Completeness Scorecard
One health number per vertical, decomposed into five weighted dimensions. The score is a prototype heuristic — the weights below are ours to argue about — but every input (spine, freshness, known gaps, seasons) is the real 2026-08-10 snapshot. HLR scores lower on purpose: its freshness is unmonitored and its results table is missing score/points.
Portfolio average
Mean across 4 curated verticals, out of 100 possible points.
3 of 4 sit ≥ 85; HLR pulls the mean down.
The rubric (prototype weights)
| Dimension | Max | How it's earned |
|---|---|---|
| Spine coverage | 40 | spinePresent / 6 backbone tables × 40 |
| Freshness | 20 | current 20 · vendor_unknown 10 · behind 5 |
| Column uniformity | 20 | points present & no known gaps → 20; else −8 per gap |
| Season history | 10 | seasons ≥ 3 → 10; else proportional |
| Archetype depth | 10 | estimated 8 — depth tables exist but not all loaded |
| Total | 100 | Bands: ≥ 85 healthy · ≥ 65 watch · below at risk. |
Supercross / Motocross
asd_smx
No known column or freshness gaps — full marks on uniformity and freshness.
World Surf League
asd_wsl
No known column or freshness gaps — full marks on uniformity and freshness.
Major Arena Soccer League
asd_masl
No known column or freshness gaps — full marks on uniformity and freshness.
High Limit Racing
asd_hlr
2 known gaps scored against
Why the score is a prototype
The five weights above are a heuristic we picked for this prototype — reasonable people would tune them. What is not negotiable is the inputs: spine presence, freshness status, known gaps, and season counts all come verbatim from the snapshot.
Archetype depth is scored as an estimate (8/10) for every sport — depth tables exist per archetype, but not all are loaded into the product spine, so we neither claim full credit nor zero.
Pulled with AWS_PROFILE=asd against Redshift Serverless (us-west-1 / default-workgroup / dev).
Verify a line yourself
Every earned point traces to a row you can count. Run the same queries the score reads from and check the freshness and gap inputs first-hand.
Open the Query Playground