About PortfolioAtlas

PortfolioAtlas answers one simple question: "If I retire with $X, what kind of life can I actually live?" We break it down city by city, dollar by dollar.

How We Calculate

We estimate each city's costs across three lifestyle tiers. The familiar 4% rule (the "Trinity Study" benchmark — spending × 25) is shown only as a reference number. Whether a city counts as "you can retire here" is decided by the historical backtest below, not the rule.

For each lifestyle tier, we research and estimate realistic monthly costs for housing, groceries, dining out, healthcare, transportation, entertainment, utilities, and — at higher tiers — domestic help and luxury extras. Those costs feed both the 4% rule benchmark and the historical backtest that produces the verdict.

The FIRE Score

Each city receives a FIRE Score from 1 to 10 based on a weighted combination of four quality-of-life factors: safety (30%), healthcare quality (30%), infrastructure (20%), and expat friendliness (20%). Cost of living is intentionally excluded — it's already addressed through the affordability filters and lifestyle tiers.

A higher score means a better quality of life for retirees. Cities like Singapore, Tokyo, and Amsterdam score highest thanks to excellent safety, world-class healthcare, and strong infrastructure. The FIRE Score helps you answer: "If I can afford this city, how good is it to actually live there?"

How the retire verdict works

Whether a city says "You can retire here" is decided by a historical backtest, not a rule of thumb. We take your actual plan — your portfolio, the selected lifestyle tier's annual cost, and your retirement horizon — and run it through every historical period of that length since 1871, using real US market returns from Robert Shiller's dataset (1871–2022). The portfolio is a 75% stock / 25% bond mix rebalanced once a year, and each year we raise the withdrawal for that period's actual inflation.

The result is a survival rate: the share of those historical periods in which the plan never ran out of money. It is a percentage, not a yes or no.

We turn that survival rate into a plain verdict, against the confidence level you pick (default Balanced, 90%):

• "You can retire here": survived at or above your confidence level.
• "Close — worth a closer look": survived within 10 points below it.
• "Not quite yet": survived lower than that.

The homepage map is binary: a city shows green when it clears the same confidence line. The middle "Close" band appears only on the city page.

You still see the 4% rule benchmark because it is the number most people know, but it is only a sizing shortcut: it assumes roughly 30 years and leaving your principal intact forever. The headline figure, "Needed to retire here", is instead solved from the backtest — the portfolio that actually reaches your confidence level over your retirement length — so it always agrees with the verdict.

Two controls under "Refine your projections" shape this. Confidence level (Aggressive 80% / Balanced 90% / Conservative 95%) sets the bar for both the verdict and the target. Retirement length sets how long each backtested period runs.

A few honest limits:

• It is based on US market history. Past performance does not guarantee future results.
• The allocation is a fixed 75% stocks and 25% bonds, and costs assume a single adult.
• We do not model moving mid-retirement. Each city's number assumes you stay there. The gap between two cities' survival rates is itself the value of geographic flexibility, which you can see by comparing them rather than us simulating it.

Important Disclaimers

The cost data on PortfolioAtlas is researched and estimated for informational purposes only. Actual costs vary based on personal lifestyle choices, exchange rate fluctuations, inflation, and individual circumstances.

This site does not provide financial, tax, or legal advice. Always consult with qualified professionals — a financial advisor, tax specialist, and immigration attorney — before making any relocation or retirement decisions.

Cost data reflects approximate values as of early 2026 and will be periodically reviewed.