MLB Model Validation — July 2026 Research Summary
MLB Model Validation — July 2026 Research Summary
> Signal Syndicate Research Library · Model Validation · Educational intelligence · Not betting advice
This is the first scheduled monthly validation edition (July 2026). Quant close for the public Route B preview remains the June 2026 research sample until a July slate batch is graded and frozen. Figures are historical estimates from graded research samples — not live P&L.
Prior edition: MLB Model Validation — June 2026.
Executive summary
Public board: ROUTE_B (v1 broad totals, |edge| ≤ 2.5 cap) stays in research preview. No shadow lane promoted.
June close (reporting baseline): n=105 capped plays, 50.48% win rate, −3.64% ROI across 19 June slates.
Early July: Monitoring window only — no new frozen Route B July sample in this edition. Performance tab and timeline JSON reflect the June close until quant refresh.
Verdict: MONITOR — honest reporting, gates unchanged, CLV tracking not yet live.
Public board (ROUTE_B)
| Rule | Behavior | ||
|---|---|---|---|
| Model | Team/Pitcher v1 broad totals scan | ||
| Edge cap | Plays with \ | model_edge\ | > **2.5 runs** excluded from public payload |
| Confidence | Edge **magnitude**, not win probability | ||
| P43 shadow | **Not** on the board |
June 2026 research sample (current public preview)
| Metric | Value | Note |
|---|---|---|
| Capped plays (Route B) | **105** | Research sample only |
| Win rate | **50.48%** | Not your bets |
| ROI | **−3.64%** | Not a profit claim |
| Archive context | 19 June slates · 177 uncapped graded plays | Cap removes strong-edge tail |
The edge cap is a trust fix from shadow analysis (uncapped strong-edge tail showed negative ROI). It is not presented as a performance optimization.
Shadow lane: P43 summer UNDER
| Metric | Value |
|---|---|
| Status | **MONITOR ONLY** |
| OOS bets | **18** (promotion review minimum: **75**) |
| Win rate | **50.0%** |
| ROI | **−4.55%** |
| On public board | **No** |
P43 remains below the 75-bet promotion gate. Small-sample volatility is expected — do not apply shadow outputs to the public board.
Shadow lane: Hits props V1 / V1.3
No new OOS pull this month. Status unchanged from June close:
| Year | V1 ROI @ edge 0.3 | n (edge-filtered) | Verdict |
|---|---|---|---|
| 2024 pilot | **+5.22%** | 1,504 | Baseline only |
| **2025 OOS** | **−22.79%** | 1,073 | Sign flip — do not promote |
| **2026 OOS** | **+9.87%** | 1,284 | Monitor — early-season slates |
V1.3 (85/15 Phase D blend) tied V1 on 2025/2026 OOS because Phase D batter cache was not refreshed for new seasons. No Signal Board wiring.
Total hits lane credits to date: 1,120 (390 pilot + 730 OOS).
K props — RETIRED
The pitcher strikeouts props lane remains RETIRED. Do not spend Odds credits on K. See: Why We Retired K Props (May–Jul).
CLV tracking
Closing Line Value (CLV) measures whether a bet beat the market close — a process metric, not win rate.
Weekly and monthly CLV arrays are empty in the current timeline. Live CLV tracking starts when the shadow forward log has 30+ graded plays with closing lines. We do not fabricate CLV figures.
Read: CLV Validation Study Memo.
P43 CLV sample size: 0 (insufficient closing-line pairs).
Methodology (July)
No production model-parameter changes this month. Platform Performance validation v2 shipped in June — CLV above the fold, backtest tabs by season, and monthly archive strip on `/performance`.
Grading rules, edge cap, and promotion gates are unchanged from the June memo.
August watchlist
1. Route B July slates — grade and freeze when quant schedules shadow refresh; update timeline + next monthly memo.
2. P43 OOS — accumulate toward 75 graded shadow bets before promotion review.
3. Hits V1.3 — Phase D cache refresh for 2025/2026 if founder approves credit spend.
4. CLV — begin populating weekly/monthly series once 30+ closing-line pairs exist.
Promotion gates (unchanged)
1. 75 OOS shadow bets minimum before promotion review
2. Walk-forward / sign-flip discipline
3. Multi-AI courtroom + founder sign-off
4. No production config changes without explicit YES
Compliance
For informational and educational purposes only. Not betting advice. Past research samples ≠ future results.
Research only · estimates only · not betting advice.