the portfolio. The plan compares the current withdrawal rate with the starting rate to decide whether to cut or raise.
- The 69 historical retirements
- One 30-year retirement starting each January from 1928 to 1996, using actual returns and inflation. They overlap heavily, so they contain only a few truly different bad stretches.
- Reshuffled history
- 5,000 extra 30-year retirements, each stitched together from runs of real consecutive years (10 years long on average) picked at random from 1928–2025. Stocks, bonds and inflation from the same year stay together. They test how a plan does on sequences of good and bad years that history could have produced but didn't. A plan tuned only on real history can look much safer than it is; see the study.
- Smooth-spending score
- The steady yearly amount a retiree would value as much as the plan's actual ups and downs. It is higher for plans that spend more, more steadily and earlier, and much lower for plans with lean years or that run out. Use it to compare plans, not as a spending figure.
- Typical
- The middle (median) retiree: half did better, half did worse.
What the explorer leaves out
- Fees and taxes. None are charged. A 0.25% yearly fee raises the recommended plan's run-out rate from about 2% to about 3% in reshuffled history.
- Other income. Social Security, pensions and annuities are not included. They make any portfolio plan safer.
- Lifespan. Every retirement lasts 30 years. Living longer is covered only by the money left at the end.
- The future. Results come from 1928–2025 U.S. and developed-market returns. The future may be better or worse than anything in that record.
- Your situation. The explorer is an educational tool, not financial advice. Talk to a qualified adviser before making decisions.
Tips
- Compare against a preset. Load the earlier plan or Bengen 4% to see why the recommended plan was chosen.
- Watch two tiles together. "Typical spending, years 1–10" shows the reward; "Ran out, reshuffled history" shows the risk.
- Raise the floor carefully. A higher floor feels safer but is the most common way to make a plan run out.
- On a phone, results come first and the settings follow; use "Change the settings" at the top to jump down.
- Keyboard: Tab to a (?) icon and press Enter to open its help; Escape closes it.