Backtesting a play by play based model with good results. However, 2025 tests poorly despite of getting good open to close line movement agreement. Did anyone here have a surprisingly bad 2025 CFB season despite getting decent CLV (on full game sides).
