How the score works
Each measure is rescaled from 0 to 100 across the 50 states: the best state on that measure gets 100 and the worst gets 0. The overall score is the weighted average of those eight scores, so a state that led on every measure would score 100.
| Measure | Weight | Better when | Year and source |
|---|---|---|---|
| Regional price parity | 20 | Lower | 2024 · BEA RPP |
| Effective property tax rate | 10 | Lower | 2024 · Census ACS |
| Violent crime rate | 15 | Lower | 2025 · FBI UCR |
| Life expectancy | 15 | Higher | 2022 · CDC NCHS |
| Poverty among adults 65 and older | 10 | Lower | 2024 · Census ACS |
| Healthcare practitioner and technical occupations | 10 | Higher | 2024 · Census ACS |
| Population age 65 and older | 10 | Higher | 2024 · Census ACS |
| Average temperature | 10 | Higher | 1991–2020 average · NOAA nClimDiv |
Living costs carry the most weight because most retirement incomes are fixed. Average temperature is scored as “warmer is better”, a common retirement preference but not a universal one; the rank range shows how much each state depends on it.
How stable is each rank?
A composite depends on its choices. To show how much, we recalculated the ranking eight times, leaving out one measure each time. The rank range in the table shows the best and worst result for each state. For 1 of the 50 states, the rank moves by two places or fewer, so their position does not hinge on any single measure. A wide range means one measure matters a lot for that state.
What the ranking leaves out
State taxes on retirement income, the cost of long-term care and the quality of particular hospitals are not scored, because each needs more detail than a single statewide figure. The measures come from different periods, from the 1991–2020 climate average to 2025 crime figures, because each agency publishes on its own schedule. Survey margins of error are not carried into the score.
We publish this ranking once per data edition and record any change to the weights in the publication log.