Our Methodology NBA-Lineups.com Data Desk

Opening-night fixtures use the NBA schedule release. The imported preseason schedule and roster entries use ESPN team feeds, labeled with season and retrieval date. The historical availability excerpt uses the official NBA injury PDF. Sources are kept separate so an old injury row cannot silently attach itself to a current roster.

The local SQLite snapshot stores team and game payloads with their checked date. A build renders those records into static pages. That local data layer is not a server scheduler and does not establish a live refresh cadence.

A projection requires documented recent starts, rotation context and availability information. Confirmation requires a named source containing the opening five. The clock reaching T−30 does not change a badge. No supported player projection or live confirmation pipeline is operating in this preview.

The current position view groups players using the source’s position labels and alphabetizes each group. It is not a coaching depth order. The planned ranking requires a last-ten-game sample of starts and minutes, with the actual sample size displayed. Broad G and F labels are not narrowed without evidence.

The intended cadence is a nightly schedule refresh, thirty-minute report checks from noon ET through the final tip-off and ten-minute starter checks from T−90 through the start. These jobs are not installed or running. Current pages show fixed snapshots.

A failed source should preserve its last known timestamp or show Source delayed. It must not turn missing data into zero injuries or confirmed starters. Daily archives retain their own date instead of being overwritten by the next slate.

Counts describe the displayed source scope. Archive counts refer to the player entries in the named official file; homepage samples remain explicitly selected rows. Imported fixture counts describe imported coverage rather than an asserted complete league slate.

Thin or incomplete daily pages stay noindex. A production publication check must verify factual coverage, metadata, source currency and route integrity before indexing. The current preview is blocked from indexing as a whole.

Learn about the publisher and the evidence behind each page.