Retire.SimplifiedMethodology

How the math works

What goes into the projection, where the data comes from, and the things this tool intentionally doesn't try to model. If you're working with an advisor or just like reading the fine print, start here. To see the engine checked against hand-worked IRS and SSA examples, and why other calculators may differ, read How we check the math.

Last reviewed: November 3, 2026. We re-check assumptions each January when the IRS, SSA, and CMS publish next year's figures.

Tax brackets, deductions, and capital gains

Federal income tax brackets, standard deduction, additional standard deduction for age 65+, and long-term capital-gains brackets all use 2026 figures from IRS Rev. Proc. 2025-32.

OBBBA senior bonus deduction: $6,000 per qualifying individual age 65+, on top of the existing additional standard deduction. Effective tax years 2025–2028 only (expires after 2028 absent congressional extension). Phases out 6¢ per dollar of MAGI over $75K single / $150K MFJ; reaches zero at $175K / $250K. The projection engine applies this deduction year-by-year and stops applying it after 2028.

Marginal rates and bracket thresholds update each January when the IRS publishes new inflation adjustments.

Later years: brackets, the standard and 65+ deductions, capital-gains brackets and IRMAA tiers grow with your inflation assumption, the way the IRS and SSA index them each year. Thresholds the law never indexes stay fixed: Social Security taxation ($25,000 / $32,000), the 3.8% NIIT threshold, the SALT cap and the OBBBA senior deduction phase-out. State brackets grow with inflation in the states whose law indexes them (California, Minnesota, Wisconsin and Massachusetts's surtax threshold); New York, New Jersey and Hawaii set theirs in statute, so they stay fixed. Oregon indexes all but its top bracket; the engine holds all of Oregon's brackets fixed, which slightly overstates Oregon tax in later years.

2027 status (checked November 3, 2026): SSA has announced the 2027 cost-of-living raise (3.6%) and wage base ($192,300). The IRS has not yet published 2027 brackets, deductions or 401(k) limits, and CMS has not published 2027 Medicare premiums, so the engine uses 2026 figures grown by inflation until they are out.

Sources: IRS Rev. Proc. 2025-32 (2026 inflation adjustments) · IRS — One Big Beautiful Bill Act senior deduction · Tax Foundation: 2026 federal brackets

Retirement contribution limits

401(k) elective deferral: $24,500 (2026); age-50+ catch-up: $8,000; SECURE 2.0 super catch-up for ages 60–63: $11,250.

IRA contribution: $7,500; age-50+ catch-up: $1,100. HSA: $4,400 individual / $8,750 family; age-55+ catch-up: $1,000.

Sources: IRS Notice 2025-67 (2026 retirement plan limits)

Social Security

Benefit calculations use the 2026 PIA bend points ($1,286 / $7,749) and the 2026 SS wage base of $184,500.

Early-claiming reduction follows the IRS two-tier formula: 5/9 of 1% per month for the first 36 months before FRA, then 5/12 of 1% per month thereafter. Delayed-retirement credits accrue at 8% per year past FRA, capped at age 70.

Cost-of-living adjustment (COLA) projected at 2.5%/yr.

Sources: SSA: 2026 COLA fact sheet · SSA: PIA formula bend points

Medicare and IRMAA

2026 Medicare Part B base premium: $202.90/month. Average Medigap Plan G: $165/month. Average Part D premium: $45/month.

IRMAA (income-related Part B surcharge) thresholds and surcharges use 2026 SSA POMS HI 01101.020 figures. Tier 1 single starts at MAGI > $109,000; tier 1 MFJ starts at > $218,000.

The projection deducts IRMAA as spending for each person on Medicare. Like SSA, it uses MAGI from two years earlier; future thresholds are indexed to inflation and the surcharge grows with healthcare inflation. Only the Part B surcharge is modeled (the smaller Part D surcharge is not).

Sources: SSA POMS HI 01101.020 — IRMAA Sliding Scale Tables · CMS: 2026 Medicare Parts A & B Premiums

Required Minimum Distributions

RMDs begin at age 73 per the SECURE Act 2.0. Divisors come from the IRS Uniform Lifetime Table (2022 revision).

Sources: IRS Pub 590-B — Uniform Lifetime Table

Net Investment Income Tax (NIIT)

The 3.8% NIIT applies to the lesser of (a) net investment income, or (b) MAGI over $200,000 (single) / $250,000 (MFJ). The projection engine applies this on top of regular capital-gains tax in years where MAGI crosses the threshold.

These thresholds are statutory (Affordable Care Act, 2013) and have never been indexed for inflation, so more taxpayers cross into NIIT territory each year via wage growth and cumulative inflation.

Sources: IRS Topic No. 559 — Net Investment Income Tax

State income tax

California, New York, New Jersey, Oregon, Hawaii, Minnesota, Massachusetts, and Wisconsin use full graduated bracket math rather than a flat effective rate. High earners in these states see liability matching their actual top-bracket exposure. Massachusetts's modeling includes the 4% 'Fair Share' surtax on income above the inflation-indexed threshold (~$1.1M for 2026).

All other states use a single effective rate — accurate within a few hundred dollars for flat-tax states (PA, IL, IN, etc.) and for typical retiree income levels.

Sourcing confidence varies by state: Hawaii and Minnesota's brackets are confirmed directly against tax.hawaii.gov and revenue.state.mn.us for tax year 2026. Massachusetts's surtax threshold ($1,107,750) and Wisconsin's 2026 bracket thresholds come from secondary sources that agree with each other, because the official 2026 tables weren't reachable (Wisconsin's revenue site still shows 2025 as of November 3, 2026). Both will be checked against the states' own figures when available.

What this tool does not yet model: NYC's additional 3.876% city tax; California's 1% Mental Health Services tax surcharge on income above $1M; local income taxes in PA, OH, MI cities.

Sources: CA FTB — 2025 California Tax Rates · NY DTF — 2025 NY State Tax Rates · NJ DOT — 2024 NJ Tax Rate Schedules · Hawaii Dept. of Taxation — Tax Rate Schedules · Minnesota Dept. of Revenue — 2026 Income Tax Brackets · Massachusetts DOR — 4% Surtax on Taxable Income · Wisconsin Dept. of Revenue — Individual Income Tax Rates (2025 schedule; 2026 thresholds from secondary sources)

Portfolio assumptions and inflation

Default expected return: 7%/yr, treated as a long-run growth rate (the steady projection is the middle case).

Default retirement allocation shift: 60% of working-years return — i.e., a 7% expected return drops to 4.2% in retirement, modeling a more conservative bond-heavier allocation. This is editable on the Plan tab; users with longer time horizons or higher risk tolerance should bump this up.

Default inflation: 2.5%/yr. Healthcare inflation: 3.5%/yr.

The retirement allocation shift activates only when both household members are retired — gap years where one spouse is still working keep the higher pre-retirement return rate.

Healthcare bridge (pre-Medicare)

ACA marketplace premium estimates use national-average benchmark second-lowest-cost silver plan premiums by age (KFF marketplace data, scaled to household size).

Subsidy calculations follow the post-ARPA (now permanent under IRA extension through 2025; reverification needed for 2026+) premium-tax-credit formula. The 400% FPL cliff is currently uncapped per the IRA extension; verify status each year.

2026 Federal Poverty Level guidelines (HHS): $15,650 for a household of 1; $21,150 for 2.

Capital-gains assumption

When the projection draws from a taxable brokerage, it estimates the embedded gains ratio using a simple curve: 30% gains in the first year of retirement, growing 2 points/yr, capped at 80%. This avoids hard-coding either an unrealistically optimistic (low gains) or pessimistic (all gains) assumption.

Annuity withdrawals use a flat 30% gains ratio (exclusion-ratio approximation).

Survivor analysis (couples mode)

When one spouse dies (modeled at their longevity age), the projection applies three IRS rules going forward: (1) Social Security step-up — the survivor automatically claims the higher of their own benefit or the deceased's; the deceased's line drops to zero. (2) Filing status flip: the year of death itself stays MFJ per IRS rule; the year after flips to single. Single brackets compress at the top, so the survivor's effective tax rate often rises even on the same income. (3) Spousal rollover: the deceased's tax-deferred and Roth balances roll into the survivor's name (no immediate distribution required, preserves tax-deferred status).

RMDs are computed per-spouse, using each spouse's own age and own 401(k) balance — the previous combined-pool approach over-RMD'd younger spouses with separate accounts.

Pension survivor benefit: if a joint-and-survivor option was elected at retirement (50% / 75% / 100%), the surviving spouse continues receiving that fraction of the deceased's pension. Default is 0% (single-life payout) since most users don't elect survivor benefits.

What survivor analysis does not yet model: qualifying widow(er) tax filing for 2 years post-death (assumes no dependent children — survivor flips straight to single); cost-basis step-up on inherited taxable brokerage assets (we use the same gains-ratio curve, which slightly overstates capital-gains tax in survivor years); inherited IRAs for non-spouse heirs (the 10-year rule applies to children but not surviving spouses, who roll over).

Sources: IRS Pub 559 — Survivors, Executors, and Administrators · SSA — Survivors Benefits

Plan odds (Monte Carlo) and the history check

The headline "your money lasts your whole plan in X% of markets" runs the full projection above — taxes, Social Security, required withdrawals, healthcare, long-term care, survivor years — through 500 market sequences, not a simplified model. A market fails if any year runs short of money. The random seed is fixed, so the same plan always shows the same percentage; sampling alone moves it by about ±3 points.

Yearly returns are drawn from a normal distribution around your expected return. The plan stores only a return, so the size of yearly swings is read off typical stock/bond mixes (7% is treated like a roughly 50/50 mix swinging about 9% a year). In retirement, returns and swings are both scaled by the retirement allocation shift. The average year is lifted by half the variance so the middle outcome compounds at your stated rate.

The history check replays every complete run of years since 1928 (S&P 500 and 10-year Treasury, Damodaran/NYU Stern) in their real order, centered on your expected return so it tests bad timing rather than a different average.

Known limitation: normal-distribution returns underestimate left-tail risk vs. fat-tailed empirical distributions. Real markets have crashes that random draws don't fully capture — the history check is there to cover that.

What this tool intentionally does NOT model

Estate tax (federal exemption is $15,000,000 in 2026 — made permanent by the One Big Beautiful Bill Act, which removed the 2026 TCJA sunset. Relevant only to households well above mass-affluent).

Alternative Minimum Tax (AMT) — relevant to ISO exercises and high-SALT states.

Net Unrealized Appreciation (NUA) on company stock in employer 401(k)s.

K-1 income, deferred compensation, RSU vesting schedules — relevant to executives and business owners.

Concentrated single-stock risk.

Long-term care insurance policies. Long-term care itself is modeled: $3,600/month in today's dollars by default (about a 65% chance of needing care × typical assisted-living cost), ramping up over the last ~5 years of the plan and rising with healthcare inflation. You can change or remove it under Expenses.

Reminder. This tool is an educational planning estimator, not financial advice. The math is honest about its assumptions and citations, but every household has nuances (deferred comp, K-1 income, large taxable concentrations, divorce, kids, business interests, etc.) that a competent fee-only advisor or CPA will handle better than any free calculator. If you're managing $1M+, talk to one.