CFB Betting Edge

2025 · Week 5 · Mid-American · backtest — what the model predicted vs what happened, and vs the market
Week 5 backtest · Mid-American · 6 games — what the model predicted vs what happened
Straight up
5–183%
ATS vs close
2–3 · 1P40%
Model margin MAE
10.4
Market margin MAE
7.7
→ 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 Missouri 43.2 · actual Missouri 36
winner Missouri
ATS vs close
leaned Massachusetts +44.5Cover ✓
margin err: model 7.2 · mkt 8.5 · closer
Akron @ ToledoFinal 345
Model
pred Toledo 20.6 · actual Toledo 42
winner Toledo
ATS vs close
leaned Akron +21.5Miss ✗
margin err: model 21.4 · mkt 20.5 · mkt closer
Model
pred Ohio 6.6 · actual Ohio 15
winner Ohio
ATS vs close
leaned Bowling Green +7.5Miss ✗
margin err: model 8.4 · mkt 7.5 · mkt closer
UConn @ BuffaloFinal 2017
Model
pred UConn 1.7 · actual UConn 3
winner UConn
ATS vs close
leaned Buffalo +3.0Push
margin err: model 1.3 · mkt 0.0 · mkt closer
Model
pred Eastern Michigan 2.7 · actual Central Michigan 11
winner Eastern Michigan
ATS vs close
leaned Eastern Michigan +3.0Miss ✗
margin err: model 13.7 · mkt 8.0 · mkt closer
Model
pred San Diego State 13.5 · actual San Diego State 3
winner San Diego State
ATS vs close
leaned San Diego State -1.5Cover ✓
margin err: model 10.5 · mkt 1.5 · 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.