How Do Insurance Companies Determine Wildfire Risk?
They look at your property's terrain, vegetation, nearby fuel sources, local fire history, and distance from a fire department or hydrant.
Free to compare
What your non-renewal letter is actually about
If a wildfire risk model flagged your property, the letter rarely spells that out. It may just say the company is reducing exposure in your area. Behind that decision is usually a score built from your address. Slope of the land, how close trees and brush sit to the house, how far you are from a fire station or hydrant, and whether your area has burned before all feed into it. Some insurers also factor in your roof material and whether there's defensible space cleared around the structure. None of this shows up on your declarations page, so you won't find it there. What you can do is look at your property the way that model does, and fix what's fixable before you apply again.
Every insurer runs its own wildfire model against the same terrain and fuel data, so a company that dropped you can still look at your address and return a very different score. Some weigh defensible space more heavily. Others care more about hydrant distance or fire history. Checking a few before your lender's deadline closes is how you find one that actually prices your risk closer to what it is.
What actually goes into the score
Wildfire models start with public data. Satellite imagery, topographic maps, vegetation surveys, and decades of fire perimeter records from state and federal agencies all feed in. A company doesn't send someone to walk your property before sending a renewal notice. The model estimates your risk from the air and from history, then a person may follow up with an inspection if the score comes back borderline.
Slope matters more than most homeowners expect. Fire moves faster uphill, so a house on a hillside with brush below it scores worse than the same house on flat ground, even with identical landscaping. Models also weigh what's called the fuel load within a set radius of the structure, meaning the density and dryness of trees, grass, and brush that could carry fire toward the house. A single dead tree fifty feet away isn't what moves the number. A continuous stretch of untrimmed vegetation from the property line to the wall is.
Distance from a fire station and from a reliable water source, usually a hydrant, gets scored separately from the vegetation risk. These two numbers reflect how fast a fire can be fought once it starts, not how likely it is to start. A property deep in a canyon with no hydrant access can score poorly here even if the surrounding land was burned clear two years ago and has little fuel left. Insurers also pull local and regional fire history going back ten to thirty years, since an area that has burned once tends to burn again as vegetation regrows.
Roof and siding materials factor in too, though usually as a smaller piece of the overall score. A composite or metal roof reads differently to a model than wood shake. Stucco or fiber cement siding scores better than wood siding sitting close to a fence line that touches the house. None of these factors are visible to the homeowner on paper, which is why two neighbors can get different letters from the same company. The model isn't scoring the neighborhood. It's scoring the specific lot, the specific structure, and the specific few hundred feet around it.
Why two houses on the same street get different letters
Wildfire models start with public data. Satellite imagery, topographic maps, vegetation surveys, and decades of fire perimeter records from state and federal agencies all feed in. A company doesn't send someone to walk your property before sending a renewal notice. The model estimates your risk from the air and from history, then a person may follow up with an inspection if the score comes back borderline.
Slope matters more than most homeowners expect. Fire moves faster uphill, so a house on a hillside with brush below it scores worse than the same house on flat ground, even with identical landscaping. Models also weigh what's called the fuel load within a set radius of the structure, meaning the density and dryness of trees, grass, and brush that could carry fire toward the house. A single dead tree fifty feet away isn't what moves the number. A continuous stretch of untrimmed vegetation from the property line to the wall is.
Distance from a fire station and from a reliable water source, usually a hydrant, gets scored separately from the vegetation risk. These two numbers reflect how fast a fire can be fought once it starts, not how likely it is to start. A property deep in a canyon with no hydrant access can score poorly here even if the surrounding land was burned clear two years ago and has little fuel left. Insurers also pull local and regional fire history going back ten to thirty years, since an area that has burned once tends to burn again as vegetation regrows.
Roof and siding materials factor in too, though usually as a smaller piece of the overall score. A composite or metal roof reads differently to a model than wood shake. Stucco or fiber cement siding scores better than wood siding sitting close to a fence line that touches the house. None of these factors are visible to the homeowner on paper, which is why two neighbors can get different letters from the same company. The model isn't scoring the neighborhood. It's scoring the specific lot, the specific structure, and the specific few hundred feet around it.
Before you apply again
Clear brush back from the house, note your roof material, and measure roughly how far you are from a hydrant or fire station. Then compare quotes with those details in hand, before your current coverage ends.
