Every calculator on this site runs on one shared engine, so the same inputs give the same answer whichever page you are on. The engine is about a hundred lines of arithmetic, not a black box, and this page states all of it: the formulas, the compounding convention, the defaults each form ships with, the tax year constants quoted in the guides, and the research the 4 per cent withdrawal rate rests on. It also states what the method deliberately leaves out, which matters more than the parts it includes.
What formula does each calculator use?
Five formulas cover the whole site. Everything else is presentation.
FIRE number = annual spending ÷ withdrawal rate
Coast FIRE number = FIRE number ÷ (1 + real return)years until retirement
Barista FIRE number = (annual spending − part-time income) ÷ withdrawal rate
Savings rate = (take-home income − spending) ÷ take-home income
real return = nominal return − inflation − fees
Lean FIRE and Fat FIRE are the FIRE number formula with a different spending figure in the numerator, not a different calculation. The Lean FIRE calculator ships with $25,000 of annual spending and the Fat FIRE calculator with $100,000, against $40,000 on the FIRE number calculator. The 4% rule calculator runs the FIRE number formula in both directions at once: portfolio multiplied by the rate gives the income, income divided by the rate gives the portfolio. The early retirement calculator does not add a formula either. It projects your balance forward and reports the age at which it first crosses the FIRE number.
Does the projection compound monthly or annually?
Monthly, and the two conventions agree by construction. The engine converts the annual real return into a monthly rate of (1 + real) raised to the power of one twelfth, minus 1. Each month it multiplies the balance by that rate and then adds your monthly contribution, so contributions land at the end of the month rather than the start, and twelve monthly steps come to exactly one year of the annual real return. The chart plots one point a year from that monthly series.
The Coast FIRE discount is applied annually rather than monthly, because it is a single division rather than a path: your FIRE number divided by (1 + real return) raised to the number of years left. Since the monthly and annual conventions agree over a whole year, the two sit together without drift.
The real return itself is a subtraction: nominal return minus inflation minus fees. At the shared defaults of 7 per cent, 3 per cent and 0 per cent that is exactly 4.00 per cent. The exact Fisher form, 1.07 divided by 1.03 minus 1, is 3.88 per cent, so the subtraction is roughly 0.12 percentage points optimistic at these levels and more optimistic as inflation rises. That is a real simplification and it is stated here rather than buried: it is kept because it is the convention almost every FIRE spreadsheet uses, and because you can see where the number came from. Every figure the calculators display is in today's money for exactly this reason.
What are the default assumptions?
These are the values the shared form ships with. All of them are editable, and the four assumption sliders sit under the Advanced assumptions toggle on each calculator.
| Input | Default | Notes |
|---|---|---|
| Current age | 30 | Shared across the core calculators |
| Retirement age | 65 | Gives a 35 year horizon at the default age |
| Annual spending in retirement | $40,000 | $25,000 on Lean FIRE, $100,000 on Fat FIRE |
| Currently invested | $120,000 | Only affects progress and the crossing age, never the target |
| Monthly contribution | $0 | $1,000 on every calculator except Coast FIRE, which starts at zero on purpose |
| Annual part-time income | $20,000 | Barista FIRE only |
| Investment return, nominal | 7% | Your assumption, not a forecast |
| Inflation | 3% | Subtracted from the return |
| Investment fees | 0% | Also subtracted; set it to your fund's real cost |
| Safe withdrawal rate | 4% | See the sources section below |
Run the site at those untouched defaults and you get a FIRE number of $1,000,000 and a Coast FIRE number of $253,415. The Lean FIRE pair is $625,000 and $158,385, the Fat FIRE pair is $2,500,000 and $633,539, and the Barista FIRE number is $500,000, exactly half of full FIRE because the part-time income covers half the spending. Those five pairs are the fastest way to check that a page is telling you the truth. If a figure on this site does not reconcile with the formulas above, it is an error and we want to hear about it.
Why does Coast FIRE discount at 4 per cent real?
Because the discount rate and the growth rate have to be the same number, or the calculator contradicts itself. Coast FIRE asks what balance today would grow into your FIRE number by retirement with nothing added. That is the projection run backwards, so it uses the same real return the projection uses forwards: 7 per cent nominal minus 3 per cent inflation minus 0 per cent fees, which is 4 per cent real. Change the return slider and the Coast number moves immediately, which is the point of exposing the slider at all. The level matters far less than the consistency: a plan built on 4 per cent real that is checked against 3 per cent real is worth more than a plan built on one number and never stress-tested. The Coast FIRE by age grid is the same discount applied at seven starting ages, which is why each five year step raises the number by roughly a fifth.
Why does the savings rate calculator use 5 per cent real?
The savings rate calculator is the one tool on the site whose real return slider does not start at 4 per cent. It starts at 5 per cent, and that is a deliberate convention rather than an oversight. That tool is not sizing a portfolio, it is showing the shape of the relationship between the share of income you keep and the years it takes to stop needing income, and it keeps its own slider so you can compare it against the version of that table you have already seen elsewhere.
Here is exactly what the choice costs you. At the shipped defaults of $60,000 take-home, $42,000 of spending and nothing invested yet, the tool reports a 30 per cent savings rate, $18,000 invested a year, a $1,050,000 target and 27.7 years. Slide the real return to 4 per cent and the same savings rate takes 30.4 years. The difference is not small, so if you are comparing that page against the Coast FIRE or FIRE number calculators, move the slider to 4 per cent first and the two will agree.
Which figures are tied to a tax year?
Some numbers quoted in the guides are set annually by a government or a regulator. Each one below is stated with the tax year it belongs to and a link to the document it comes from, so you can check whether it has moved before you rely on it.
- United States retirement plan limits, for tax year 2026. The elective deferral limit for 401(k), 403(b), governmental 457 plans and the Thrift Savings Plan is $24,500. The IRA limit is $7,500. The catch-up limit for those aged 50 and over in workplace plans is $8,000, the IRA catch-up is $1,100, and the higher catch-up for ages 60 to 63 stays at $11,250. Source: IRS Notice 2025-67 and the IRS announcement of the 2026 limits.
- Health Savings Account limits, for tax year 2026. $4,400 for self-only high deductible cover and $8,750 for family cover. The additional contribution for people aged 55 and over is $1,000 and is fixed in statute rather than indexed. Sources: IRS Revenue Procedure 2025-19 and 26 U.S. Code § 223(b)(3). These are the figures used on the HSA for early retirement.
- United Kingdom State Pension, for tax year 2026/27. The full new State Pension is £241.30 a week. Source: GOV.UK, the new State Pension. This is the figure the UK Coast FIRE calculator offsets against your spending, and it assumes a complete National Insurance record.
- Early access to United States retirement accounts. The exceptions to the 10 per cent additional tax, including the rule of 55 and substantially equal periodic payments, are listed by the IRS. These drive the retire at 55 guide and the Roth conversion ladder.
- United States health cover before Medicare. Premium savings on the Marketplace depend on your income in the year you claim them. Source: HealthCare.gov, used in part-time jobs with health insurance.
- Country calculator constants. Contribution caps, access ages, state pension rates and inflation targets for Australia, Canada, Germany, India, Ireland, Malaysia, the Netherlands, New Zealand, Pakistan, Singapore, South Africa and the UAE are stated on each country page next to the regulator that sets them, because they move on different calendars. Start at the country calculator index. They are not repeated here, so there is only ever one copy to keep current.
- Balances used for comparison. Median and mean household retirement balances come from the Federal Reserve Survey of Consumer Finances, and the contribution and participation behavior from Vanguard, How America Saves. Both are used on average retirement savings by age.
Where does the 4 per cent rule actually come from?
It comes from two separate pieces of work that are often merged into one. In 1994 William Bengen tested withdrawal rates against United States market and inflation history and found that a first-year withdrawal of 4 per cent of the starting portfolio, raised with inflation each year afterwards, survived every rolling 30 year window he examined. In 1998 Cooley, Hubbard and Walz published portfolio success rates across a range of stock and bond mixes and payout periods in the AAII Journal, the paper almost always called the Trinity study, and they updated it in 2011 with data through 2009.
- Bengen, W. P., Determining Withdrawal Rates Using Historical Data, Journal of Financial Planning, 1994 (FPA reprint, March 2004, member access)
- Cooley, Hubbard and Walz, Retirement Savings: Choosing a Withdrawal Rate That Is Sustainable, AAII Journal, February 1998, the Trinity study
- Cooley, Hubbard and Walz, Portfolio Success Rates: Where to Draw the Line, Journal of Financial Planning, 2011, the update
Two limits are worth stating plainly. First, both rest on the history of one country's markets over one long and unusually good stretch, and a rule fitted to that history is not a law of nature. Second, the headline figure is a 30 year horizon, and someone retiring at 40 is planning for closer to 50. Longer horizons and a worse opening decade both push the sustainable rate down, which is why the withdrawal rate is a slider on every calculator here rather than a constant. The sequence of returns risk guide and the withdrawal strategies comparison both cover what a single fixed rate hides.
What this method does not do
No tax, at either end. The calculators do not model income tax on your contributions now or on your withdrawals later, and they do not know which of your accounts a given pound or dollar is sitting in. No simulation: one steady real return runs the whole projection, so sequence of returns risk, the single largest danger in the first decade of drawdown, is invisible here by construction. No knowledge of your health, your dependants, your housing, or your job security. And a flat real spending assumption, which almost nobody's actual spending obeys. These are sizing tools for testing how sensitive a plan is to each assumption, and they are educational rather than advice. Before you resign, take the output to a qualified professional in your own jurisdiction.
Common questions about the method
Why does the real return come from subtraction rather than division?
Because it is legible and it is what the rest of the field does. The engine sets the real return to nominal return minus inflation minus fees, so the 7 per cent, 3 per cent and 0 per cent defaults give exactly 4.00 per cent. The exact Fisher form, 1.07 divided by 1.03 minus 1, is 3.88 per cent, so the subtraction is about 0.12 percentage points optimistic at these levels. That gap widens as inflation rises, which is why the country calculators that assume higher inflation are the ones where it matters most. If you want the conservative version, drop the return slider by a quarter of a point.
Do the projections compound monthly or annually?
Monthly, and the two agree by construction. The annual real return is converted to a monthly rate of (1 plus real) to the power of one twelfth, minus 1. Each month the balance is multiplied by that rate and then the monthly contribution is added, so twelve steps come to exactly one year of the annual real return. The Coast FIRE discount is applied annually because it is a single division rather than a path: your FIRE number divided by (1 plus real return) raised to the years left.
Why does the savings rate calculator default to a 5 per cent real return?
It is a deliberate convention and the one place on the site where the default real return is not 4 per cent. The savings rate tool exists to show the shape of the savings rate to years relationship rather than to size a portfolio, and it keeps its own slider. At 5 per cent real with a 4 per cent withdrawal rate, a 30 per cent savings rate reaches financial independence from zero in 27.7 years. At 4 per cent real the same savings rate takes 30.4 years. Move the slider to 4 per cent and the table redraws to match the rest of the site.
Where does the 4 per cent withdrawal rate come from?
From two pieces of work, not one. William Bengen tested withdrawal rates against United States market history in the Journal of Financial Planning in 1994 and found that 4 per cent of the starting portfolio, raised each year with inflation, survived every rolling 30 year period he tested. Cooley, Hubbard and Walz then ran success rates across stock and bond mixes in the AAII Journal in 1998, the work usually called the Trinity study, and updated it in the Journal of Financial Planning in 2011. Both are linked in full below. Neither is a guarantee, and both are drawn from one country's history.
What do these calculators deliberately not do?
They do not model tax, in accumulation or in drawdown. They do not run market simulations: one steady real return is assumed, which hides sequence of returns risk entirely. They do not know your health, your dependants or your housing costs, and they assume your spending stays flat in real terms. They are sizing tools. Use them to see how sensitive your plan is to each assumption, then take the output to a qualified professional before you act on it.
Change log
- 4 September 2026. First publication. Formulas, compounding convention, defaults table and tax year 2026 constants written directly from the shipped engine and checked against it.
Same engine, same defaults, your inputs.