Historical hail archive: January 2016–December 2025. NOAA radar estimates within 25 miles of each city centre. ZIP matches use the first three digits; this is not a property-level report or a live alert service. Coverage and source.
Annual report · generated from NOAA radar data
The 2017 hail year across 50 US metros
Radar detected hail over the metros we track on 1463 metro-days in 2017. 336 of those reached hen-egg size (2 in) or larger.
Data last refreshed: · source: NOAA NCEI Severe Weather Data Inventory, dataset nx3hail
Most hail days in 2017
When it fell
Hail seasons run earlier in the south and later in the northern plains, so a national month curve flattens something that is sharp in any one place. The per-metro pages show the local shape.
The fifteen biggest stones radar estimated in 2017
| Date | Metro | Radar max | Metric | Comparison |
|---|---|---|---|---|
| April 22, 2017 | Huntsville, AL | 4.00 in | 10.2 cm | Grapefruit |
| April 29, 2017 | Lubbock, TX | 4.00 in | 10.2 cm | Grapefruit |
| May 18, 2017 | Lawton, OK | 4.00 in | 10.2 cm | Grapefruit |
| June 9, 2017 | Amarillo, TX | 4.00 in | 10.2 cm | Grapefruit |
| June 11, 2017 | Minneapolis, MN | 4.00 in | 10.2 cm | Grapefruit |
| June 11, 2017 | St. Paul, MN | 4.00 in | 10.2 cm | Grapefruit |
| June 15, 2017 | Lubbock, TX | 4.00 in | 10.2 cm | Grapefruit |
| June 16, 2017 | Des Moines, IA | 4.00 in | 10.2 cm | Grapefruit |
| June 17, 2017 | Topeka, KS | 4.00 in | 10.2 cm | Grapefruit |
| June 26, 2017 | Amarillo, TX | 4.00 in | 10.2 cm | Grapefruit |
| June 29, 2017 | Sioux City, IA | 4.00 in | 10.2 cm | Grapefruit |
| June 30, 2017 | Lawton, OK | 4.00 in | 10.2 cm | Grapefruit |
| June 30, 2017 | Omaha, NE | 4.00 in | 10.2 cm | Grapefruit |
| July 1, 2017 | Lubbock, TX | 4.00 in | 10.2 cm | Grapefruit |
| July 21, 2017 | Chicago, IL | 4.00 in | 10.2 cm | Grapefruit |
Radar estimates the largest stone the algorithm believes was aloft in the storm cell. Ground truth is smaller more often than it is larger, and it varies across a few hundred metres. How we handle that.
Common questions
What counts as a hail day in the 2017 report?
One UTC calendar day on which NEXRAD Level-III detected at least one hail signature within 25 miles of one of the 50 metro centroids we track. Two storms on the same day in the same metro count once.
Why is my city's total different from the news?
Media totals usually count storm reports — people phoning in what they saw — while this counts radar detections in a fixed radius. The two measure different things and neither is wrong.
Does a big number mean a bad claims year?
Not directly. Damage depends on where the hail landed, how big the stones were at ground level, and the age of the roofs underneath. A single well-placed storm over a dense suburb costs more than ten storms over farmland.
Get quotes from roofers who work in your area
We pass your details to licensed contractors in our partner network. Nothing is booked and nothing is charged here.