Data · Markets
Legal market data by place and practice
One place and one practice area at a time: what the courts and agencies count, how many law offices serve it, how many people need help in Spanish, and the ad rules that apply. Every figure is a record from one of our open datasets, with its source, version and DOI.
What can it answer?
The markets layer joins our datasets by place (state, county, federal court district, metro area) and by practice area. It answers six kinds of question, as pages, downloads and tools for AI assistants:
- Resolve: turn "Harris County, Texas, truck accidents" into the keys every dataset uses (county FIPS code, federal district, metro area, practice id).
- Market profile: every figure for one place and practice: demand, lawyer supply, Spanish-language demand, prices and, where a practice matches, our benchmarks.
- Compare markets: the same figures for several places side by side, with rates per resident where the population is known.
- Supply gap: where demand per law office is high, with both counts shown.
- Demand trend: which counts are rising or falling, and where.
- Rules check: the ad rules, state laws on drug and device injury ads, filing duties, referral rules, limitation periods and fee caps for a state and practice.
Questions it answers
- Which California counties have rising personal injury filings? The trend section of personal injury in California, or the demand_trend tool with place CA, kind "filings" and direction "rising": each county's superior-court injury filings in the last two fiscal years, from the California courts dataset.
- Where is Spanish-language legal demand underserved? The supply_gap tool with state US and metric "spanish": states ranked by Spanish speakers with limited English per Spanish-speaking lawyer, with each margin of error; with a state code, its counties per law office.
- What are the advertising restrictions for a drug-injury campaign in Texas? The rules_check tool with state TX and practice drug-device-injury: the Texas ad rules, the Texas statute on drug and device injury ads, the filing duty, the referral rule, the limitation periods and recent bar opinions, each quoted with its source.
- Which markets should an employment firm research next? The supply_gap tool with state US and practice employment-law ranks the states by EEOC discrimination charges per law office; a state's page lists every practice, and employment law in Texas shows the full profile.
What goes in
The datasets the practice taxonomy links to at least one practice area, plus three context datasets the layer links itself (marked "context" below and in every answer). Versions and DOIs come from the data catalog when this page is built.
The practice links come from the crosswalks dataset (version 1.1.0), which maps every practice area to the datasets that measure it.
How a figure is joined
- One record, copied: each figure is one value from one record of the dataset named with it, with that record's id, period, status and last-checked date. Nothing is computed except the ratios and rates below, and those show their inputs.
- Place: a figure keeps the geography its dataset uses. County figures show at county level; datasets with no county field (federal court districts, immigration courts, Social Security hearing offices, metro areas) show at their own level and are labelled with it.
- Practice: the crosswalks dataset says which datasets measure a practice. Inside a dataset, each measure is tagged with the practices it counts (fatal motorcycle crashes count for motorcycle accidents and for personal injury, not for family law).
- Gaps: a linked dataset with no figure for the place is listed as a gap, and a value the source withholds stays empty with its status. We never estimate or fill them in.
- Supply gap: one demand count divided by one supply count (private law offices, or VA-accredited attorneys and agents for veterans' disability), rounded to two decimals. Places with no law office are listed apart. There is no blended score.
- Per resident: a count divided by the residents in the Census survey (legal-need dataset), per hundred thousand, with both inputs.
- Trend: rising or falling is the sign of the change between the last two periods a source publishes, with no threshold and no smoothing. County series keep their last two periods; the full series is in the record.
Limits
- A count shows what a court or agency recorded: filings, crashes, complaints, inspections, offices. It is not the number of people who need or will hire a lawyer.
- Sources cover different years and months; every figure states its own period. Compare like with like.
- Survey figures (Census) come with a margin of error; small places can have large ones or be withheld.
- State court filings are in the data for California and Texas so far; elsewhere, federal and agency counts stand in, and the page says so under gaps.
- Federal court districts and immigration courts do not follow county lines.
- Rules are summaries of official texts with their quotes and links. They are not legal advice and never say an ad is allowed.
- Our benchmarks appear only where a practice matches: Probative benchmarks, 2025–2026, aggregated, rounded, and small groups suppressed.
Pages, downloads and AI tools
Every state has a page listing its practice areas, and every practice area has a market page with its sources, a CSV and JSON download of the rows behind it, and a citation line. Pages with too few public demand counts are kept out of search results.
AI assistants can ask the same questions through our free MCP server (no account): https://probativegrowth.com/mcp/markets. Its tools are resolve, market_profile, compare_markets, supply_gap, demand_trend and rules_check; every answer carries the dataset, version, DOI and record id of each number.
The files behind the layer: markets index (states, counties, practices, dataset versions) and one file per state under /data/v1/markets/places/. The market pages sitemap lists every indexed page.
Every state
- Alabama
- Alaska
- Arizona
- Arkansas
- California
- Colorado
- Connecticut
- Delaware
- District of Columbia
- Florida
- Georgia
- Hawaii
- Idaho
- Illinois
- Indiana
- Iowa
- Kansas
- Kentucky
- Louisiana
- Maine
- Maryland
- Massachusetts
- Michigan
- Minnesota
- Mississippi
- Missouri
- Montana
- Nebraska
- Nevada
- New Hampshire
- New Jersey
- New Mexico
- New York
- North Carolina
- North Dakota
- Ohio
- Oklahoma
- Oregon
- Pennsylvania
- Rhode Island
- South Carolina
- South Dakota
- Tennessee
- Texas
- Utah
- Vermont
- Virginia
- Washington
- West Virginia
- Wisconsin
- Wyoming
How to cite
The layer adds no data of its own, so cite the datasets each answer names: every answer and page ends with a credit line listing them with their versions and DOIs, under CC BY 4.0. To report an error, see the corrections log.