How to Read a Hail Swath Map

A hail swath map shows the path a hailstorm took across the ground and shades that path by the largest hail size estimated anywhere inside it. You read it in three steps: find the band that covers the location you care about, read the legend to see what hail size that band’s colour represents, and then treat that number as the storm’s peak rather than as the size that fell on any particular roof. The last step is the one almost everyone skips, and it is the one that decides whether the map tells you something true.

This page explains what the bands mean, what the size numbers are and are not, and where the map stops being able to answer the question you actually have.

A swath is a corridor, not a county

Hail does not fall evenly over an area. A storm cell lays down a long, relatively narrow corridor of hail as it moves, and the stones are concentrated in part of that corridor rather than spread across it. A swath map draws that corridor.

This is the whole reason a swath is more useful than a county-level storm report. A report tells you hail occurred somewhere in a county that may be a thousand square miles. A swath tells you it occurred along a specific track, and lets you see that one side of a neighbourhood sits inside it and the other side does not. Two houses a few blocks apart can genuinely have different outcomes from the same storm, and the swath is what makes that visible.

The colour band is a size estimate, and it is a peak

Every published hail map shades its swaths by estimated hail diameter, with a legend mapping colours to size ranges. Two properties of that number matter far more than the palette.

It is the largest hail estimated in that band, not the average. A band is an upper bound over the area it covers. Most points inside it received less than its headline figure, some considerably less.

It is an estimate rather than a measurement. Nobody measured the hail at your house. The size is inferred from radar and, on some products, revised afterwards by a meteorologist looking at the same storm.

That second point is worth putting a number on, because it is usually glossed over. In our own archive of named storms, 114 carry both a meteorologist-reviewed hail size and an algorithm-derived one for the same storm. The two agreed exactly on 48 of them. On the other 66 they did not, with the largest single disagreement running to 1.5 inches and six storms differing by a full inch or more.

That is not an error rate. There is no ground-truth hail measurement sitting in the archive to score either estimate against, so it does not establish that the algorithm is wrong or that the human is right. What it does establish is narrower and more useful: a hail size on a swath map is a provisional figure that a reviewer revises more often than not. Treat it accordingly.

What the size numbers actually mean

Across the 120 named hail storms we tracked between 31 July 2025 and 22 July 2026, the median peak hail size was 2.25 inches. Thirty-nine of them, just under a third, peaked at 3 inches or larger; 23 peaked below 1.5 inches.

Two cautions travel with that distribution and it is misleading without them. The sizes are peaks, so they describe the worst cell in each storm rather than a typical point inside it. And the population is selected: a storm is in that archive because it was significant enough to be named, so this is a distribution of notable storms and not of hail in general.

The other number you will see on nearly every hail map is a 1-inch line. That threshold comes from the National Weather Service, which defines a severe thunderstorm as one producing a tornado, winds of at least 58 mph, and/or hail at least 1 inch in diameter. It is a criterion for issuing a warning about a storm. It is not a roofing threshold, and no federal source states the size at which a roof is damaged, because that depends on the roof covering, its age, the impact angle and the wind. If a page tells you a single number at which roofs fail, it is telling you something nobody has established.

Where the map stops

A swath map can tell you that a hailstorm passed over an area, roughly where its track ran, what day it happened, and an estimated upper bound on the stones inside it. That is genuinely valuable, and it is where the map’s competence ends.

It cannot tell you what fell on one roof, because it reports peaks over bands. It cannot tell you whether a roof is damaged, because that is a function of the roof rather than the storm. And its size figure is provisional, as the disagreement between the two estimates above shows.

So the correct last step of reading a swath map is to stop reading the map. Check the specific address against the storm history, get the date, and then have someone look at the roof.

Checking one address

SwathIQ publishes a free per-address hail history lookup that returns the storm dates and estimated sizes recorded for a single point, which is the question a swath map cannot answer by eye. The full interactive hail map is there if you want to see the swaths themselves.

We should be direct about who we are, because it changes how you should read this page: The Call Center Doctors do not repair roofs and never will. We run the phones for roofing contractors, which on a storm week means storm damage appointment setting. A homeowner reading this is not a customer of ours, which is exactly why we have no reason to tell you that a swath over your house means you have a claim.

If you are a roofing contractor rather than a homeowner, the map is the start of a different problem: deciding whether a swath is worth working and then reaching the people inside it before anyone else does. That is covered in storm restoration appointment setting and in how to get roofing leads after a storm.

The half of that decision a map cannot help with is whether you can work the swath once you have chosen it. The appointment capacity planner works out how many appointments your reps can run in a month and how many setters it takes to fill that calendar, which is usually the number that decides how wide a swath is worth chasing.

Limitations of everything above

The hail-size distribution is drawn from named storms in one rolling year and is a distribution of peaks, so it should not be read as what hail typically does. The agreement figures compare two estimates against each other and not against measured hail, so they bound confidence rather than accuracy. The 1-inch criterion is quoted from the National Weather Service glossary and describes storm warnings only. Nothing here is an assessment of any particular roof, and nothing here is insurance advice.