CFB Betting Edge

2025 · Week 8 · Sun Belt · backtest — what the model predicted vs what happened, and vs the market
Week 8 backtest · Sun Belt · 7 games — what the model predicted vs what happened
Straight up
4–357%
ATS vs close
4–357%
Model margin MAE
12.9
Market margin MAE
14.8
→ model's margins were closer than the market
Results — model read vs the final score + how the closing-line lean held up
Model
pred Georgia Southern 14.9 · actual Georgia Southern 17
winner Georgia Southern
ATS vs close
leaned Georgia Southern -6.5Cover ✓
margin err: model 2.1 · mkt 10.5 · closer
Model
pred App State 10.6 · actual Coastal Carolina 8
winner App State
ATS vs close
leaned App State -10.5Miss ✗
margin err: model 18.6 · mkt 18.5 · mkt closer
Model
pred James Madison 4.1 · actual James Madison 36
winner James Madison
ATS vs close
leaned James Madison -3.0Cover ✓
margin err: model 31.9 · mkt 33.0 · closer
Model
pred South Alabama 3.7 · actual Arkansas State 1
winner South Alabama
ATS vs close
leaned Arkansas State +8.5Cover ✓
margin err: model 4.7 · mkt 9.5 · closer
Model
pred Marshall 3.5 · actual Marshall 3
winner Marshall
ATS vs close
leaned Marshall +3.0Cover ✓
margin err: model 0.5 · mkt 6.0 · closer
Model
pred Louisiana 0.3 · actual Southern Miss 12
winner Louisiana
ATS vs close
leaned Louisiana +4.5Miss ✗
margin err: model 12.3 · mkt 7.5 · mkt closer
Troy @ UL MonroeFinal 3714
Model
pred Troy 2.6 · actual Troy 23
winner Troy
ATS vs close
leaned UL Monroe +4.5Miss ✗
margin err: model 20.4 · mkt 18.5 · mkt closer
The model beats SP+ but not the closing line (ATS ≈ break-even). “Model−Mkt” is the gap between our projected home margin and the market's implied margin — a read on disagreement, not a profitable edge. Upcoming games have no closing line yet, so the market number marked ~ is the consensus midpoint of the best posted spreads across books.