Every retirement plan is built around one fear: running out. But when I replayed a conservative plan across a century of market history, the common outcome wasn't depletion. It was dying with many times more than you started. The real risk for careful savers may be the opposite of the one they're planning for.
Ask anyone planning an early retirement what keeps them up at night and the answer is the same: outliving the money. It's the right thing to respect, and it's why this calculator uses a conservative rate and a cash buffer. But respecting a risk and correctly sizing it are different things, and the backtest tells a story that almost nobody plans around. At a conservative withdrawal rate, the typical retirement doesn't squeak by. It ends with a fortune.
On the 50-year backtest, a 3.5% inflation-adjusted withdrawal survived 47 of 48 historical windows on its own, and all 48 with a small cash buffer. "Survived" is the headline, but it undersells what happened. Here's where those retirements actually ended, in multiples of the starting portfolio, after fifty years of withdrawals:
| Conservative plan (3.5%, 50 yrs) | Median ending | Worst 10% ending | Windows survived |
|---|---|---|---|
| No cash buffer | 8.8× | 1.15× | 47/48 |
| 2-year buffer | 7.9× | 1.32× | 48/48 |
| 3-year buffer | 7.5× | 1.06× | 48/48 |
Endings are inflation-adjusted multiples of starting wealth after 50 years of 3.5% withdrawals. From the cash buffer study. "Worst 10% ending" is the 10th-percentile outcome.
Read the median column again. A retiree who started with $1 million and drew 3.5% for half a century ended, in the typical historical case, with the inflation-adjusted equivalent of nearly $8 million, after a lifetime of spending. That's not the exception. That's the middle of the distribution.
The more striking column is the worst 10%. Even the unluckiest tenth of historical retirements, the ones that started right before a crash, ended at or above where they began — 1.32× with a two-year buffer. Only a single start in the entire record, 1929, depleted a no-buffer portfolio at all, which is precisely why the calculator keeps a buffer in the first place. The asymmetry is the whole point: the downside of a conservative plan was, with one historical exception, ending roughly where you started, while the upside was ending with eight times more. People plan as if those odds were reversed.
This is the quiet problem Bill Perkins named in Die With Zero: the goal was never to accumulate the largest possible pile and then stop existing on top of it. A median outcome of 8× starting wealth is, viewed honestly, a planning failure of a different kind — years of foregone experiences, trips not taken, generosity deferred, all to guard against a tail that mostly didn't arrive. The richest person in the graveyard didn't win. They over-saved against a fear that, for conservative plans, the data simply doesn't support at the scale people feel it.
It's not irrational, it's structural. You only retire once, into one specific sequence of markets, and you can't know in advance whether you'll get the median or the 1929 start. To survive the worst case you have to spend as if you might get it, which means that in every other case, the overwhelming majority, you systematically underspend. Conservative withdrawal math guarantees that the typical retiree dies with a large surplus, because the rate is set by the worst sequence, not the median one. The safety is real. So is its price.
Not "spend recklessly." The tail is real and one historical start did fail. But the data argues for confronting the trade instead of defaulting to maximum caution. A few honest levers: spend more in your early, healthier years when experiences are worth most; consider a slightly higher withdrawal rate with eyes open about what it costs in worst-case safety; or front-load discretionary spending and taper later. There's no free lunch — a higher rate does raise the odds of the bad tail, as the 4% rule page shows over long horizons. The point is to choose deliberately. The calculator helps by showing your whole range of outcomes across every market since 1928, not just a single average, so you can see whether you're on track to run dry or, far more likely, to die with millions you meant to enjoy.
See your full range of outcomesThe 50-year backtest has the full survival data, the cash buffer study covers the endings by buffer size, and the safe withdrawal rate page explains how the rate is chosen in the first place.
Not financial advice. Consult a fee-only fiduciary CFP for personalized guidance. Backtest results are historical and don't guarantee future outcomes.