A probability is served because evidence supports it, not because nothing else answered

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
2026-09-02 21:03:16 -04:00
parent 9dda9df132
commit a8de676756
10 changed files with 1359 additions and 1 deletions
@@ -193,6 +193,12 @@ function rowsFromSides(base, sides, ctx = {}) {
      // rows is silently dropped for the whole batch. Filled by
      // `mergeChainShadow` after enrichment; never read by anything served.
      chain_shadow: null,
+
+      // The PROBABILITY CONTRACT shadow — what the certified serving contract
+      // WOULD serve, beside what was actually served. Declared always, for the
+      // same first-row-shape reason as chain_shadow. Filled by
+      // `mergeProbabilityContract`; SHADOW ONLY, read by nothing served.
+      probability_contract: null,
    });
  }
  return rows;
@@ -825,6 +831,61 @@ function mergeEnrichment(rows, enrichedGrades) {
 * keeping, and a fabricated third of a triple would poison the adjudication this
 * column exists to enable.
 */
+/**
+ * PROBABILITY CONTRACT SHADOW.
+ *
+ * Records, per row: the raw model belief, what the certified contract would
+ * serve, the state, the estimator identity, and what EV/Kelly/VALUE would be
+ * under the actionability law. It writes to its own column and mutates nothing
+ * a user sees — `p_win`, `confidence`, `grade`, `ev_pct`, `value` and
+ * `takeable` on the row are untouched.
+ *
+ * The raw probability is ALWAYS carried, including on refusals: it is model
+ * evidence, and a row that records only the refusal cannot be re-adjudicated.
+ */
+function mergeProbabilityContract(rows, contract) {
+  if (!Array.isArray(rows) || !rows.length) return rows || [];
+  if (!contract || typeof contract.resolve !== 'function') return rows;
+  const pc = require('./model/probabilityContract');
+
+  return rows.map((r) => {
+    let res;
+    try {
+      res = contract.resolve({ model_version: r.model_version, p_win: numOrNull(r.p_win) });
+    } catch { return r; }
+    if (!res) return r;
+    let derived;
+    // The SIDE'S price. A retention row carries book_odds plus both sides'
+    // prices; using the wrong side would price the opposite bet.
+    const sideOdds = r.book_odds != null ? r.book_odds
+      : (String(r.side) === 'under' ? r.under_odds : r.over_odds);
+    try { derived = pc.derivedClaims(res, sideOdds); } catch { derived = null; }
+    return {
+      ...r,
+      probability_contract: {
+        raw_model_probability: res.raw_model_probability,
+        served_probability: res.served_probability,
+        probability_state: res.probability_state,
+        estimator_type: res.estimator_type,
+        estimator_version: res.estimator_version,
+        certification_version: res.certification_version,
+        model_version: res.model_version,
+        reason: res.reason,
+        fitted_through: contract.fitted_through || null,
+        fit_n: contract.fit_n || null,
+        derived: derived ? {
+          available: derived.available,
+          ev_pct: derived.ev_pct,
+          kelly_pct: derived.kelly ? derived.kelly.pct : null,
+          value: derived.value,
+        } : null,
+        // SHADOW. Nothing here has been served to anyone.
+        servable: false,
+      },
+    };
+  });
+}
+
 function mergeChainShadow(rows, shadow) {
  if (!Array.isArray(rows) || !rows.length) return rows || [];
  const byKey = shadow && shadow.byKey;
@@ -1183,6 +1244,7 @@ module.exports = {
  createCollector,
  mergeEnrichment,
  mergeChainShadow,
+  mergeProbabilityContract,
  persist,
  attachLineage,
  commitPublication,