Ball Movement
Data Lineage
Column-level lineage from star schema to source endpoints
Generated dependency map
Use the generated lineage map for code-sourced coverage
This page is the exhaustive replay pulled from code metadata. Use it when you need total dependency coverage, then switch to curated lineage pages when you want the possession slowed down and explained.
trace code-sourced upstream and downstream coverage without guessing
keeps the dependency view aligned with transformer definitions
regenerate from lineage metadata instead of hand-editing
Start with the table-level view for blast radius
Use the top of the generated page to see whether a change affects one table or an entire downstream family.
Use column lineage only when the problem is local
If the breakage is one metric or key, the curated column-lineage page is the better explanatory cut.
Reconnect the replay to source scouting when needed
Once the dependency chain is known, jump upstream to endpoints if the real question is about the feed.
Quick scan
Read the generated replay in the right order
Start with blast radius, then slow the tape down to one field only when the problem is truly local. These routes keep the exhaustive map readable.
Use the generated table-level map first when you need to know how wide the dependency ripple runs before debugging one metric or field.
If the issue is one rename, metric, or key, move from the broad dependency map into the narrower column-level replay and naming guides.
Data Lineage
Table-Level Lineage
flowchart LR
BoxScoreAdvancedV3[BoxScoreAdvancedV3]
BoxScoreHustleV2[BoxScoreHustleV2]
BoxScoreMatchupsV3[BoxScoreMatchupsV3]
BoxScoreMiscV3[BoxScoreMiscV3]
BoxScorePlayerTrackV3[BoxScorePlayerTrackV3]
BoxScoreScoringV3[BoxScoreScoringV3]
BoxScoreSummaryV2[BoxScoreSummaryV2]
BoxScoreTraditionalV3[BoxScoreTraditionalV3]
CommonPlayerInfo[CommonPlayerInfo]
CommonTeamRoster[CommonTeamRoster]
CommonTeamYears[CommonTeamYears]
…Column-Level Lineage
See lineage.json for machine-readable column-level tracing.
Companion cuts
Switch from exhaustive replay to narrated debugging
These companion pages help you move from total coverage into the right debugging lens for the next question.
Zoom out to full table movement
Use the table replay when you need to understand full upstream and downstream impact.
Slow the tape down to one field
Best next stop when the problem is a rename, foreign key, or single metric rather than a whole table.
Reconnect the chain to the NBA API family
Jump upstream when the dependency answer still needs extractor or result-set context.
Next possession
Turn exhaustive lineage into practical debugging
The generated lineage page gives you code-sourced coverage. These curated follow-ups help you replay the dependency chain at the table level, slow it down to a single field, or reconnect it to the source feed.
Table Lineage
Trace full upstream and downstream table chains when a model or dashboard looks wrong.
Column Lineage
Replay one field at a time when the issue is a rename, metric, or foreign key.
Endpoints
Reconnect the lineage chain to the nba_api source families that started the possession.