NBA Over/Under Analysis - 2026-02-03
Summary
| Metric | Value | |||||||||
|---|---|---|---|---|---|---|---|---|---|---|
| Total Picks | 6 | |||||||||
| Correct | 3 | |||||||||
| Accuracy | 50% | |||||||||
| Skipped | 4 | |||||||||
| HIGH Conf | 2/3 correct (67%) | |||||||||
| MEDIUM Conf | 1/3 correct (33%) | Results | Game | Pick | Confidence | Vegas | Predicted | Actual | Result | Margin |
| --------- | ----------- | ---------- | ----- | --------- | ------ | ------ | ------ | |||
| DEN @ DET | OVER 228.9 | MEDIUM | 228.9 | 231.0 | 245 | OVER | +16.1 | |||
| UTA @ IND | SKIP | - | 236.6 | - | 253 | OVER | - | |||
| NYK @ WAS | SKIP | - | 227.6 | - | 233 | OVER | - | |||
| ATL @ MIA | UNDER 241.1 | MEDIUM | 241.1 | 238.0 | 242 | OVER | +0.9 | |||
| LAL @ BKN | SKIP | - | 222.4 | - | 234 | OVER | - | |||
| BOS @ DAL | UNDER 223.2 | HIGH | 223.2 | 219.0 | 210 | UNDER | -13.2 | |||
| CHI @ MIL | UNDER 223.4 | MEDIUM | 223.4 | 220.0 | 246 | OVER | +22.6 | |||
| ORL @ OKC | SKIP | - | 219.6 | - | 220 | OVER | - | |||
| PHI @ GSW | UNDER 221.4 | HIGH | 221.4 | 218.0 | 207 | UNDER | -14.4 | |||
| PHX @ POR | UNDER 218.5 | HIGH | 218.5 | 215.0 | 255 | OVER | +36.5 |
All 10 games went OVER or UNDER vs Vegas:
- OVER: 8 games (80%)
- UNDER: 2 games (20%)
- Massive OVER night across the league
- Pick: OVER 228.9 | Predicted: 231.0
- Actual: 245 (DEN 121, DET 124) | Margin: +16.1 over Vegas
- Factors: Denver fast pace, both teams 100% OVER trend
- Takeaway: Direction was correct but magnitude was much larger than predicted. Denver's fast pace factor was on point. Actual was 14 pts above our prediction — we were too conservative even picking OVER.
- Pick: UNDER 223.2 | Predicted: 219.0
- Actual: 210 (BOS 110, DAL 100) | Margin: -13.2 under Vegas
- Factors: Tatum OUT, Irving OUT, Dallas decimated roster
- Takeaway: Massive UNDER — even more than predicted. Dual star absences (50+ PPG removed) was the correct signal. We predicted 219, actual was 210. The injury factor was even stronger than our -6 to -8 pt adjustment suggested. Dallas scored only 100 with their gutted roster.
- Pick: UNDER 221.4 | Predicted: 218.0
- Actual: 207 (PHI 113, GSW 94) | Margin: -14.4 under Vegas
- Factors: PHI away B2B, Curry OUT, Butler OUT, GSW roster gutted
- Takeaway: Another massive UNDER. B2B + dual star absences was the strongest UNDER signal on the slate. GSW scored only 94 — their roster was completely depleted. The B2B + star injury combo is our most reliable HIGH confidence signal.
- Pick: UNDER 241.1 | Predicted: 238.0
- Actual: 242 (ATL 127, MIA 115) | Margin: +0.9 over Vegas
- Factors: Herro OUT, Rozier OUT (~37 PPG removed from MIA)
- Margin of error: Only 0.9 pts! This was essentially a push.
- Why wrong: Miami's injuries DID reduce their scoring (115 is below average for a 241.1 total game), but Atlanta exploded for 127 points. The fast-pace Hawks offense overwhelmed the injury adjustment. Without B2B fatigue as a primary factor, the injury-only thesis was too thin.
- Lesson: Injuries alone without B2B or pace alignment are not enough for UNDER picks. MIA scoring 115 confirmed the injury impact, but we can't control the other team.
- Pick: UNDER 223.4 | Predicted: 220.0
- Actual: 246 (CHI 115, MIL 131) | Margin: +22.6 over Vegas
- Factors: Giannis OUT (30+ PPG), Giddey OUT
- Why wrong: CATASTROPHIC miss. Milwaukee scored 131 WITHOUT Giannis(!). This completely defies the star-absence logic. The Bucks role players stepped up massively, and/or their defensive identity collapsed without Giannis, leading to a track meet.
- Lesson: Star absence does NOT always mean lower scoring. Giannis is also an elite defender — his absence may INCREASE total scoring because both teams score more easily. The guidelines only account for offensive impact of star absences, not the defensive void they create. Milwaukee's depth and system can maintain/exceed scoring even without their star.
- Pick: UNDER 218.5 | Predicted: 215.0
- Actual: 255 (PHX 130, POR 125) | Margin: +36.5 over Vegas
- Factors: Booker OUT, Lillard OUT (50+ PPG removed), Henderson/Avdija doubtful
- Why wrong: SYSTEM-BREAKING miss. This is the largest error in system history by a massive margin. Both star scorers OUT and the total was 255 — 36.5 pts ABOVE Vegas and 40 pts above our prediction. This game obliterates the assumption that "star scorers OUT = lower scoring."
- Root cause analysis: 1. Replacement player effect: When stars sit, replacement-level players may play faster, take more shots, and create a chaotic, high-scoring game 2. Defensive void: Booker and Lillard are also solid defenders. Their absence likely led to poor defense on both ends 3. Pace change: Without primary ball-handlers to slow the game down, both teams may have played at a frantic pace with more possessions 4. Small sample arrogance: We had 0% accuracy on PHX and no data on POR, yet assigned HIGH confidence based solely on the injury factor
- CRITICAL LESSON: Dual star absences WITHOUT B2B is NOT a reliable UNDER signal. The replacement player effect can cause HIGHER totals, not lower.
- OVER picks today: 1/1 (100%)
- UNDER picks today: 2/5 (40%)
- Cumulative OVER: 3/4 (75%)
- Cumulative UNDER: 9/20 (45%)
Correct Picks (3)
DEN @ DET OVER 228.9 (MEDIUM)
BOS @ DAL UNDER 223.2 (HIGH)
PHI @ GSW UNDER 221.4 (HIGH)
Wrong Picks (3)
ATL @ MIA UNDER 241.1 (MEDIUM)
CHI @ MIL UNDER 223.4 (MEDIUM)
PHX @ POR UNDER 218.5 (HIGH)
Factor Performance Analysis
Factors That Worked
| Factor | Games Used | Correct | Accuracy | |||||
|---|---|---|---|---|---|---|---|---|
| B2B Away + Star Injury | 1 (PHI@GSW) | 1 | 100% | |||||
| Dual Star Absences + B2B | 1 (PHI@GSW) | 1 | 100% | |||||
| Fast Pace + OVER Trend | 1 (DEN@DET) | 1 | 100% | |||||
| Dual Star Absences (no B2B) | 2 (BOS@DAL, PHX@POR) | 1 | 50% | Factors That Failed | Factor | Games Used | Correct | Accuracy |
| ------------------------- | -------------------- | ------- | -------- | |||||
| Star Injury Only (no B2B) | 2 (ATL@MIA, CHI@MIL) | 0 | 0% | |||||
| Roster Depletion (no B2B) | 2 (CHI@MIL, PHX@POR) | 0 | 0% |
Key Insight: B2B Is Still the Primary Signal
The only UNDER pick that worked WITH star absences was PHI@GSW, which ALSO had B2B as a factor. The two UNDER picks based purely on star absences without B2B (CHI@MIL and PHX@POR) both lost. BOS@DAL worked but that was a unique situation with BOTH teams' rosters decimated AND confirmed UNDER trends.
Confidence Level Performance
| Level | Total | Correct | Accuracy | Notes | |||||
|---|---|---|---|---|---|---|---|---|---|
| HIGH | 3 | 2 | 67% | PHX@POR was the catastrophic miss | |||||
| MEDIUM | 3 | 1 | 33% | Only DEN@DET OVER hit | Vegas Deviation Analysis | Deviation | Games | Correct | Accuracy |
| --------- | ---------------------- | ------- | -------- | ||||||
| 0-2 pts | 1 (DEN@DET: 2.1) | 1 | 100% | ||||||
| 2-4 pts | 4 (ATL, CHI, PHI, PHX) | 2 | 50% | ||||||
| >4 pts | 1 (BOS@DAL: 4.2) | 1 | 100% |
OVER vs UNDER Direction
OVER continues to massively outperform UNDER in our system.
Skipped Games Analysis
| Game | Vegas | Actual | Would OVER have hit? | |||||
|---|---|---|---|---|---|---|---|---|
| UTA @ IND | 236.6 | 253 | YES (+16.4) | |||||
| NYK @ WAS | 227.6 | 233 | YES (+5.4) | |||||
| LAL @ BKN | 222.4 | 234 | YES (+11.6) | |||||
| ORL @ OKC | 219.6 | 220 | YES (+0.4) | All 4 skipped games went OVER. This was an extreme OVER night league-wide. Cumulative Stats (Updated) | Metric | Before | After | Change |
| ----------------- | ---------- | ---------- | ------ | |||||
| Total Predictions | 18 | 24 | +6 | |||||
| Total Correct | 9 | 12 | +3 | |||||
| Overall Accuracy | 50% | 50% | 0% | |||||
| HIGH Confidence | 100% (1/1) | 75% (3/4) | -25% | |||||
| MEDIUM Confidence | 47% (7/15) | 44% (8/18) | -3% | |||||
| OVER Accuracy | 67% (2/3) | 75% (3/4) | +8% | |||||
| UNDER Accuracy | 47% (7/15) | 45% (9/20) | -2% |
Lessons Learned
1. B2B remains the ONLY reliable UNDER primary signal. Without B2B, injury-based UNDER picks are 0/2 today and trending poorly overall. 2. Star absences WITHOUT B2B can go EITHER way. The PHX@POR game (255 total with Booker+Lillard OUT) proves replacement players can create higher-scoring chaos. 3. Dual star absences only work for UNDER when COMBINED with B2B (PHI@GSW) or extreme roster depletion on BOTH sides (BOS@DAL where DAL scored only 100). 4. OVER picks continue to outperform UNDER (75% vs 45% cumulative). The system should lean OVER more aggressively. 5. Giannis OUT = potential OVER, not UNDER. His defensive presence may suppress scoring more than his offensive production generates. His absence opened up a 246-point game. 6. MEDIUM confidence on UNDER without B2B should be downgraded to SKIP. 7. PHX@POR is the system's biggest lesson: 0% accuracy on PHX entering the day, no data on POR, yet we assigned HIGH confidence based on injuries alone. Never assign HIGH without B2B or proven team trend data.
Rule Adjustments Recommended
1. HIGH confidence UNDER now REQUIRES B2B — star absences alone are insufficient 2. Star absence adjustment cap: -3 pts max per star even for 25+ PPG players (replacement effect offsets some) 3. Consider the DEFENSIVE impact of star absences — defensive stars OUT may INCREASE total 4. OVER picks deserve more weight — 75% accuracy vs 45% for UNDER 5. Never assign HIGH on teams with 0% prediction accuracy regardless of injury factors
_Generated: 2026-02-04_ | _System Version: O/U Predictor v1.0_