Opening the research record…
Opening the research record…
Exploratory pilot · Week 1 settled
A frozen editorial heuristic meets a market benchmark—only after every line-blind forecast is hashed and committed.
Experimental · not guaranteed betting advice
A fair-line midpoint difference shows direction, not a calibrated cover probability or verified betting edge. The v0.2 probability framework remains research-only.
Week 1 · 2026
Every Week 1 matchup between two Power Four programs or Notre Dame. No game was selected because of its spread.
Sep 4, 2026, 12:00 AM UTC
Sep 5, 2026, 1:00 AM UTC
Sep 5, 2026, 7:30 PM UTC
Sep 5, 2026, 11:30 PM UTC
Sep 6, 2026, 2:30 AM UTC
Sep 6, 2026, 11:30 PM UTC
Sep 6, 2026, 11:30 PM UTC
Sep 7, 2026, 11:30 PM UTC
Frozen v0.1 decision rules
The edge is the blind fair home margin minus the signed market home margin. A market inside the complete fair range can never exceed a lean. These official calls remain unchanged while the key-number-aware v0.2 framework is reviewed.
Read the forecast methodology| Classification | Rule | Standard risk |
|---|---|---|
| No position | Under 1.5, genuine toss-up, or integrity veto | 0 |
| Lean | 1.5–2.9, or any edge with market inside range | 0.55 to win 0.50 |
| Large lean | 3.0–4.9 and market outside range | 0.825 to win 0.75 |
| Play | 5.0+ and market outside range | 1.10 to win 1.00 |