Methodology: How We Estimate Car Insurance Costs

Every number on this site is either a published statistic (with its source named) or an estimate produced by the model below. The model is deliberately simple and fully disclosed so you can judge it — and it is never a quote.

Data vintage. NAIC Auto Insurance Database Report (latest edition used: see data sources)

1. What we estimate

We estimate a typical annual premium for three coverage levels — state minimum liability, standard liability (100/300/100 with uninsured motorist coverage) and full coverage (standard liability plus collision and comprehensive) — for a given state, city or ZIP code, vehicle and driver profile. The default profile is a driver in their 30s with a clean record, 7,500–12,000 miles a year and $500 deductibles.

2. State base premiums (NAIC)

The starting point is the NAIC Auto Insurance Database Report, which publishes, for every state, the average expenditure per insured vehicle and the average premiums for liability (L), collision (C) and comprehensive (K) among drivers who buy each coverage. Countrywide values are used for national figures (e.g. L = $737, C = $464, K = $238).

3. Coverage level

The NAIC liability average reflects the mix of limits people actually buy, which we treat as equivalent to 50/100 bodily-injury limits. To price other limits we use increased-limits factors (ILF), a standard rating concept:

Bodily injury limit15/3020/4025/5030/6025/6530/6550/100100/300250/500
ILF0.900.961.001.041.031.051.201.451.72
  • Minimum liability = L × ILF(state minimum) ÷ ILF(50/100)
  • Standard liability = L × ILF(100/300) ÷ ILF(50/100)
  • Full coverage = standard liability + C + K (after the location, vehicle and deductible factors below)

Required PIP, uninsured motorist and similar coverages are included in the NAIC liability average for each state, so they are part of the liability figure.

4. Location factor (city and ZIP code)

Insurers price by territory. We approximate territory relativities with public data that correlates with claim frequency:

  • Density factor = (city density ÷ state reference density)0.055. The state reference is the population-weighted geometric-mean density of where residents live (every place of 5,000+ people plus the rest of the state as one rural area), from Census population estimates and Gazetteer land areas.
  • Commute factor = 1 + 0.009 × (city mean commute − state mean commute), in minutes, from the Census American Community Survey (ACS).
  • Location factor = density × commute, limited to 0.80–1.45. It applies to liability and collision.
  • Weather factor for comprehensive = 1 + 0.0025 × (county weather score − state reference), limited to 0.85–1.30. The weather score is the mean FEMA National Risk Index score for hail, tornado, hurricane, inland flooding, coastal flooding and strong wind; the state reference is the population-weighted mean across the cities we cover in that state.
  • ZIP code factor (relative to its city) = (ZIP density ÷ city density)0.050 × (1 + 0.008 × (ZIP commute − city commute)), limited to 0.85–1.20, using ACS data for ZIP Code Tabulation Areas.

5. Vehicle factor

Each model starts from a factor for its segment (1.00 = the average insured vehicle), separately for liability, collision and comprehensive:

SegmentLiabilityCollisionComprehensive
Subcompact car1.020.900.85
Compact car1.030.950.90
Midsize car1.021.000.95
Full-size car1.001.081.02
Sports car1.121.351.15
Luxury car0.981.401.25
Luxury sports car1.081.851.60
Small SUV0.950.920.88
Midsize SUV0.951.000.95
Large SUV1.001.101.12
Luxury SUV0.961.381.30
Minivan0.920.920.85
Pickup truck1.000.981.05
Heavy-duty pickup1.051.081.15
Van1.000.950.90
Wagon0.970.950.90

Adjustments: electric vehicles × 1.12 collision and × 1.05 comprehensive; models from luxury brands outside a luxury segment × 1.05 collision; NHTSA 5-star overall × 0.97 liability, 3 stars or fewer × 1.04; NICB national top-10 most-stolen × 1.25 comprehensive.

HLDI data takes priority. When the Highway Loss Data Institute publishes relative insurance losses for a model (100 = all-vehicle average), we blend them in with a weight of 0.85 for collision (collision index) and comprehensive (comprehensive index). Liability uses the average of the bodily-injury and property-damage-liability indices with a lighter weight of 0.60, because HLDI's bodily-injury figure measures claim frequency only, while its property-damage, collision and comprehensive figures measure overall losses. HLDI results cover a specific span of recent model years and are shown on each model page.

Vehicle age. Collision and comprehensive are reduced by 4.5% per model year older than the newest in our data (floor 50%), reflecting lower vehicle values.

6. Driver profile and policy choices

Age multipliers
16-19 (own policy)× 2.75
20-24× 1.60
25-29× 1.15
30-39× 1.00
40-49× 0.96
50-59× 0.92
60-69× 0.95
70+× 1.12
Driving record
Clean record× 1.00
One speeding ticket× 1.22
One at-fault accident× 1.45
DUI / DWI× 1.85
Annual mileage
Under 7,500 miles× 0.93
7,500-12,000 miles× 1.00
12,000-15,000 miles× 1.04
Over 15,000 miles× 1.09
Collision/comprehensive deductible
$250× 1.12
$500× 1.00
$1,000× 0.86
$1,500× 0.80
$2,000× 0.74

These multipliers are modeling assumptions based on the typical relativities seen in publicly filed rating plans and state insurance-department premium comparison guides. Every insurer's rating plan differs, sometimes a lot. An SR-22 filing adds a flat $25 in our model; the larger cost comes from the violation surcharge above.

7. Ranges and rounding

Results are rounded to the nearest $5. The range shown is ±18% around the central estimate to reflect normal differences between insurers. Monthly figures are the annual figure ÷ 12.

8. Worked example: Toyota Camry in Houston, TX

  1. Texas NAIC averages: liability $798, collision $529, comprehensive $400.
  2. Standard liability limits: $798 × 1.208 = $964.
  3. Houston density 3,730 vs. state reference 670; commute 27.2 vs. 26.7 min → location factor 1.104 (comprehensive 1.121 with weather).
  4. Toyota Camry vehicle factor: liability 1.095, collision 1.162, comprehensive 1.088 (HLDI-based).
  5. Driver in their 30s, clean record, average mileage, $500 deductibles: × 1.00.
  6. Result: about $2,330 a year ($1,910–$2,750), or $194 a month.

9. Home insurance estimates

We start from the NAIC average homeowners (HO-3) premium for the state and scale it by (home value ÷ state median home value)0.80, using Census ACS median values. Construction type and deductible multipliers are applied, and a hazard factor of 1 + 0.006 × (county hazard score − state reference) (limited to 0.70–1.80) uses FEMA National Risk Index scores for ten hazards. Flood and earthquake coverage are excluded, as they are from standard policies.

10. Limitations

  • We do not see your address, claims history, credit-based insurance score (where allowed), prior coverage or an insurer's own loss data — all of which matter.
  • NAIC averages lag behind current rates by the time needed to compile and publish them; premiums have generally risen since.
  • Location factors are proxies built from public data, not insurers' actual territory rates.
  • HLDI data describes specific recent model years; older model years may have had different loss experience.
  • We do not rank insurance companies. Verifiable company-by-company rate data isn't published consistently for every state, so any ranking would not meet our standards.

11. Updates and corrections

We update the data when sources publish new editions and review state rules against each state insurance department. Found an error? Tell us — see our editorial policy and the full list of data sources.