CFB Betting Edge

2025 · Week 6 · American Athletic · backtest — what the model predicted vs what happened, and vs the market
Week 6 backtest · American Athletic · 6 games — what the model predicted vs what happened
Straight up
3–350%
ATS vs close
2–3 · 1P40%
Model margin MAE
14.5
Market margin MAE
10.3
→ market was closer (model can't beat the close ATS)
Results — model read vs the final score + how the closing-line lean held up
Tulsa @ MemphisFinal 745
Model
pred Memphis 23.5 · actual Memphis 38
winner Memphis
ATS vs close
leaned Memphis -21.0Cover ✓
margin err: model 14.5 · mkt 17.0 · closer
Model
pred South Florida 17.5 · actual South Florida 28
winner South Florida
ATS vs close
leaned Charlotte +28.0Push
margin err: model 10.5 · mkt 0.0 · mkt closer
Air Force @ NavyFinal 3134
Model
pred Navy 15.8 · actual Navy 3
winner Navy
ATS vs close
leaned Navy -13.5Miss ✗
margin err: model 12.8 · mkt 10.5 · mkt closer
Model
pred Rice 13.3 · actual Florida Atlantic 6
winner Rice
ATS vs close
leaned Rice -4.5Miss ✗
margin err: model 19.3 · mkt 10.5 · mkt closer
Army @ UABFinal 3113
Model
pred UAB 1.9 · actual Army 18
winner UAB
ATS vs close
leaned UAB +6.5Miss ✗
margin err: model 19.9 · mkt 11.5 · mkt closer
UTSA @ TempleFinal 2127
Model
pred UTSA 4.1 · actual Temple 6
winner UTSA
ATS vs close
leaned Temple +6.5Cover ✓
margin err: model 10.1 · mkt 12.5 · 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.