1,907 settled bets · 5.0% ROI · profitable in 5/5 seasons
Week 1 / Game totals
The total is
the argument.
A regulation-score model, a final-score model and the betting market all get a vote. The board publishes the answer only after separating the prediction from the price.
What changed
Let the under signal be an under signal.
This card is not forced to be directionally balanced. It currently finds 19 unders and 10 overs.
The core model targets the final total and recommends a side when it differs from the consensus by more than one point. A second CatBoost model targets regulation points, then converts its edge into an over probability using spread-binned isotonic calibration. That probability layer can add a recommendation where the core model passes; it never replaces a core side. On a Core model card, the point edge makes the call and probability is only a cross-check. On aProbability add card, calibrated probability makes the call.
The displayed book is the most favorable named-book total in the saved snapshot: lowest for an over, highest for an under. The model and probability are still evaluated against the provider consensus.
Sequential replay · 2021–2025
The policy earned the right to make the call.
Each historical season was forecast from earlier seasons. These are research replays, not a live betting record; the current card is the first forward slate for this exact combination.
426 bets where core passed · positive in 4/5 seasons · 98.7% season-cluster support
2,333 settled bets · 5.3% ROI · core always keeps priority
The current card
29 recommendations across 43 games.
UL Monroe at Mississippi State
- Final-total model
- 57.8
- Consensus total
- 56.0
- Model edge
- +1.8
- Regulation model
- 58.0
- Probability cross-check
- Supports · P(over) 57.2%
- Expected possessions
- 11.57
- Weather adjustment
- 0.0
Ball State at Ohio State
- Final-total model
- 57.8
- Consensus total
- 56.5
- Model edge
- +1.3
- Regulation model
- 59.1
- Probability cross-check
- Supports · P(over) 57.2%
- Expected possessions
- 11.04
- Weather adjustment
- 0.0
UTEP at Oklahoma
- Final-total model
- 51.6
- Consensus total
- 50.5
- Model edge
- +1.1
- Regulation model
- 52.9
- Probability cross-check
- Supports · P(over) 57.2%
- Expected possessions
- 11.90
- Weather adjustment
- 0.0
Coastal Carolina at West Virginia
- Final-total model
- 53.1
- Consensus total
- 56.3
- Model edge
- −3.1
- Regulation model
- 54.5
- Probability cross-check
- Supports · P(under) 53.8%
- Expected possessions
- 11.72
- Weather adjustment
- 0.0
Florida International at South Florida
- Final-total model
- 50.6
- Consensus total
- 53.8
- Model edge
- −3.1
- Regulation model
- 52.1
- Probability cross-check
- Supports · P(under) 53.1%
- Expected possessions
- 12.06
- Weather adjustment
- 0.0
Texas State at Texas
- Final-total model
- 57.6
- Consensus total
- 59.8
- Model edge
- −2.1
- Regulation model
- 58.5
- Probability cross-check
- Supports · P(under) 51.2%
- Expected possessions
- 11.55
- Weather adjustment
- 0.0
Florida Atlantic at Florida
- Final-total model
- 58.1
- Consensus total
- 59.5
- Model edge
- −1.4
- Regulation model
- 57.9
- Probability cross-check
- Supports · P(under) 51.2%
- Expected possessions
- 11.30
- Weather adjustment
- 0.0
Boise State at Oregon
- Final-total model
- 49.9
- Consensus total
- 51.8
- Model edge
- −1.9
- Regulation model
- 52.0
- Probability cross-check
- Supports · P(under) 51.2%
- Expected possessions
- 11.30
- Weather adjustment
- 0.0
North Texas at Indiana
- Final-total model
- 53.9
- Consensus total
- 55.5
- Model edge
- −1.6
- Regulation model
- 55.7
- Probability cross-check
- Supports · P(under) 51.2%
- Expected possessions
- 11.40
- Weather adjustment
- 0.0
Miami (OH) at Pittsburgh
- Final-total model
- 48.2
- Consensus total
- 49.5
- Model edge
- −1.3
- Regulation model
- 50.0
- Probability cross-check
- Supports · P(under) 51.2%
- Expected possessions
- 11.67
- Weather adjustment
- 0.0
Miami at Stanford
- Final-total model
- 46.2
- Consensus total
- 47.5
- Model edge
- −1.3
- Regulation model
- 47.0
- Probability cross-check
- Supports · P(under) 51.2%
- Expected possessions
- 11.34
- Weather adjustment
- −0.1
Washington State at Washington
- Final-total model
- 49.1
- Consensus total
- 50.8
- Model edge
- −1.6
- Regulation model
- 51.3
- Probability cross-check
- Supports · P(under) 51.1%
- Expected possessions
- 11.10
- Weather adjustment
- 0.0
UNLV at Hawai'i
- Final-total model
- 56.0
- Consensus total
- 58.5
- Model edge
- −2.5
- Regulation model
- 55.9
- Probability cross-check
- Supports · P(under) 50.8%
- Expected possessions
- 11.48
- Weather adjustment
- −0.7
Tulane at Duke
- Final-total model
- 49.2
- Consensus total
- 50.8
- Model edge
- −1.5
- Regulation model
- 49.6
- Probability cross-check
- Supports · P(under) 50.4%
- Expected possessions
- 11.29
- Weather adjustment
- 0.0
Boston College at Cincinnati
- Final-total model
- 50.3
- Consensus total
- 51.8
- Model edge
- −1.4
- Regulation model
- 51.5
- Probability cross-check
- Supports · P(under) 50.4%
- Expected possessions
- 11.23
- Weather adjustment
- 0.0
Liberty at James Madison
- Final-total model
- 49.9
- Consensus total
- 52.8
- Model edge
- −2.8
- Regulation model
- 51.6
- Probability cross-check
- Opposes · P(under) 49.7%
- Expected possessions
- 11.54
- Weather adjustment
- 0.0
Baylor vs. Auburn
- Final-total model
- 56.5
- Consensus total
- 59.3
- Model edge
- −2.7
- Regulation model
- 58.4
- Probability cross-check
- Opposes · P(under) 49.7%
- Expected possessions
- 11.43
- Weather adjustment
- 0.0
Louisville vs. Ole Miss
- Final-total model
- 53.6
- Consensus total
- 55.5
- Model edge
- −1.9
- Regulation model
- 54.7
- Probability cross-check
- Opposes · P(under) 49.7%
- Expected possessions
- 11.64
- Weather adjustment
- 0.0
San José State at Eastern Michigan
- Final-total model
- 52.7
- Consensus total
- 54.0
- Model edge
- −1.3
- Regulation model
- 53.0
- Probability cross-check
- Opposes · P(under) 49.7%
- Expected possessions
- 11.43
- Weather adjustment
- 0.0
Colorado at Georgia Tech
- Final-total model
- 49.7
- Consensus total
- 50.8
- Model edge
- −1.0
- Regulation model
- 50.4
- Probability cross-check
- Opposes · P(under) 49.7%
- Expected possessions
- 11.40
- Weather adjustment
- 0.0
Central Michigan at New Mexico
- Final-total model
- 45.5
- Consensus total
- 46.8
- Model edge
- −1.3
- Regulation model
- 47.0
- Probability cross-check
- Opposes · P(under) 47.7%
- Expected possessions
- 11.07
- Weather adjustment
- 0.0
Wyoming at Colorado State
- Final-total model
- 48.3
- Consensus total
- 47.5
- Model edge
- +0.8
- Regulation model
- 49.2
- Probability gate
- P(over) 60.8%
- Expected possessions
- 11.51
- Weather adjustment
- 0.0
Wisconsin vs. Notre Dame
- Final-total model
- 46.6
- Consensus total
- 47.5
- Model edge
- −0.9
- Regulation model
- 49.2
- Probability gate
- P(over) 57.2%
- Expected possessions
- 11.33
- Weather adjustment
- −0.0
Kent State at South Carolina
- Final-total model
- 54.0
- Consensus total
- 53.3
- Model edge
- +0.8
- Regulation model
- 55.0
- Probability gate
- P(over) 57.2%
- Expected possessions
- 11.58
- Weather adjustment
- 0.0
Northern Illinois at Iowa
- Final-total model
- 46.0
- Consensus total
- 45.5
- Model edge
- +0.6
- Regulation model
- 48.3
- Probability gate
- P(over) 57.2%
- Expected possessions
- 11.24
- Weather adjustment
- 0.0
Akron at Wake Forest
- Final-total model
- 48.1
- Consensus total
- 48.5
- Model edge
- −0.4
- Regulation model
- 49.7
- Probability gate
- P(over) 57.2%
- Expected possessions
- 11.83
- Weather adjustment
- 0.0
Western Michigan at Michigan
- Final-total model
- 47.5
- Consensus total
- 47.5
- Model edge
- +0.0
- Regulation model
- 48.9
- Probability gate
- P(over) 57.2%
- Expected possessions
- 11.18
- Weather adjustment
- 0.0
Oklahoma State at Tulsa
- Final-total model
- 59.5
- Consensus total
- 60.3
- Model edge
- −0.7
- Regulation model
- 61.5
- Probability gate
- P(over) 56.8%
- Expected possessions
- 11.93
- Weather adjustment
- 0.0
Arkansas State at Memphis
- Final-total model
- 54.7
- Consensus total
- 55.5
- Model edge
- −0.8
- Regulation model
- 53.9
- Probability gate
- P(under) 53.1%
- Expected possessions
- 11.66
- Weather adjustment
- 0.0
Every prediction
The 43-game board.
Passing is a prediction too. The table retains every lined FBS–FBS game, including the 14 games where neither layer clears its gate.
Read before betting
A prediction is not a price guarantee.
Totals can move quickly, and a half-point can change both the edge and the recommendation. This page reflects a saved market snapshot, assumes standard −110 pricing, and does not account for limits or availability. Available CFBD forecasts are treated as the realized game conditions and translated through the historical weather response. Forecasts can change, so recheck the named book, price and weather before acting.
See the Week 1 spread board