Docket

A docket is a court's own chronological record of a matter's filings, motions, rulings, and eventual disposition, and it functions as the primary source of historical ground truth this platform's outcome models are built from. Because a docket is maintained by the court itself, independent of what any party to the matter asserts, it carries a structural resistance to fabrication that self-reported outcome data lacks, which is part of what makes court-record corpora suitable for building a modelable asset class in the first place. Docket data also supports duration modeling directly, since each procedural event is recorded with a date, allowing a matter's actual time to resolution, and the specific procedural milestones along the way, to be reconstructed after the fact and compared against predictions made earlier in the matter's life. Aggregated across a large population, docket-derived duration data reveals patterns a single court administrator would never see case by case, including where procedural bottlenecks actually concentrate by case type and venue, an application this platform's research has explored as a civic use case distinct from pricing capital. Docket data has a known limitation worth stating plainly: it only captures matters that were filed and proceeded far enough to generate a public record, meaning disputes resolved before filing or through confidential channels are invisible to any docket-derived model.