CFB Betting Edge

2025 · Week 2 · Sun Belt · backtest — what the model predicted vs what happened, and vs the market
Week 2 backtest · Sun Belt · 9 games — what the model predicted vs what happened
Straight up
7–278%
ATS vs close
5–456%
Model margin MAE
15.1
Market margin MAE
14.0
→ market was closer (model can't beat the close ATS)
Results — model read vs the final score + how the closing-line lean held up
Model
pred Alabama 45.4 · actual Alabama 73
winner Alabama
ATS vs close
leaned Alabama -34.3Cover ✓
margin err: model 27.6 · mkt 38.7 · closer
Troy @ ClemsonFinal 1627
Model
pred Clemson 30.1 · actual Clemson 11
winner Clemson
ATS vs close
leaned Troy +30.3Cover ✓
margin err: model 19.1 · mkt 19.3 · closer
Model
pred USC 25.3 · actual USC 39
winner USC
ATS vs close
leaned Georgia Southern +28.8Miss ✗
margin err: model 13.7 · mkt 10.2 · mkt closer
Model
pred Marshall 24.3 · actual Missouri State 1
winner Marshall
ATS vs close
leaned Marshall -7.3Miss ✗
margin err: model 25.3 · mkt 8.3 · mkt closer
Model
pred Arkansas 13.8 · actual Arkansas 42
winner Arkansas
ATS vs close
leaned Arkansas State +24.0Miss ✗
margin err: model 28.2 · mkt 18.0 · mkt closer
Model
pred Louisville 11.1 · actual Louisville 14
winner Louisville
ATS vs close
leaned James Madison +15.3Cover ✓
margin err: model 2.9 · mkt 1.3 · mkt closer
Model
pred South Alabama 2.8 · actual Tulane 2
winner South Alabama
ATS vs close
leaned South Alabama +13.5Cover ✓
margin err: model 4.8 · mkt 11.5 · closer
Texas State @ UTSAFinal 4336
Model
pred Texas State 8.8 · actual Texas State 7
winner Texas State
ATS vs close
leaned Texas State +4.5Cover ✓
margin err: model 1.8 · mkt 11.5 · closer
Model
pred Memphis 9.7 · actual Memphis 22
winner Memphis
ATS vs close
leaned Georgia State +15.0Miss ✗
margin err: model 12.3 · mkt 7.0 · 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.