Model performance
144 graded· 2 awaiting result· 26 upcoming· CLV on 139/144· last graded Nov 1, 11:03 PM CT
Winners
41-19
68%
Totals
40-44
48%
Spread
26-34
43%
Avg CLV
0.0%
beat the close
Shop gain
+1.40%
best vs avg book (ML)
Work in progress: NBA is in the red. Small samples swing hard, but closing-line value near zero says that model isn’t beating the market yet, so we're retuning it. Every pick stays in the record as graded.
- NBA (Oct 28): The model now leans on the market until teams have ~20 games of this season (it was still rating last year’s rosters), and skips games where it disagrees with the market most.
By model
| Model | Moneyline | ML CLV | Totals | Tot CLV | Spread | Spread CLV |
|---|---|---|---|---|---|---|
| nba_elo | 41-19 · 68.3% | 0.0% | — | — | 26-34 · 43.3% | -0.3% |
| nba_totals | — | — | 40-44 · 47.6% | 0.2% | — | — |
+EV picks — flagged plays only
the card's green ● plays: 2–15% EV| Model | +EV picks | Record | Win% | Avg odds | P/L (u) | ROI | CLV |
|---|---|---|---|---|---|---|---|
| nba_elo · Moneyline | 30 | 15-15 | 50% | -141 | -5.83u | -19.4% | 0.2% |
| nba_elo · Spread | 34 | 9-25 | 26% | -101 | -16.45u | -48.4% | -0.1% |
| nba_totals | 50 | 21-29 | 42% | +100 | -8.13u | -16.3% | 0.2% |
Thin edges (0–2% EV) are listed on the board without a ● and aren't in the record above. Tracked anyway: 19-13, +2.91u, avg CLV +0.4% over 32 plays (small sample).
Plays above 15% EV are hidden from the picks as likely bad lines, and aren't in the record above. Tracked anyway: 3-1, +1.97u, avg CLV -1.3% over 4 plays (small sample).