About the data

Range Insights publishes aggregates of data Range Lab already collects to run its tools: rental listings gathered daily for the Rental Search Tool (RST), and the community-resource directory maintained in Pathfinder and served by Rangefinder. A snapshot runs every day; each page shows the date of the snapshot it reflects.

Housing market

• What a listing is. A rental unit advertised on a listing site or property-manager website that RST collects from. A unit advertised on several sites is usually merged into one record, though some duplicates remain. Listing counts match the county counts on rentalsearchtool.com: a listing is placed in a county by its city, using RST's county list plus a few towns it does not yet include. RV spaces are counted as listings but excluded from every rent, unit-size and FMR figure; so is an RV space or a mobile or manufactured home lot listed without bedrooms. Per-bed leases (student housing that prices one bedroom of a shared unit) are handled the same way: a listing with two or more bedrooms counts as per-bed when it says so (“by the bed,” “individual leases”) or lists under 150 sq ft per bedroom. Commercial space (an office, shop or warehouse listed with no bedrooms, or a listing that takes “commercial inquiries only”) and garages or storage units listed as rentals are handled the same way. A listing with no bedroom count is treated as a studio when it calls itself one and gives no bedroom count anywhere; otherwise it counts toward the all-sizes total but not toward any unit size.
• Active. Still advertised. A listing counts as off the market once it has been missing from its source for five days, so removals appear with a short lag.
• Asking rent, not contract rent. Rents are what landlords advertise. Signed leases, renewals and units that never get advertised are not included. Asking rents usually run above what existing tenants pay.
• Rent statistics use listings with an asking rent between $200 and $15,000 a month. A median or range is published only when at least 5 priced listings back it; smaller groups show counts only. ZIP × size rows with fewer than 3 listings are omitted.
• New listings are those first seen in the last seven days. When a new source is added, hundreds of existing listings appear on the same day; a day with more than 40 first sightings and more than 8× the county's usual daily count is treated as a bulk load and excluded. “New” counts are withheld for the first 14 days after a county starts being tracked.
• Price cuts are listings whose asking rent dropped in the last seven days. Changes that look like a listing switching between sale and rent prices are ignored.
• Coverage. Collection is county by county. Listings on sites RST does not collect from are missing, and small landlords who advertise offline are under-represented. Counts are best read as a trend within a county, not as a census of vacant units.
• History before tracking began is rebuilt from each listing's first and last sighting and its recorded price changes, and is marked “rebuilt from listing history” wherever it appears.

HUD Fair Market Rent and income limits

• Fair Market Rents (FMR) come from HUD for the county, by unit size, for the fiscal year the date falls in (HUD fiscal years start October 1). If that year is not loaded yet, the most recent earlier year is used and labeled.
• FMR is a gross rent, including utilities. Advertised rents often exclude some utilities, so “at or below FMR” somewhat overstates what a voucher can reach once a utility allowance is applied.
• Voucher payment standards are set by each housing authority, usually 90–110% of FMR. Small Area FMRs, where they apply, can differ from the county figure.
• Affordable rent at 30%, 50% and 80% of area median income uses HUD income limits and HUD's sizing convention of 1.5 people per bedroom (one for a studio), at 30% of monthly income.

Community services

• Counts are public, active listings in the Range Lab directory: exactly what Rangefinder shows. Draft, private and archived records are never counted.
• A listing counts toward a county when its service area is that county. Statewide programs have no county and are counted on the state page only.
• Essential services are twelve categories every county is expected to have locally: food, shelter, housing assistance, medical, mental health, addiction, domestic violence, legal aid, employment, transportation, childcare and utilities. “None listed” means none in the directory, which can mean a real gap or a listing we do not have yet.
• Hours-based shares (evening, weekend) are out of listings with structured hours. “Reviewed” means a staff review within the last 90 days; many listings, especially those imported from 211 directories, have not been individually reviewed.
• Newly listed means a listing became public in the directory that week. When a large import happens, many existing services become “new” at once; this reflects the directory, not services opening.
• Per-resident rates (services per 10,000 residents, “one for every N residents”) use Census ACS 2024 5-year total population (table B01003), the same figure shown as the county's population.

Affordability tab

• Affordable listings per 100 low-income renter households divides today's priced listings whose asking rent is at or below 30% of the income limit for a household at 50% of area median income (HUD income limits, sized at 1.5 people per bedroom) by the number of renter households at or below 50% of AMI in HUD's CHAS tabulation. The extremely-low-income version uses the 30% AMI limit and households at or below 30% of AMI. The numerator is today's; the denominator is a multi-year estimate, so read the figure as “how much of the market is within reach”, not as a vacancy count.
• Asking rent vs. what renters pay compares today's median asking rent with the Census ACS median gross rent, which covers renters already housed and includes utilities. The gap is what a household faces when it has to move.
• Rent burden is the share of renters for whom the Census computes a rent-to-income ratio who pay 30% or more (and 50% or more) of income in rent. CHAS reports the same by income band.
• Census 5-year estimates carry 90% margins of error; an estimate whose margin exceeds half its value is not printed. CHAS describes a five-year period ending two to three years before publication.

The other county tabs

• Economy & work. The housing wage is the full-time hourly wage at which today's median 2-bedroom asking rent is 30% of income (rent × 12 ÷ 0.30 ÷ 2,080 hours), the method the National Low Income Housing Coalition uses in Out of Reach. “Hours per week at the average wage” divides that by the county's average weekly wage from BLS. Unemployment is not seasonally adjusted and the latest month is preliminary.
• Housing supply. Permits are housing units authorized, from the Census Building Permits Survey, which has covered every county monthly since 2022; a few counties have no permit-issuing jurisdiction and show nothing. Zillow's rent index (ZORI) is a smoothed measure of asking rents and is not published for small counties; we show our own monthly average beside it so readers can see where the two agree. LIHTC units come from HUD's database of tax-credit properties placed in service.
• Homelessness & subsidized housing. HUD's Point-in-Time and Housing Inventory counts are per Continuum of Care, which can span several counties; the page names the CoC and shows its totals, never a county share we did not measure. “Listings a voucher can reach” divides today's listings at or below Fair Market Rent by the number of households using Housing Choice Vouchers in the county (HUD Picture of Subsidized Households). Waitlist figures count RST's income-restricted property listings; names are never published.
• Community. Census population groups (veterans, residents 65+, people with a disability, children under 5, Spanish speakers who speak English less than “very well”) beside the public directory listings whose category serves that group. “Residents per listing” is a coverage ratio, not a measure of capacity.
• Health & food. CDC PLACES prevalences are model-based small-area estimates from the Behavioral Risk Factor Surveillance System, not county surveys; “adults per listing” multiplies each prevalence by the Census ACS 2024 5-year population it describes: adults 18 and over, or adults 18 to 64 for the uninsured rate. The two sources are different products and years, so treat these as rough coverage ratios. Housing- and food-insecurity measures exist only for states that ran CDC's social-needs questions. HRSA shortage designations count only area designations (whole county, subdivision or tract), not single facilities.

Census figures

Every Census number on the site comes from one product and vintage, the U.S. Census Bureau's American Community Survey 5-year estimates for 2020–2024 (Census ACS 2024 5-year), from the detailed tables. Figures from different vintages are never combined: a share whose parts come from different years is not published.

A percentage is always of the group's own Census universe, which is often not total population. Veterans are counted among civilians 18 and over, disability among the civilian noninstitutionalized population, and language among residents 5 and over.

Population is table B01003 (total population). Children under 5, residents 65 and over and adult populations are from table B01001 (sex by age). Where Census publishes no margin of error because the estimate is controlled to population estimates, none is shown.

Public data sources

Published statistics joined to Range Lab's own data, with the period each currently covers.

• American Community Survey 5-year estimates (2024), U.S. Census Bureau — through 2024
  Tables B01001, B01003, B17001, B18101, B19013, B21001, B25003, B25004, B25064, B25070, B25119, C16001. 90% margins of error kept; derived sums use root-sum-square. Shares are computed within each table's own universe (lib/census.ts).
• Building Permits Survey (county, monthly), U.S. Census Bureau — through 2026-08
  New privately owned housing units authorized by building permits, by building size (1, 2–4, 5+ units). Includes imputed values for non-reporting permit offices. Not seasonally adjusted.
• Comprehensive Housing Affordability Strategy (CHAS), U.S. Department of Housing and Urban Development — through 2018-2022
  Custom tabulation of ACS 5-year data: households by share of HUD area median family income, tenure, and housing cost burden.
• Health Professional Shortage Areas (HPSA), Health Resources and Services Administration — through 2026
  County is flagged when any geographic or population HPSA component (county, subdivision or tract) with status Designated or Proposed For Withdrawal lies in it. Facility-only designations (FQHCs, clinics, prisons) are excluded. Score is HRSA's 0–25 (dental 0–26) shortage score, the maximum among the county's area HPSAs of that discipline (hpsa_<discipline>_score; null when not designated).
• Local Area Unemployment Statistics, U.S. Bureau of Labor Statistics — through 2026-07
  County unemployment rate, unemployed persons and labor force; not seasonally adjusted. Preliminary for the latest month.
• Low-Income Housing Tax Credit database (through 2024), U.S. Department of Housing and Urban Development — through 2024
  Projects and units by county as reported by state housing finance agencies, including projects no longer monitored. "Last 10y" counts placed-in-service years 2015–2024; projects with an unknown year are in the totals only.
• Occupational Employment and Wage Statistics, U.S. Bureau of Labor Statistics — through 2025
  Employment and hourly wage percentiles by major occupation group, May reference month. Published for metro and nonmetropolitan areas, not counties; county pages show their area's figures, labelled with the area name.
• Picture of Subsidized Households (2025), U.S. Department of Housing and Urban Development — through 2025
  County totals by program: units under contract, occupancy, average household income. Project-based sums every multifamily program (Section 8, Mod Rehab, 236/BMIR, 202/811). Cells HUD suppresses for small counts are null.
• PLACES: Local Data for Better Health (2025 release), Centers for Disease Control and Prevention — through 2025
  Model-based county estimates from BRFSS 2023; crude prevalence, percent of adults. Dataset swc5-untb.
• Point-in-Time count and Housing Inventory Count, by Continuum of Care, U.S. Department of Housing and Urban Development — through 2025
  Reported per Continuum of Care (county_fips = "CoC:" + code); join through coc_counties. PIT is a one-night January count; HIC is year-round beds by program type.
• Quarterly Census of Employment and Wages, U.S. Bureau of Labor Statistics — through 2026-Q1
  County totals, all ownerships and industries: average weekly wage, employment, establishments.
• Zillow Home Value Index (ZHVI), county, Zillow Research — through 2026-08 (Data from Zillow Research (zillow.com/research/data))
  All homes (single-family and condo/co-op), mid-tier (35th–65th percentile), smoothed, seasonally adjusted. Typical home value in current dollars.
• Zillow Observed Rent Index (ZORI), county, Zillow Research — through 2026-08 (Data from Zillow Research (zillow.com/research/data))
  All homes plus multifamily, smoothed, seasonally adjusted. Typical asking rent for the 35th–65th percentile of the market, in current dollars.
• Fair Market Rents and Income Limits, U.S. Department of Housing and Urban Development — annual.

Privacy

Everything on this site is an aggregate. No page shows an individual renter, landlord, listing address, or search. Range Lab's case-management data (Pathfinder clients, households and referrals) is not used here.

Citing Range Insights

“Range Insights, Range Lab. [County], [State]. Data as of [date on the page]. insights.rangelab.io.” Every table can be downloaded as CSV. Questions or corrections: hello@rangelab.io.