What it actually said
William Bengen's 1994 study asked a narrow question: what fixed, inflation-adjusted withdrawal rate would have survived every rolling 30-year retirement in US market history, including the ones that began in 1929 and 1966? The answer was about 4% of the starting portfolio — withdrawn in year one, then the same dollar amount adjusted for inflation each year after, from a portfolio of roughly half stocks, half intermediate bonds. The later Trinity study broadened the portfolios and reached the same neighbourhood.
Note what that is: a historical worst-case survival rate for one country's best century, over exactly 30 years. Not a promise, not a plan, and not a spending rule anyone actually follows — nobody mechanically raises withdrawals through a crash because the formula says so.
Where it holds
As a planning conversion between "annual spending" and "capital required", it remains the most useful single line in the field: spending × 25 equals the 4% number. It is conservative in most historical paths — the median 1994-rule retiree died with more money than they started with — and it prices the right thing: the portfolio that makes work optional, which is a different and more honest goal than "retirement".
Where it lies
Three places. First, the 30-year clock: at longer horizons — early retirement, or simply long life — survival at 4% degrades, which is why the house instrument prints 3% and 3.5% beside it. Second, valuations: starting withdrawals into expensive markets has historically underperformed the average path, and every era believes itself exceptional. Third, the fixed-real-spending assumption is nothing like human behaviour; real people cut spending in bad years, which is precisely why real people do better than the back-test's worst cases.
The house position
Use it as a conversion, not a command. The Work-Optional Number instrument runs your honest spending through 3, 3.5 and 4 percent side by side — the spread between those three columns is a truer picture than any single rate, because it shows how much of the answer is assumption. The 4% figure is the optimistic edge of a sensible range, not the middle of it.
And one line worth keeping: the rule prices spending, so every permanent $1,000 of annual spending you remove deletes roughly $25,000–$33,000 from the capital you need. Frugality compounds at the same rate as returns, with none of the volatility.