Methodology revision (October 2026)
In October 2026 we refreshed the underlying data and revised the scoring method (v4.3). City scores and rankings have changed as a result.
Data refresh
Each indicator now uses statistics published in 2025–2026. The year of each indicator is listed below.
| Indicator | Year | Source |
|---|---|---|
| Population | 2025 | Census |
| Crime rate | 2025（一部 2024） | Prefectural police |
| Land price | 2026 | Official land prices (MLIT) |
| Rent | 2023 | Housing and Land Survey |
| Public transit use | 2020 | Census |
| Retail and restaurants | 2021 | Economic Census |
| Medical facilities | 2023 | Survey of Medical Institutions |
| Schools | 2024 | School Basic Survey |
| Taxable income | 2025 | Municipal tax survey (MIC) |
| Net migration | 2025 | Internal migration report |
| Residency rate | 2025 / 2026 | Census ÷ resident register |
Main changes to the scoring
- Deviation values are based on the 503 municipalities with a 2025 census population of 50,000 or more, so the baseline does not move when the list of featured cities changes.
- Highly skewed indicators such as land price, rent and income are log-transformed before scoring.
- The overall score is a weighted average of the five axes, shown on the same scale as the axes: safety and cost weigh 0.5, the other three weigh 1.
- Grades are S, A, B and C, set by the rank of the overall score among municipalities with 50,000+ residents: S is the top 7%, A the top 30%, B the top 70%.
- Towns where far fewer people actually live than are registered (residency rate below 90%) carry a note; this is not used in the score.
- Crime rates now use municipal figures published by each prefectural police instead of prefecture-wide values. Where 2025 municipal figures are unavailable, 2024 figures are used; otherwise safety is not scored.
Why scores moved
- Schools are now counted per habitable km², which favors built-up cities with many schools nearby.
- With municipal crime rates, cities in the same prefecture no longer share the same safety score.
The article text for each city is being rewritten with the new data, city by city.
← Back to methodology