Parker Logged 11 Big-Hail Days in Three Years. Lakewood Logged One. How to Read a ZIP's Hail History.
A ZIP code's hail history is the fastest free check a roofer can run before buying a list, opening an office, or sending a crew. It's also easy to misread. Here's what the count, the dates, the sizes and the ground reports each tell you, with real Colorado and Texas ZIPs.
Before you buy a lead list, rent a yard sign route, or send three reps to a new suburb, there's a two-minute check worth running: how often has it actually hailed there?
We publish that for free, by ZIP, on the hail history lookup. It covers every ZIP we track in Colorado and Texas and lists each day radar flagged hail over it in the last three years, with a radar-estimated size and a count of ground reports. No signup.
The numbers are simple. Reading them well isn't. Here's what we'd look at, using real ZIPs from this morning's data.
Start with the count, then distrust it
Pull up a few Front Range ZIPs and the spread is wild. Counting only days with a radar-estimated 1 inch or larger (the size where roof damage becomes likely), over the last three years:
| ZIP | Area | Hail days | 1"+ days | 2"+ days |
|---|---|---|---|---|
| 80134 | Parker | 16 | 11 | 3 |
| 80504 | Longmont | 10 | 10 | 2 |
| 80918 | Colorado Springs | 7 | 3 | 0 |
| 80020 | Broomfield | 4 | 3 | 0 |
| 80013 | Aurora | 2 | 2 | 0 |
| 80226 | Lakewood | 1 | 1 | 0 |
Parker had eleven days with 1-inch-plus stones estimated over it. Lakewood, 25 miles west, had one. That's a real difference. Parker sits in the southeast-metro corridor where Palmer Divide storms tend to organize.
But before you read that table as a ranking of markets, notice what the count is actually measuring: how many days any part of the ZIP got hail. A ZIP that covers more ground catches more storms. That's why the top of our Colorado list is rural. Grover, Calhan and Deer Trail each show 50 to 66 big-hail days in three years, because those ZIPs span hundreds of square miles of plains with very few roofs on them.
So the count answers "does hail come through here?" It doesn't answer "how many roofs got hit?" A 1-inch day that clipped the empty northeast corner of a ZIP counts the same as one that went straight down Main Street.
How to use it: compare ZIPs of similar size and density. Suburb vs suburb, not suburb vs ranchland.
Then look at the dates
The count is a climate question. The dates are a sales question, because insurance runs on a clock. Most policies want storm damage reported within roughly a year of the storm (we covered the mechanics in the 12-month claim window post). A ZIP with ten great hail days, all from 2024, is mostly a retail and aging-roof market now, not a storm market.
Compare two DFW suburbs:
- Frisco (75034): 1"+ days on June 2, 2026 (2" radar-estimated), March 25, 2025, and September 25, 2024 (2.5"). The 2024 storm is two years old today. Only the June storm is comfortably inside a typical claim window.
- Parker (80134): 1"+ days on September 20, September 6, and August 26 of this year. All three are weeks old.
Similar-sized suburbs, very different answers. Parker has fresh, stacked storm dates right now. Frisco has one live date and history.
How to use it: count only the days inside the last 12 months when you're sizing a storm push. Count all three years when you're deciding whether a town is worth a permanent presence.
Repeat hits cut both ways
Longmont (80504) shows ten 1-inch-plus days in three years. That sounds like a roofer's dream. It's more complicated.
A neighborhood that got a claim-quality storm in 2024 may already have new roofs from that claim. The 2026 storm then lands on a 2-year-old roof that probably has more life left and a harder claim ahead of it. Stacked storm history raises the odds that any given roof has recently been replaced. That's exactly why storm dates alone make a poor lead list. Roof age decides which homes under a repeat-hit swath are still worth a stop. We walked through that problem here.
The flip side: in a ZIP with frequent hail, the homes that didn't replace after 2024 are now carrying two storms' worth of damage on an older roof. Those are among the best doors in the metro, if you can find them.
The size is the storm's, not the roof's
Every size on those pages is marked radar-estimated, and it describes the storm over the ZIP, not any one house. That distinction matters more than most roofers think.
Radar sees hail inside the storm. It is very good at telling you that hail fell somewhere. In our check against 785 spotter reports, radar caught 99% of Colorado's confirmed hail this year. It's less precise about how big it was at a particular address, and the classic "biggest stone near this home" number runs high. We found the max-around-a-house figure reads about 1.25 inches larger than what ground spotters measured. The full accuracy breakdown is here.
So a 2-inch day on a ZIP page means "a strong hail core crossed this ZIP." It doesn't mean every roof inside took 2-inch stones.
Ground reports are the tie-breaker
The last column counts ground reports: trained spotters, the public, and NWS reports logged near the ZIP that day. They're sparse. Plenty of real hail days have zero, especially overnight or in lightly populated areas. McKinney (75070) shows seven 1-inch-plus radar days and no ground reports at all.
When a day does have them, it's strong corroboration. Parker's history has ground reports on six of its sixteen days. Broomfield's has them on three of four. A radar-estimated 1.25-inch day backed by two ground reports is a much firmer bet than the same number with none.
How to use it: radar tells you where to look. Ground reports tell you how hard to believe it.
What a ZIP page can't tell you
A ZIP history is triage. It's the right tool for four questions:
- Is this area worth opening an office or a route in?
- Is there any storm inside the claim window here right now?
- Is this a one-storm town or a repeat-hit corridor?
- Is the homeowner at the door telling me about a real storm date? (Useful for checking a date before you commit an inspection slot.)
It can't tell you which homes to knock. A single ZIP in the Denver suburbs holds 10,000 to 20,000 homes. The storm touched part of it. Of the homes it touched, the roof age, the stone size at that specific address, and the carrier all decide whether there's a claim. We ran that drop-off in the swath-to-leads funnel. A 43,000-home footprint came down to about 1,500 worth knocking.
That last step, from ZIP to address, is the one that costs roofers the most money when they skip it.
The two-minute routine
Before you commit a crew to a new area:
- Look up the ZIP and its neighbors on the hail history lookup.
- Throw out anything older than 12 months for storm work.
- Compare it only against ZIPs of similar size.
- Give extra weight to days with ground reports.
- Check the state hub pages (Colorado, Texas) for the season-level picture.
If the ZIP passes, the next question is which roofs. That's what our territories do: every home inside the radar footprint, scored on radar-estimated stone size at the address and roof age, for one roofer per city. See what's radar-estimated near you on the Colorado hail map, run your own numbers in the cost-per-job calculator, and if your city is still open, check the territory map. First come.
One roofer per city. Real leads from real storms.
Exclusive Colorado territories with NOAA-verified storm scoring and TCPA-compliant outreach data.
Every DFW roofing site says September through November brings a second hail peak. It does. We counted ten years of spotter reports to see how big it actually is — and what a Metroplex roofer should do with October instead of waiting on it.
Radar found 99% of Colorado's 2026 spotter-confirmed hail. But the size number on the map is a different animal — and the 'biggest stone near this house' figure most maps show you runs about 1.25 inches high. Here's how to read one like a roofer.
Colorado logged roughly 320 hail reports in 2026 and hasn't had a significant one in three weeks. Texas logged 734 and its second season hasn't started yet. Here's the October-through-February math on a storm backlog that's still entirely inside the claim window.