How to read this

This page covers every Oklahoma town that could be located — 102 of them — and all 77 Oklahoma counties. It counts hail detected by weather radar between January 2013 and August 2026, within 10 miles of each town centre.

Hail days per year

The average number of days each year on which radar detected hail near that town. A day counts once, however long the storm lasted or however many times radar saw it. This is the column the table is ranked by.

2″+ storms

The number of separate days on which the hail was estimated at two inches across or larger. Two-inch hail is roughly the size of a hen’s egg, and is the point at which damage to roofs and vehicles becomes common. This is a count of days over the whole period, not a yearly average.

Total since 2013

Every hail day counted across the full period, about 13.7 years. This is the sample the yearly average is drawn from — a town with 200 total days has a firmer average than one with 20.

County

The county the town sits in. Counties are shaded on the map by their own hail day count, which is measured differently — see note 3 below.

Notes on method

1. The source is NOAA’s Severe Weather Data Inventory, specifically the NEXRAD Level‑3 hail detection product. Each record marks a moment when weather radar identified a probable hail signature, with an estimated maximum stone size.
2. A hail day is one calendar day on which radar detected hail of at least 0.75 inches within 10 miles of the town centre, at 50 percent confidence or better. A storm tracked across dozens of radar sweeps counts once. Days with fewer than five separate detections are excluded, because isolated single‑sweep signatures are frequently radar artifacts rather than hail.
3. County figures on the map count hail anywhere inside the county boundary, so they are not comparable to the town figures, which use a fixed 10‑mile radius. A large rural county will show more hail days than any single town inside it.
4. Maximum hail size is deliberately not reported. The radar product estimates stone size, but that estimate is capped, and over thirteen years almost every county reaches the ceiling at least once. A “largest hail” column would therefore read the same nearly everywhere and would tell you nothing. Storm counts at one inch and two inches are used instead, because those thresholds sit well below the cap and do discriminate between places.
5. Coverage begins in 2013. The NEXRAD dual‑polarisation upgrade, completed that year, substantially improved the ability to tell hail from heavy rain. Earlier records are materially less reliable and mixing them in would distort the comparison.
6. These are radar estimates, not confirmed ground reports. Radar infers probable hail from reflectivity aloft; it does not observe stones on the ground. Places far from a radar site may be undercounted, because the beam sits higher above ground at distance. Read the rankings as a strong indicator of relative hail exposure rather than an exact count of hailstorms.
7. Sources. Hail records from NOAA National Centers for Environmental Information. County boundaries from the US Census Bureau. Town coordinates from OpenStreetMap. Compiled by soonerhomeservices.com.