Skip to content
Home Insurance Index

Methodology

How the index is built

Public sources, checked against what they publish themselves, turned into bands and ranges. Every number names its source and year. November 2026 edition, data as of 2026-11-14.

Ranges below the state level

State and national figures are published exactly, with their source and year. Everything below the state level is a range: a county is “$1,400–$1,600,” never a single dollar figure, and a ZIP code is a band on the map. That's deliberate.

  • A range is honest about what the data can say. Survey estimates for one county carry real margins of error, and a range doesn't pretend otherwise.
  • The exact figures for your own area, and where your own policy ranks among homeowners in your ZIP, live at What My Neighbors Pay. They open up when you share what you pay. The public tables behind them are linked below; what you unlock is our cleaning and adjustments, the exact figures, the regulator comparison, and figures from neighbors who shared what they pay.
  • Rates people share with What My Neighbors Pay are never published here at ZIP level, and never as exact figures below the state level.

Each build runs an automated check that fails if any figure finer than a range appears on a county or ZIP page.

The color bands

Each map layer has seven bands with round-number breaks, set once in our data pipeline for the whole country so colors compare from coast to coast. The breaks are chosen to follow where people live (weighted by homeowners or policies, not by how many small counties there are) and to spread out the top end, where costs are changing fastest. Every county and ZIP gets the band its value falls in, from the same source everywhere. The legend shows each band's range, so color is never the only signal. Counties without a reliable value are hatched.

What homeowners report paying a year (median)
  1. Under $1,000
  2. $1,000–$1,200
  3. $1,200–$1,400
  4. $1,400–$1,700
  5. $1,700–$2,100
  6. $2,100–$2,800
  7. $2,800 or more
  8. Not enough data yet

U.S. Census Bureau, American Community Survey 2020–2024 (owner-reported; includes flood insurance).

Share of homes with flood insurance (NFIP)
  1. Under 0.5%
  2. 0.5%–1%
  3. 1%–2%
  4. 2%–5%
  5. 5%–10%
  6. 10%–20%
  7. 20% or more
  8. Not enough data yet

FEMA, NFIP policies in force; Census housing units. Data: September 2026.

Typical flood policy, per year (NFIP)
  1. Under $800
  2. $800–$900
  3. $900–$1,000
  4. $1,000–$1,100
  5. $1,100–$1,250
  6. $1,250–$1,500
  7. $1,500 or more
  8. Not enough data yet

FEMA, NFIP policies in force. Data: September 2026.

Official averages as ranges

Where a regulator publishes a county average, we show it on a fixed grid: $100-wide ranges under $1,000, $250-wide under $2,000, and $500-wide above that. Changes over time appear as coarse directions (“up 25%–50% since 2020”).

Shares and rates

The share of homeowners paying $4,000 or more appears in words (“between 1 in 5 and 1 in 3”). Nonrenewal rates appear as ranges: under 0.5%, 0.5%–1%, 1%–2%, 2%–3%, 3%–5%, 5%–10%, 10%–20%, and 20% or more.

The headline figure

Each state's headline is the best source it has: the regulator's own average premium where one exists, labeled as the official average; otherwise the median that homeowners report to the Census Bureau's American Community Survey. The two measure different things. A regulator average divides premium by policies; the Census figure is what owners say they pay, and it includes flood insurance.

Why our figures are lower than the ones in the news

Our national and Florida figures sit well below the numbers most news stories quote. Both can be right, because they measure different things:

  • Median, not average. Ours is the middle of what homeowners report paying. Most published figures are averages per policy, which a minority of very expensive homes pull up.
  • Every paying owner, condos included. The Census figure covers every owner who reports paying $100 or more, including condo owners, whose policies cover only the inside of the unit and cost less. Many published figures cover standard homeowners policies on houses only.
  • Reported, not quoted. Many rankings price one sample house (often $300,000 of dwelling coverage) with several companies. That's a quote, not what owners actually pay.
  • Timing. County colors use the 2020–2024 survey, a period when premiums rose fast. State and national headlines use the newer 2025 survey.

Florida is the clearest case: the state regulator's average homeowners policy with wind coverage is far above the median Florida owners report to the Census. We show both, labeled, and never blend them.

Sample sizes and margins of error

The Census figures are survey estimates, not counts. Each comes with a margin of error, and for a small county or ZIP it can be wide. We publish a county or ZIP figure only when at least 30 paying owners stand behind the estimate and its margin of error is no more than half the estimate; everything thinner is hatched. Flood figures need at least 10 priced policies. Even where we publish, a band is the honest reading: the true figure may sit near either edge.

How we grade states

We reviewed what every state's insurance regulator publishes about home insurance prices (November 2026):

  • A: average premiums by county, ZIP or rating territory, in a form we can load.
  • B: company-by-company sample-rate guides, or local averages only inside a PDF.
  • C: statewide averages only.
  • D: nothing useful.

This edition: 6 states graded A, 10 states graded B, 5 states graded C, 30 states graded D.

Which pages are indexed

A county page is offered to search engines only when it has at least two data layers that pass the margin-of-error check. Thinner pages still exist for people, marked noindex.

Sources and caveats

  1. U.S. Census Bureau, American Community Survey, table B25141

    State, county, ZIP (ZCTA)

    The yearly cost of fire, hazard and flood insurance that homeowners report, in cost ranges. Our national base layer and the colors on the home map (5-year estimates); the 1-year estimate is the state and national headline where no regulator figure exists.

    • People report it themselves; it isn't an insurer figure.
    • It includes flood insurance.
    • It covers owners only.
    • It comes in cost ranges, so medians are estimated within a range.
    • The top range is “$4,000 or more,” so where many homes pay that, the median breaks down. We report the share paying $4,000 or more as its own figure.
    • Margins of error are wide in small areas; those are hatched as “not enough data.”

    Source: U.S. Census Bureau, American Community Survey, table B25141.

  2. FEMA, OpenFEMA NFIP Redacted Policies (v3)

    State, county, ZIP

    Every National Flood Insurance Program policy in force: what flood policies cost, how many homes carry one, and FEMA's full price for the risk (Risk Rating 2.0).

    • Private flood policies aren't included.
    • Policies still below FEMA's full price for the risk can rise by up to 18% a year for primary homes, and up to 25% for many second homes and rentals.
    • Take-up is NFIP policies divided by Census housing units, so it's a share of all homes, not just those in flood zones.
    • This product uses the Federal Emergency Management Agency's OpenFEMA API, but is not endorsed by FEMA.

    This product uses the Federal Emergency Management Agency's OpenFEMA API, but is not endorsed by FEMA.

  3. U.S. Treasury, Federal Insurance Office (from NAIC market data), 2018–2022

    ZIP (used for context only)

    Insurer-reported HO-3 and HO-5 premiums, claims and nonrenewals by ZIP.

    • Covers 2018–2022 only.
    • Inflation-adjusted by Treasury.
    • Provided “as is” by Treasury.

    Source: U.S. Treasury, Federal Insurance Office, from NAIC Property & Casualty Market Intelligence data, 2018-2022. Provided "as is" by Treasury.

  4. U.S. Senate Budget Committee, “Next to Fall” (December 2024)

    County

    Nonrenewal rates by county, 2018–2023, from insurers' responses to the Committee.

    • 23 of 41 insurers asked responded (about 65% of the market).
    • Condo and co-op policies excluded.
    • We show the latest year as a range, and the direction since 2018.

    Source: U.S. Senate Budget Committee, "Next to Fall" (December 2024), insurer data 2018-2023.

  5. Texas Department of Insurance

    State, county

    Average homeowners premiums by county (Texas Statistical Plan), and HB 2067 counts of nonrenewals, cancellations and declinations.

    • TDI publishes average premiums by county only. Its ZIP-level data is policy counts and nonrenewals, with no prices.
    • The 2025 figures are preliminary.
    • County averages shown are for policies that include wind coverage. In coastal counties many homes leave wind out and buy it separately from the Texas Windstorm Insurance Association (TWIA).
    • No license is stated on TDI's pages; we credit TDI and treat them as public records.

    Source: Texas Department of Insurance (Texas Statistical Plan for Residential Risks; HB 2067 data on data.texas.gov).

  6. Florida Office of Insurance Regulation

    State, county

    County average premiums from the Property Insurance Stability Report.

    • MIR data is unaudited.
    • County averages shown are for policies that include wind coverage.

    Source: Florida Office of Insurance Regulation. MIR data is unaudited. Not endorsed by OIR.

  7. Alabama Department of Insurance, Property Insurance Clarity Act data

    State, county

    Premium, policies and losses by county and ZIP.

    • The average combines homeowners, condo, renters, dwelling fire and mobile home policies, so it runs lower than a homeowners-only figure.

    Source: Alabama Department of Insurance, Property Insurance Clarity Act data. Not endorsed by ALDOI.

  8. Massachusetts Division of Insurance

    State

    The annual home insurance report's Statistical Supplement.

    • The latest edition we could read is 2020; newer editions weren't downloadable.

    Source: Massachusetts Division of Insurance. Not endorsed by the Division.

  9. Missouri Department of Commerce and Insurance

    State, county

    The Homeowners Insurance Report, county averages.

    • Homeowners policies; written premium over exposures.

    Source: Missouri Department of Commerce and Insurance. Not endorsed by DCI.

  10. California Department of Insurance

    State, county

    The SB 824 wildfire data file: earned premium per dwelling-year by ZIP and county.

    • Earned premium divided by earned exposure, so it's a per-house-year average.
    • Data through 2023.

    Source: California Department of Insurance. Not endorsed by CDI.

House rules

  • Code computes every number. Words are written around them.
  • No savings promises and no carrier rankings.
  • Carrier names appear only with a regulator's own figures. Shared rates are never broken out by carrier.
  • We never sell data, and we never publish an address, a household, a driver or a policy.

Who runs this

What My Neighbors Pay, which publishes this index, is run by MicroEndeavor LLC. The index is free, and we never sell your data or share it with anyone unless you ask. The index is not an insurance company, and it never names a company to buy from.

Updates and corrections

The index is rebuilt when the data pipeline runs (monthly) and frozen as a press edition each quarter. Map data as of 2020–2024. Spotted an error? Tell us through What My Neighbors Pay.