FINAL: Howard 86 — UMBC 83. Our Monte Carlo simulation projected Howard 78.9 - UMBC 78.9 (Howard at 56.7% win probability). The spread is 1.5 and the total is 141.5.
Howard
78.9
Projected Score
VS
O/U 141.5
UMBC
78.9
Projected Score
Win Probability
HowardUMBC
+1.5
Spread (Howard)
141.5
Total Line
10,000
Simulations
UMBCHoward W4
Calibrated accuracy at this confidence: 59.7% (4,284 games)
Projected Points Range 10th – 90th percentile
UMBC
667992
Howard
667992
Projected
Howard 78.9 — UMBC 78.9
Actual
Howard 86 — UMBC 83
Pick Results
Over 141.5totalWIN+1.36u
Spread Analysis
Howard Cover
59.0%
UMBC Cover
41.0%
ATS Edge: +2.5 pts
AI Intelligence Analysis
NEUTRALGREEN ZONE80.6% WR (n=19)
The near pick'em market (-105 Howard / -107 UMBC) appropriately prices this game — the model also sees it as a coin flip (Howard 56.7% win prob) with only a 0.6-pt spread edge, making this an unprofitable bet profile.
Key Factors
- Market lines: Howard -105 / UMBC -107 — essentially a pick'em with both sides near 50/50
- Model win prob: Howard 56.7% — near-coin-flip territory, well below our 65%+ preferred ML threshold
- Predicted total 157.7 vs market 141.5 = +16.2 pts gap (massive over signal) but OVER zone is RED (24.6% WR)
- Spread edge only +0.6 pts (model: Howard -0.9, market: Howard -1.5) — zero meaningful spread value
- 0-5% edge bucket: 23.5% WR over 30 days (17 bets) — worst performing edge bucket in current regime
Risk Factors
- Away ML (UMBC) profitability zone is RED: CBB|ml|away|underdog|any = 18.6% WR (333 bets)
- Total discrepancy is massive (+16.2 pts) but OVER is a documented money pit (RED zone, 24.6% WR)
- Both teams near pick'em — model has no directional edge to exploit
MODEL MARKET CONFLICTTOTALS VALUE
Edge Analysis
Moneyline
Howard 56.7%
--
Spread
+1.5
--
Total
141.5
--
How this prediction was generated: This page shows output from the Olympus Bets College Basketball Monte Carlo engine. Each game is simulated 10,000 times using real-time team data, injury reports, and current odds. Probabilities are calibrated using Bayesian methods and sized via the Kelly Criterion. Full methodology →