builtbykev
a8de676756
The band gate was blocked for its `else` branch. It read:
candidate = F(raw)
served = inCertifiedBand(candidate) ? candidate : RAW
and above raw 0.60 the model is measured overconfident — holdout raw 0.80-0.90
predicts 0.843 and realizes 0.639. So "the calibrator is not supported here" was
being answered with a number already proven wrong. Unsupported calibration does
not make raw true.
Four candidates were adjudicated on ONE split — fit on the earliest 60% of
train, decide support on the last 40%, evaluate on a holdout that saw neither:
A low-param 80.2% coverage 0.24374 REFUTED — its extra region
(raw 0.80-0.90) certified on cert (err +0.040, n=55) and
refuted on holdout (served 0.754 vs observed 0.639), and it
leaves a hole at 0.70-0.80 while serving the island above it
B isotonic 91.3% coverage 0.24337 CERTIFIED, contiguous raw [0.50,0.80)
C empirical band 91.3% coverage 0.24335 REFUTED — refitted point-in-time on
current-model hits the realized rates INVERT in grade order
(B+ 0.593 < B 0.614 < C+ 0.623), so the served function steps
down at raw 0.78. Its shipped constants come from 3,417 props
pooled across four batter stats and do not reproduce here
D raw identity 43.1% coverage 0.24866 certifies raw 0.50-0.60 and only there
Raw is candidate D, not a fallback. It earns exactly one region (holdout error
+0.010 on n=1,316), which is why the law is "raw must earn its region" rather
than "raw is never true". B already covers that region, so no hybrid is built.
Above raw 0.80 nothing is certified and nothing is served. That is the region
where raw is most wrong, isotonic over-corrects (cert err -0.093) and its LODO
mapping at 0.95 has spread 0.180. 8.7% of holdout rows land there.
The registry did not need changing. `serves(stat, p)` already tested certified
bands against the RAW p_win — support in the input domain, the correct question —
and returned {serve:false, reason}. It never said "serve raw". The output-space
gate and the raw fallback were both invented downstream in calibrationService.
ACTIVATION IS OFF. PROBABILITY_CONTRACT_SHADOW defaults to 0, CALIBRATION_DEPLOYED
stays frozen empty, and every served field is byte-identical. This releases the
support first, which is the required order. The shadow records raw belief, the
candidate served value, the state, the estimator identity, and what EV/Kelly/VALUE
would be under the actionability law — into its own column, read by nothing.
Migration 051 was applied to production BEFORE retentionService named the column.
PostgREST builds a bulk insert from the first row's shape, so a key whose column
does not exist 400s the whole batch silently — that is how migration 038 took
retention down for three days.
The user-facing contradiction is NOT fixed here. A B+ still says "realized about
66%" beside a confidence of 84. Fixing that is activation, and activation costs
32% of VALUE flags and 46% of Kelly recommendations on the holdout.
Suite 401/401, 5,580 passed, 4 skipped, deterministic across three runs.
Teeth 23/23, each independently injected and restored byte-identically.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01CQJeAG8vcDoL5zkiaJyVb8
86 lines
2.6 KiB
JSON
86 lines
2.6 KiB
JSON
{
|
|
"table": "model_snapshots",
|
|
"generated_by": "scripts/generate-schema-contract.js",
|
|
"derived_from": "supabase/migrations/*.sql applied in order to a disposable postgres:15-alpine",
|
|
"note": "Insertable columns only (no generated/identity columns). This is the contract the retention writer must satisfy. Regenerate after any migration that touches model_snapshots; scripts/verify-schema-contract.js detects drift against a live database.",
|
|
"known_production_drift": {
|
|
"columns_in_production_not_in_migrations": [
|
|
"quarantine_reason",
|
|
"re_settled_at",
|
|
"settlement_source"
|
|
],
|
|
"explanation": "Present in production but not created by any committed migration - added out of band, same class as the migration-014 debt. The contract deliberately uses the MIGRATION-derived set, which is the stricter of the two: a writer that stays within it is valid against both.",
|
|
"classification": "PROD-ONLY COLUMN -> DRIFT WARNING / RECORDED DEBT (not release-authorized)"
|
|
},
|
|
"column_count": 65,
|
|
"columns": [
|
|
"actual_value",
|
|
"archetype",
|
|
"book",
|
|
"book_odds",
|
|
"canonical_event_id",
|
|
"captured_at",
|
|
"chain_shadow",
|
|
"change_type",
|
|
"claim_digest",
|
|
"claim_schema_version",
|
|
"code_sha",
|
|
"confidence",
|
|
"confidence_basis",
|
|
"created_at",
|
|
"cycle_hour_utc",
|
|
"devig_method",
|
|
"digest_algorithm_version",
|
|
"edge_pct",
|
|
"ev_pct",
|
|
"event_identity_method",
|
|
"event_identity_source",
|
|
"event_identity_version",
|
|
"event_occurrence",
|
|
"factor_inputs",
|
|
"fair_odds",
|
|
"fair_prob",
|
|
"features",
|
|
"game_date",
|
|
"game_id",
|
|
"grade",
|
|
"grade_11",
|
|
"id",
|
|
"line",
|
|
"lineage_action",
|
|
"lineage_state",
|
|
"lineage_version",
|
|
"model_version",
|
|
"opponent",
|
|
"outcome",
|
|
"over_odds",
|
|
"overround",
|
|
"p_win",
|
|
"player_key",
|
|
"player_name",
|
|
"probability_contract",
|
|
"projection",
|
|
"publication_id",
|
|
"published",
|
|
"published_at",
|
|
"read_id",
|
|
"read_natural_key",
|
|
"recaptures_id",
|
|
"refusal_reason",
|
|
"refused",
|
|
"revision_ordinal",
|
|
"settled_at",
|
|
"side",
|
|
"snapshot_id",
|
|
"sport",
|
|
"stat",
|
|
"supersedes_id",
|
|
"takeable",
|
|
"team",
|
|
"under_odds",
|
|
"value"
|
|
],
|
|
"authority": "RELEASE-AUTHORIZED model_snapshots INSERT CONTRACT",
|
|
"authority_note": "Derived from the migration chain, which is the release authority. Production is not. A column present in production but in no migration is DRIFT / RECORDED DEBT and is NOT release-authorized; it must never be copied into this contract from live schema."
|
|
}
|