Colorado's Last Big Stone Fell August 19. Here's What You Sell Until March.

The last hail report worth a roofer's attention in Colorado this year came on August 19, near Arriba out on the eastern plains — a 3-inch stone in a county with almost nobody living in it. Three weeks later, nothing has followed it. Colorado's hail season runs roughly April through September with the peak in June, and for practical purposes the 2026 version is over.

Which means the annual ritual is about to start. Sometime in the next four weeks, a lot of Front Range roofers will let the canvass crews go, stop paying for data, and tell themselves they'll spin back up in March.

That's a defensible move if your only product is a fresh storm. It's an expensive one if you understand what October actually is.

The inventory didn't go anywhere

Colorado logged about 320 hail reports across roughly 30 counties in 2026 — spotter and NWS local storm reports, the public record, not anybody's proprietary map. El Paso County led with 37. The biggest stone of the year was 4.5 inches near Genoa. Texas logged 734 reports, topping out at a 5-inch stone at Camp Wood on April 29, with Dallas County alone carrying 27.

Not one of those events has expired.

Most homeowner policies require notice "as soon as practicable," and as a practical matter carriers want a roof claim reported inside about a year of the date of loss. The policy language governs, and it varies — but work from that one-year rule of thumb and lay the 2026 season against a calendar:

Now look at what that means for the five months in front of you. From October through February, every storm of the 2026 season is still live, and nothing new is landing on top of it. The events that actually time out during those five months are storms from October 2025 through February 2026 — which in Colorado is very close to an empty set.

That is the only stretch of the year when your list holds still. In June you're triaging a fresh core every week or so while last spring's storms quietly hit their deadline behind your back. In November nothing is expiring and nothing is arriving. You get five months to work a full season's backlog at your own pace, against roughly zero competing crews, and the homeowners you reach have mostly never heard from anybody about the July storm. We wrote up the mechanics of that closing-window lane in the 12-month clock; October is when you finally have the hours to run it.

If you're in Texas, don't copy Colorado's calendar

Colorado shuts down. North and Central Texas do not.

Texas gets a second severe season in October and November. As the jet stream migrates south, the temperature contrast sharpens and the fronts get stronger, and DFW picks up a genuine secondary peak — including a well-documented October tornado bump. These systems aren't as widespread as the April–May run, but they produce hail and damaging wind on streets full of roofs.

So the Texas version of this post is the opposite advice: your off-season starts in December, not October, and the roofers who furlough their canvass crews on October 1 hand the fall storms to whoever stayed staffed. Same for wind — a fall front doesn't need a hailstone to strip a corridor of shingles, which is the case we made in the San Antonio wind piece.

What actually changes in winter: your channel, not your effort

Here's the honest constraint nobody puts in these posts. Door-knocking gets structurally worse in winter, and it has nothing to do with motivation.

Denver's sunset moves from about 8:30 PM in early July to about 4:35 PM in early December. Run a practical knock window of 10 AM to dusk and you go from roughly ten hours of daylight to roughly six and a half. That's a 35% cut in door-hours per canvasser per day, before you account for the days nobody opens the door because it's 22 degrees.

The correct response is not "knock harder." It's to shift weight onto the channels that don't care about daylight — mail, email, and manual dials — and to work a tighter, better-scored list because you have fewer swings.

That makes winter the season where compliance actually bites, because you're on the phone more. Colorado and Texas each have their own quiet hours and consent rules, and if you sell into Maryland the Stop the Spam Calls Act is stricter than federal law on both the calling window and automated contact. The field guides are here: Colorado, Texas, Maryland. General information, not legal advice — talk to your own counsel.

The other winter lane is the one that has no storm in it at all. Carriers are steadily moving 15-to-20-year roofs to depreciated coverage and non-renewing the rest, and that homeowner has a deadline of their own that ignores the weather entirely. That's the insurability cliff, and it's a retail conversation you can have in January.

You're not selling installs. You're selling March.

The standard objection is that you can't roof in the cold. Mostly you can — asphalt shingles just don't self-seal reliably below about 40 degrees, so cold-weather work means hand-sealing tabs and following the manufacturer's cold-weather instructions instead of trusting the sun.

But that objection misses the actual play. A contract signed in December doesn't have to be installed in December. It fills your March and April calendar — the exact weeks when every other roofer in the metro is starting from a cold list and a zero backlog. The off-season isn't production. It's the pipeline you walk into spring holding.

Run the number for your own shop

Take a Mid City territory at $699 a month. October through February is five months — $3,495.

A Front Range insurance re-roof commonly lands somewhere in the $18,000–$30,000 range. Call it $22,000 at a 30% gross margin: about $6,600 of gross profit on one job. That single contract covers the entire off-season roughly 1.9 times over. Two jobs across five months and the data paid for itself plus a chunk of next spring.

Those are illustrative numbers, not a forecast — your ticket, your margin, your close rate, and nothing here promises anybody a job. Run it with your own figures on the cost-per-job calculator before you decide. And if you're going to hold the territory through the winter anyway, note that annual billing on our pricing is ten months for twelve — which is, functionally, the two deadest months of the year for free.

The short version

Colorado's storms stopped on August 19. Colorado's leads did not. You have roughly 320 events' worth of Colorado backlog and 734 events' worth of Texas backlog sitting inside the claim window, five months of clear calendar, and almost no competition on the doors or the phones.

Sort your list by storm date rather than by score this month and look at what's been sitting there since July. Check what fell near you on the Colorado hail map or the Central Texas map. If your city is still open, take a look at the map — one roofer per city gets it, and the roofer who holds it in February is the one holding it in May.