How the Match Score works

A city’s identity, population, land area, and elevation come from live public geographic and demographic sources — OpenStreetMap Nominatim and Wikidata — read at the time of a search. City descriptions and local context are summarized from public Wikipedia content.

The fifteen category inputs (cost of living, housing, income, taxes, weather, crime, schools, healthcare, employment, commute, walkability, recreation, culture, airport access, and retirement suitability) are derived CityMove planning estimates. They are built from CityMove reference state and territory tax rules, regional cost-of-living patterns, and city-size signals — not a live proprietary feed for each individual place. The taxes category is a CityMove-generated tax comparison factor, not an official tax burden or government-published rate.

Raw inputs are bounded to plausible ranges before use so that a single outlier cannot distort a category. Population is used on a logarithmic scale so that a handful of very large metros do not overwhelm everything else, and derived figures are rounded to sensible planning precision.

Values are computed deterministically from a city’s name and state, which means the same place always produces the same numbers across visits and devices. There is no random noise between page loads.

Every category is converted to a common 0–100 scale before anything is combined. Each raw signal is mapped against a national reference range using min–max normalization: a value at the favorable end of the range approaches 100, and a value at the unfavorable end approaches 0.

For categories where a lower raw number is the better outcome — cost of living, the CityMove tax comparison factor, and commute time — the scale is inverted so that "better for you" always reads as a higher score. This shared scale is what lets categories as different as a home price and a commute time be weighed against one another fairly.

The overall Match Score is a weighted average of the fifteen normalized category scores. You set each category’s weight from 0 to 10; a weight of 0 removes the category from the calculation entirely, and higher weights pull the overall score toward that category’s performance.

Because the weights are yours, the score is personal. Two people can look at the same city and see very different Match Scores — that is the intended behavior, not an error.

When a live geographic source does not return a figure (for example, an unavailable land area or elevation), the report shows the field as unavailable rather than inventing a precise value, and the model falls back to a size- and region-based estimate for scoring purposes.

Because every category always resolves to an estimate, the model reports a data-confidence level alongside each city rather than implying measured certainty.

Live geographic and demographic values refresh on every search. The scoring model, its reference ranges, and the tax and cost-of-living tables are reviewed periodically; the current version’s last-updated date is shown on this page and on every data-heavy page.

When the model changes, previously viewed scores can shift. Treat any saved score as a snapshot of the model on the day you generated it.

The Match Score estimates how closely a place matches the preferences you selected. It is explicitly not an objective ranking of which city is "best," and it should not be used as financial, legal, or relocation advice.

Category inputs are planning estimates that can differ from current, hyper-local reality — crime and school quality in particular vary sharply between neighborhoods within the same city. Use the score to build and narrow a shortlist, then confirm the specifics that matter to you with local, official sources before making a decision.