Which jurisdictions Caseworth covers, and what coverage means

Axis one: the legal graph

The legal knowledge graph holds ingested statutes, and it is live in 26 states with 902,094 statutes indexed in total. The states are: Texas, Indiana, Washington, Alabama, Illinois, Maryland, South Dakota, Ohio, Virginia, Louisiana, Colorado, South Carolina, Connecticut, North Carolina, Utah, Maine, Florida, Arizona, Massachusetts, Missouri, Delaware, Wyoming, Vermont, Michigan, Wisconsin, Pennsylvania.

Depth is uneven between them, and the uneven part is worth stating plainly rather than averaging away. Texas alone accounts for 122,558 statutes, a substantial share of the total, because state codes differ enormously in size and structure. A state being live means its statutes are ingested and queryable, not that every state is represented to the same depth.

Where a state is not in the graph, Caseworth does not silently substitute another state’s law. The controlling authority for a matter is either present for that jurisdiction or it is not surfaced.

Axis two: the filing-deadline reference

The statute of limitations reference is the broadest thing Caseworth publishes and the only one that is genuinely nationwide. It covers all 51 US jurisdictions, meaning the 50 states and the District of Columbia, across 5 claim types: personal injury, car accident, slip and fall, medical malpractice, and wrongful death.

That is generated from a single dataset, so the pages and the sitemap cannot drift apart. It is a public reference tool available everywhere, including the jurisdictions where the consumer products are not offered, because a deadline is published law rather than an assessment of anyone’s situation.

Two naming points, because they cause confusion. Premises liability is not a separate category here; slip and fall is the premises category. Product liability is not a separate category either.

Axis three: where the consumer products are offered

The consumer chat and report products are available nationally with six exceptions: California, New York, Missouri, New Jersey, Tennessee, the District of Columbia.

The exclusions are deliberate and they are not about data. Each of those jurisdictions takes a position on the unauthorised practice of law, or on legal-services regulation generally, that we decided we would rather respect than test. Removing a product from six markets is an expensive decision to make voluntarily, and we made it because the alternative is asking users in those states to rely on something whose regulatory footing we were not confident about.

This is the axis that most often looks like a contradiction from outside. A state can be fully live in the legal graph and still not be a state where the consumer products are offered. Missouri is the clearest example: its statutes are ingested and it is live on the graph, and the consumer chat and report products are not offered there. Both facts are accurate and they are about different things.

What state law actually changes

Jurisdiction is not a formality in this subject area. It changes outcomes directly, which is why coverage is worth being precise about rather than approximating.

• Filing deadlines differ by state and by claim type, and claims against government entities frequently require a separate notice on a far shorter timetable.
• Shared-fault rules differ. Most states reduce recovery by the claimant’s share of fault; a small number bar recovery entirely once that share passes a threshold.
• Limits on non-economic damages exist in some states for some claim types, most commonly medical malpractice.
• Insurance requirements and minimum limits are set at state level, which affects what is practically collectible.
• Some states operate no-fault motor vehicle systems that change what can be claimed and when.
• Whether a comparable resolved matter is genuinely comparable depends first on whether it shares the jurisdiction.

How a state comes online

Adding a state is an ingestion and validation exercise rather than a switch. The state code is ingested and structured so that sections resolve individually, deadline rules are checked against the primary source rather than a secondary summary, comparable resolved matters are assembled to the point where a range can be reported with a count behind it, and the regulatory position on consumer legal-information products is reviewed for that jurisdiction.

The last of those is why graph coverage and product availability are tracked separately. They are answered by different work and they can land at different times.

We publish which states are live rather than which are planned. A roadmap with dates on it would be a commitment about ingestion timelines that depend on how a given state publishes its code, and the honest position is that we do not know the order in advance well enough to publish one. The coverage map is generated from the same dataset the product reads, so it is current by construction rather than by someone remembering to update it.

What we do not claim

• That coverage is uniform in depth across live states. It is not, and the statute counts show it.
• That a comparable-outcomes range exists for every claim type in every live state. Benchmark data is thinner in some combinations than others, and where it is thin the count behind the range shows that.
• That Caseworth is a law firm or provides legal advice in any jurisdiction. It is a software platform, and what it produces is educational legal information.
• That a jurisdiction being covered means an outcome can be predicted there. A benchmark describes resolved matters; it does not forecast a particular one.