Frequency & Severity — A Hurricane Insurance Concept That Wildfire Underwriting Needs
Wildfire Insurance must evolve beyond parcel-level assumptions and embrace data-driven risk profiles rooted in actual fire behavior
In the world of catastrophe modeling, the concepts of frequency and severity are foundational. In hurricane risk modeling, frequency refers to the number of storms that make landfall, while severity refers to how destructive those landfalls are — wind speed, surge damage, rainfall, and ultimately, insured loss. The industry has decades of data, strong atmospheric modeling tools, and clear distinctions between the probability of an event and the scale of damage it might cause.
Underwriting wildfire risk, by contrast, don’t benefit from this clarity.
Despite being one of the fastest-growing sources of insured loss in the U.S., wildfire risk modeling remains stuck in a structure-centric paradigm. Underwriters still base decisions on parcel-level conditions: the roof material, vent mesh size, defensible space, and other WUI standards — essentially, a checklist of building survivability. These attributes are valuable, but they say little about whether a fire is likely to occur in the first place. That’s the missing more than half of the equation.
Wildfire Frequency Isn’t Modeled the Way Hurricanes Are
When insurers assess wildfire risk, the implicit assumption is that a wildfire will occur. The traditional approach asks the question: In a wildfire, how badly will this structure fare?
The traditional approach to property insurance underwriting never asks: What is the likelihood that a wildfire even reaches this parcel?
This is the missing variable in the industry’s wildfire “calculus.” Unlike hurricane models, traditional wildfire risk assessment skips the geospatial dynamics of fire behavior. It ignores how vegetation, slope, wind exposure, and ignition history interact across a landscape. Instead, most models depend on lab-derived burn data and scattered claims histories to guess at “intensity” — the likelihood of total structure loss if fire arrives.
The result? A severe mispricing of risk. Properties in low-burn landscapes are rated the same as those in high-frequency ignition corridors. Some homes are over-insured; others are uninsurable. And in the face of climate volatility, insurers retreat — not because they fear losses, but because they can’t price them.
A New Approach: Wildfire Risk Profiles Based on Real Burn Behavior
Athena Intelligence is changing that. Our wildfire analytics platform, Voice of the Acre®, applies a geospatial modeling framework that redefines “frequency” and “severity” using empirical fire behavior, in the context of the bioregion , NOT assumptions.
Unlike a simple score, based on What happens to the structure during a wildfire? Athena starts by asking, Given everything for miles around that influences this parcel of land, how often has a wildfire occurred when this set of conditions existed anywhere in this bioregion? From this probabilities can be derived.
Traditional Models: Opaque black-box scores with limited transparency.
Athena’s Geospatial Approach: A bankable probability — explainable, traceable to raw data, and defensible by actuaries.
Athena incorporates vegetation dynamics, and landscape features well beyond the parcel to generate wildfire risk profiles that reflect how the land behaved across historic fire events. The profiles are decoupled from individual addresses and are based on statistically valid patterns across space and time.
Here is an example:
In this case, the property may look safe, but locations with the same characteristics have burned 11 time (out of 517 annual profiles) over the past 5 years. In the context of a portfolio of property insurance, the probability of this property being in a wildfire is 91 bp (1 out of 109, or slightly less than 1% chance of being in a wildfire, if a wildfire occurred anywhere in Monterey County.)
This is real frequency, not statistics and hypotheticals.
This opens the door to transparent underwriting, reserving, more accurate pricing and unit growth without adverse selection.
Insurance Doesn’t Work If It Can’t Quantify Risk
Property insurers today are retreating, and unit sales are falling. In wildfire-prone regions, carriers are pulling out — not because losses are unmanageable, but because pricing those losses accurately is impossible with legacy tools.
Athena fills this void. We provide insurers with Green/Yellow/Red underwriting decision flags and quantified risk scores that can be easily integrated into existing policy software.
Green is low risk, and is eligible for standard policies. Properties with a Yellow flag have risk low enough that they can be considered for Excess and Surplus lines. Athena has the probabilities that can be used in pricing and reserving.
Athena’s data helps insurers answer the new question, how has the risk to this property changed as the weather volatility has increased?
The answers are no longer speculative.
We’re not modeling an imagined future. Athena can show how a very specific terrain profiles actually burned.
This ties environmental risk of the location directly to the balance sheet of insurer, mortgage holder, and property owner, with a precise probability.
Borrowing the language of hurricane modeling — frequency and severity — Athena helps insurers reframe how wildfire risk should be priced, managed, and ultimately insured. Move beyond checklists for a parcel and embrace insights from geospatial analysis.
With Voice of the Acre®, insurers can finally see wildfire not just as a natural catastrophic hazard, but as a measurable, quantified risk. With the right data, measurable risk becomes insurable again.
Athena Intelligence is a data vendor with a geospatial, conditional, profiling tool that pulls together vast amounts of disaggregated wildfire and environmental data to generate spatial intelligence, resulting in a digital fingerprint of wildfire risk.
Clients include financial services companies, including insurance, electric utilities, communities and homeowners associations (HOAs). Athena’s geospatial intelligence is incorporated into multiple products that can be accessed through an online portal. Athena’s data is currently used in wildfire mitigation plans (WMP) and public safety power shutoffs (PSPS), Community Wildfire Protection Plans (CWPP), property insurance underwriting and portfolio risk optimization.
You can learn more on our YouTube Channel @Athena_Intelligence Athena Intelligence (AthenaIntel.io) — YouTube Reach out to me at Elizabeth@AthenaIntel.io and follow us on LinkedIn
