diff --git a/specs/audit-data/mlb-grade-degradation.md b/specs/audit-data/mlb-grade-degradation.md new file mode 100644 index 0000000..dd2f142 --- /dev/null +++ b/specs/audit-data/mlb-grade-degradation.md @@ -0,0 +1,37 @@ +# MLB Grade Pipeline Degradation — Backend Finding (2026-07-17) + +Source: phone-audit P1-7 (broken edge board) + P2-9 (B grades at 45% confidence). +Diagnosed against LIVE `GET /api/snapshot/mlb` on 2026-07-17. This is a BACKEND +grading/feature-pipeline issue, NOT a frontend display bug. The frontend guards +(sane-edge cap, statAbbrev) are damage control, not the fix. + +## The evidence (25 live MLB grades) +- **projection == 0 for 9/25.** The model has no projection reference for a third + of props — a degraded feature path (the S58 insufficient-data refusal should + have caught these; instead they graded with projection 0). +- **edge_pct on a broken scale.** Distinct values {20, 60, 100, 140}. A real + prop-market edge is single-digit %, never past ~40. These are not a market edge + (frontend now shows them absent — slateAdapter EDGE_BOARD_SANE_MAX / EdgeCell). +- **grade <-> confidence mismatch on 25/25.** Every grade's letter disagrees with + its own surfaced `confidence` vs `grade_thresholds.json`: + - B shown at confidence 55 -> 55 is the B- band (55-59), not B (60-65). + - Systematic ~5-point (one sub-tier) gap on every prop. + The surfaced `confidence` is NOT the probability that derived the letter. + +## Likely root cause (to investigate, not yet fixed) +1. `confidence` field is post-`apply_data_sufficiency_modifier` (bayesian.py), + while the grade letter was assigned from the PRE-penalty prob -> the two + disagree by one sub-tier on every prop. Either surface the same value, or + re-letter from the penalized confidence. Decide which is the source of truth. +2. projection==0 for 9/25: the MLB feature/projection path is feeding 0 instead + of refusing (S58 `insufficient_data`). Trace `projectionFor` for MLB props; + projection 0 also breaks `computeEdge` (=> the 20/60/100/140 garbage). + +## Do NOT +- Do not re-letter or shift thresholds on the frontend (would hide the bug). +- Do not "fix" edge display by rescaling 140 -> 14 (guessing the scale is its own + fabrication). The number must come out of the pipeline correct. + +## Frontend already shipped (honest guards) +- Leaderboard stat labels: `web/src/lib/statAbbrev.js` (raw snake_case -> SB/ER/TB). +- Edge board: impossible |edge|>40 rendered absent AND excluded from ranking. diff --git a/tests/unit/statAbbrev.test.js b/tests/unit/statAbbrev.test.js new file mode 100644 index 0000000..8072875 --- /dev/null +++ b/tests/unit/statAbbrev.test.js @@ -0,0 +1,36 @@ +'use strict'; + +// P2-9 — the canonical SHORT stat label. Raw snake_case ("stolen_bases U0.5") +// on the leaderboard was the display bug; lock the abbreviations the audit named. + +const { statAbbrev, STAT_ABBREV } = require('../../web/src/lib/statAbbrev'); + +describe('statAbbrev', () => { + test('the audit-named MLB stats map to short labels', () => { + expect(statAbbrev('stolen_bases')).toBe('SB'); + expect(statAbbrev('earned_runs')).toBe('ER'); + expect(statAbbrev('total_bases')).toBe('TB'); + expect(statAbbrev('home_runs')).toBe('HR'); + expect(statAbbrev('strikeouts')).toBe('K'); + }); + + test('NBA stats too', () => { + expect(statAbbrev('points')).toBe('PTS'); + expect(statAbbrev('rebounds')).toBe('REB'); + expect(statAbbrev('assists')).toBe('AST'); + }); + + test('an unknown id upper-cases its words (never leaks raw snake_case)', () => { + expect(statAbbrev('some_new_stat')).toBe('SOME NEW STAT'); + expect(statAbbrev('some_new_stat')).not.toContain('_'); + }); + + test('null / empty is safe', () => { + expect(statAbbrev(null)).toBe(''); + expect(statAbbrev('')).toBe(''); + }); + + test('no label contains an underscore (all are real abbreviations)', () => { + for (const v of Object.values(STAT_ABBREV)) expect(v).not.toContain('_'); + }); +}); diff --git a/web/src/components/ExploreHub.tsx b/web/src/components/ExploreHub.tsx index e338431..c4bce60 100644 --- a/web/src/components/ExploreHub.tsx +++ b/web/src/components/ExploreHub.tsx @@ -5,6 +5,7 @@ import SportBadge from '@/components/vyndr/SportBadge'; import GradeBadge from '@/components/vyndr/GradeBadge'; import { playerHref } from '@/lib/playerHref'; import { dedupeLeaders } from '@/lib/playerGrouping'; +import { statAbbrev } from '@/lib/statAbbrev'; /** * ExploreHub (Session 42 leaders; Session 60 night2/C — THE AGGREGATOR). @@ -147,7 +148,7 @@ export default function ExploreHub() { {r.player} {r.team && {r.team}} -