Bulls +8.5 WIN: Correctly identified that 8.5-point road favorite was too high for injury-depleted Rockets against a desperate Bulls team
Injury impact assessment: Properly weighted multiple key injuries across both Houston and Chicago rosters affecting game dynamics
WHAT MISSED
All totals went OVER: Model systematically underestimated scoring pace - HOU/CHI (256 vs 232.5), GSW/DAL (268 vs 231.5), MIL/LAC (225 vs 221.5)
Large spread favorites covered easily: BKN +15.5 and MIL +13.5 both failed badly - model overvalued injured underdogs' competitiveness against motivated favorites
MODEL CALIBRATION NOTES
Systematic UNDER bias: 3/3 totals went over by significant margins (3.5 to 36.5 points), suggesting pace calculations are consistently conservative
Injury overweighting: Model appears to overvalue how much star player absences affect competitiveness against elite teams - injured teams still got blown out
METHODOLOGY ADJUSTMENTS TO CONSIDER
Pace adjustment: Increase baseline scoring projections by 8-12 points per game to account for modern NBA pace trends
Blowout threshold: When spread exceeds 13 points, reduce injury impact weighting by 50% - talent gaps often override injury advantages in lopsided matchups