We ran our own running-weather thresholds against 20 years of climate normals for every US city we track. Here's the actual ranking, not a vibe.
Every “best cities for running” list I’ve read is really just someone’s opinion wearing a headline. So instead of writing another one, we pulled 20 years of climate normals for every US city in WeatherPerch’s own location dataset and ran them through the exact same thresholds our running weather pages use to score a live forecast: NWS’s precipitation-likelihood bands, the NWS Wind Advisory threshold, and NOAA’s Heat Index caution tiers.
No live weather went into this — normals only, because a “best cities right now” claim would be wrong by next week. This is what a typical September day looks like in each city, based on the last two decades.
The Method, Briefly
For each city, we took the September climate-normal high temperature, average rain-day frequency, and average wind speed, then ran them through the same per-factor risk assessment (rain, wind, heat/cold) our location pages already show for live forecasts. Cities are bucketed by overall risk level first, then ordered within a bucket by how close the normal high sits to a comfortable ~58°F running temperature, adjusted for rain and wind.
The Ranking
| # | City | Normal High / Low | Rain Chance | Normal Wind |
|---|---|---|---|---|
| 1 | Boston, Massachusetts | 73°F / 58°F | 37% | 11 mph |
| 2 | Columbus, Ohio | 77°F / 59°F | 38% | 9 mph |
| 3 | Chicago, Illinois | 74°F / 61°F | 40% | 13 mph |
| 4 | Baltimore, Maryland | 78°F / 62°F | 42% | 10 mph |
| 5 | Charlotte, North Carolina | 82°F / 64°F | 41% | 8 mph |
| 6 | Austin, Texas | 90°F / 72°F | 39% | 10 mph |
What This Doesn’t Tell You
A 20-year normal is a baseline, not a forecast. If you’re planning a run next Tuesday, check that city’s actual 7-day forecast — normals tell you what to generally expect from the season, not what’s coming this week. This list is useful for choosing a race destination or a training base months out, not for deciding what to wear tomorrow.
We’ll revisit this for other months and other activities (hiking, cycling) as the underlying normals data gets used more broadly across the site.