Backtest Results - Thursday, March 12, 2026
SPREAD PICKS SCORECARD
| Pick |
Result |
Outcome |
| OKC -7.5 |
Celtics 102 - Thunder 104 |
Failed to cover by 5.5 |
| DET -14.5 |
76ers 109 - Pistons 131 |
Covered by 7.5 |
| IND +8.5 |
Suns 123 - Pacers 108 |
Failed to cover by 6.5 |
Spread Record: 1-2 (-1.2u)
OVER/UNDER PICKS SCORECARD
| Pick |
Result |
Outcome |
| WAS @ ORL UNDER 233.5 |
Wizards 131 - Magic 136 (Total: 267) |
Over by 33.5 |
| DEN @ SAS UNDER 240.5 |
Nuggets 136 - Spurs 131 (Total: 267) |
Over by 26.5 |
| CHI @ LAL UNDER 238.5 |
Bulls 130 - Lakers 142 (Total: 272) |
Over by 33.5 |
O/U Record: 0-3 (-3.3u)
PARLAY RESULT
| Leg |
Pick |
Outcome |
| 1 |
Detroit Pistons -14.5 |
UNMATCHED |
| 2 |
WAS @ ORL UNDER 233.5 |
LOSS |
Parlay Result: LOSS (-0.5u)
DAILY SUMMARY
| Category |
Record |
Units |
| Spreads |
1-2 |
-1.2 |
| Over/Under |
0-3 |
-3.3 |
| Parlay |
LOSS |
-0.5 |
| Daily Total |
|
-5.0 |
QUALITATIVE ANALYSIS
WHAT WORKED
- Detroit -14.5 spread hit cleanly - Model correctly identified Philadelphia's massive injury impact (4 key players out) and properly weighted the compounding effect of multiple Tier 1 absences
- Injury adjustment methodology was sound - The 17.8 point raw adjustment (capped at 7.0) for PHI's depleted roster proved accurate in practice
WHAT MISSED
- All three totals went massively over - WAS/ORL (33.5 over), DEN/SAS (26.5 over), CHI/LAL (33.5 over) suggests systematic underestimation of pace/scoring
- OKC failed to cover -7.5 vs injury-depleted Boston - Model may have overvalued Tatum's absence impact or underestimated Boston's depth/motivation
- Indiana +8.5 vs Phoenix missed - Despite both teams being "depleted," the model didn't properly account for talent differential
MODEL CALIBRATION NOTES
- Confidence ratings weren't provided in original analysis, making calibration assessment impossible
- Clear systematic bias toward unders - All three total bets missed badly in same direction, indicating pace/efficiency models are outdated
- Injury impact appears well-calibrated for spreads but may need refinement for totals (injured teams can still play fast)
METHODOLOGY ADJUSTMENTS TO CONSIDER
- Update pace/efficiency baselines - Current totals model appears anchored to lower-scoring era; modern NBA is trending higher
- Separate injury impact for spreads vs totals - Missing stars affects competitiveness more than game flow; depleted teams often play faster, not slower
Overall: 1-5 record suggests model needs significant recalibration, particularly for totals where it's systematically low.
CUMULATIVE SEASON RECORD
| Category |
Record |
Win % |
Units |
| ATS |
42-40 |
51.2% |
-6.4 |
| O/U |
43-45 |
48.9% |
|
| Parlays |
6-27 |
|
|
| Season Total |
|
|
-6.4u |
View original analysis