BOS -11.5 WIN: Injury impact assessment was accurate - Warriors without Curry and Butler were severely compromised against strong home Celtics, though the 6-point edge was smaller than expected
LAL +1.5 WIN: Correctly identified Lakers' hot streak (6 games) as meaningful momentum against Rockets at home
DEN @ MEM UNDER 246.5 WIN: Properly factored in Grizzlies' depleted roster limiting offensive output
WHAT MISSED
TOR @ CHI UNDER 235.5 LOSS: Massively underestimated pace impact - 248 total suggests both teams played uptempo despite Bulls' injuries (expected slower game)
GSW @ BOS UNDER 217.5 LOSS: Injury logic was backwards - fewer healthy players doesn't always mean lower scoring when rotations are shortened and pace increases
Parlay failure: Compounding two correlated under bets (both involving depleted teams) created unnecessary risk
MODEL CALIBRATION NOTES
Confidence ratings were appropriately conservative - MEDIUM confidence on BOS spread hit but with smaller margin than projected
Systematic bias toward unders when injuries present - Model assumes injuries = lower scoring, but reality shows pace/rotation effects can increase totals
Missing pace adjustments when key playmakers are out (teams often play faster, sloppier)
METHODOLOGY ADJUSTMENTS TO CONSIDER
Separate pace impact from scoring impact in injury calculations - Key player absences can increase possessions even while reducing efficiency
Cap parlay correlation exposure - Avoid combining multiple bets with similar logical foundations (both unders based on injury reasoning)