Retiring Early Is Arithmetic. Staying Retired Is Engineering.

Strip every early-retirement book down to its load-bearing wall and you find one equation — and it isn’t about investment returns. Your savings rate does two jobs at once. It builds the pile. And — the half everyone misses — it shrinks the life the pile has to support. Save 10% and you need 90% of your income replaced forever. Save half and you only need half replaced — while building the pile five times as fast. The rate attacks the problem from both ends. Nothing else in the book does that.

Run it at a 5% real return, standard 25-times-spending target, starting from zero: 10% saved takes roughly fifty years. Twenty percent, about thirty-seven. A third, twenty-eight. Half, seventeen. Push to 60% and it’s twelve. That table is the entire genre in four lines — every bestseller on the shelf is footnotes to it — and look what’s not in it: your fund picks, your market view, your timing. At these horizons a brilliant portfolio moves the date a couple of years. The savings rate moves it by decades.

It’s also why the movement’s fans and critics keep talking past each other. The critics say “nobody can save half their income,” and for plenty of incomes that’s just true — this game is won on the big pipes, housing and cars, never the lattes. The fans reply that even 10%-to-25% pulls the date in by fifteen years. Also true. The equation doesn’t care which camp you join. It pays out per percentage point, starting with the first one.

The number powering every early-retirement spreadsheet — save 25 times your spending, draw 4% a year, never run out — has a birth certificate. In 1994 an advisor named William Bengen tested every retirement start-year in U.S. history and found the worst-case survivor: 4%, inflation-adjusted, survived every thirty-year window on record, including the ones that started in 1929 and 1966. Honest, important work. The early-retirement movement then borrowed it with one quiet edit: they kept the 4% and threw out the thirty years.

Retire at 45 and you need fifty years, and at fifty years the rule’s own math softens — more historical windows fail, and the safe number drifts toward 3.5% or under. Sounds like a footnote until you price it: at 4%, freedom costs 25× your spending. At 3.5%, about 29×. At 3%, 33×. The gaps between those multiples are measured in years of additional working. That’s the fine print, invoiced.

The second clause actually ends retirements: sequence risk. Two retirees earn identical thirty-year average returns. The one who catches a deep bear in years one through three — while withdrawing — can fail; the same crash in years twenty-five through twenty-seven is a story at dinner. Withdrawals turn temporary drawdowns into permanent ones. Readers of the leverage letter will recognize the shape: spending from a falling portfolio is the margin call you issue to yourself. The defenses are boring and they work — a year or two of cash, a written rule for what gets trimmed in a bad year, any part-time income at all. Each moves the survival odds more than another point of return.

Here’s the part the books save for chapter eleven, which is backwards, because it’s the part that needs actual engineering. American retirement has two locked gates — Medicare at 65, penalty-free retirement accounts at 59½ — and both assume you waited. Quit at 47 and you’re bridging up to eighteen years of healthcare and twelve of income, legally, cheaply, without dismantling the compounding you spent two decades building.

The healthcare bridge is the one that scares people, and the honest news is mixed. Marketplace coverage is real and often cheaper than feared — subsidies key off income, not assets, and a retiree living on savings can show a modest taxable income. But it’s also the most Washington-exposed line in the whole plan; the rules move with every Congress. So we engineer around the rules: a dedicated healthcare reserve, budgeted at full unsubsidized prices, treated as untouchable. If a subsidy shows up, it’s gravy. It’s never a load-bearing wall.

The income bridge has three standard spans, best combined. The plain taxable brokerage account — no age gates, friendly capital-gains rates, and chronically underfunded because it doesn’t come with a tax sticker. The Roth conversion ladder — convert a slice of the old 401(k) each year, pay the tax in a deliberately low-income year, wait the required five, withdraw penalty-free: a conveyor belt of accessible money, built five years ahead of need. And 72(t) payments — rigid, unforgiving of errors, occasionally exactly right. Underneath all three, one discipline: the bridge gets designed years before the resignation letter, because the ladder needs a five-year head start and the taxable account needs a decade of feeding.