Accountability

MODEL RECORD

Every published projection is graded against the final score and stays on this page, hit or miss. Two separate things are tracked: our pregame game projections and our in-game win-probability model.

Projections: Week 6, posted Oct 10, 2026 · graded from ESPN finals · win-probability metrics from held-out 2025 testing.

The B1G+ take: the sample is tiny. Calibration check: B1G+A picks carried an average pregame FPI win chance of 81%, so you'd expect about 4.1 of 5 right; it got 4. With 5 games that is far too small a sample to say the model is well calibrated, and we'll keep tracking it in the open. Straight-up pick percentages also lean on ESPN FPI, which supplies the win chance. The projected score is our own blend of the market-implied score and team scoring stats.

B1G+A RECORD

green = .500 or better · red = below .500

	ALL-TIMEsince 10/9/26 · 5 graded	THIS WEEKWeek 6 · 5 graded · 2 pending
Straight upprojected winner won	4-1.800	4-1.800
Against the spreadB1G+A lean vs. DraftKings	4-1.800	4-1.800
Over/underB1G+A lean vs. DraftKings	2-3.400	2-3.400
Avg margin errorprojected vs. actual margin	3.8pts	3.8pts

Projection scorecard · Week 6

Original post ↗

Game	Projected	Final	Pick	Winner	Spread	Total	Margin err
IU @ NEB	IU 28–24	20–17	IU 70%	Hit	NEB +7.5 Hit	OVER 47.5 Miss	1
BALL @ NU	BALL 13–40	14–45	NU 99%	Hit	BALL +35.5 Hit	UNDER 53.5 Miss	4
UCLA @ ORE	UCLA 29–34	17–15	ORE 85%	Miss	UCLA +10.5 Hit	OVER 58.5 Miss	7
ILL @ MSU	ILL 26–22	38–37	ILL 54%	Hit	ILL -2.5 Miss	OVER 47.5 Hit	3
MD @ OSU	MD 16–41	16–45	OSU 98%	Hit	MD +33.5 Hit	OVER 56.5 Hit	4
USC @ PSU	USC 27–28	—	PSU 62%	Pending	USC +1.5 Pending	OVER 54.5 Pending	—
MINN @ PUR	MINN 30–24	—	MINN 53%	Pending	MINN -2.5 Pending	OVER 49.5 Pending	—

B1G+ predicted score averages the Vegas-implied score (from the DraftKings spread and total) with a stat blend: each offense's points per game averaged with the opponent's points allowed per game, using stats through Oct. 3. Win chance from ESPN FPI. Lines from DraftKings via ESPN as of Saturday morning. Projections are estimates, not guarantees. Not projected: Iowa @ Washington (Fri Oct 9; the Week 6 card covered Saturday games only).

In-game win-probability model

tested on all of 2025, held out

Brier (lower is better)0.1242125,614 plays, 764 games

vs cfbfastR built-in WP0.1507same plays: ours 0.1239

vs ESPN live WP0.1254 / 0.1261ours / ESPN · 300 games, 49,907 plays

Brier gap vs ESPN, 95% CI-0.0049 to +0.0035includes zero: a statistical tie

Calibration, 2025

Predicted home win probability vs. how often the home team actually won, by decile. On the diagonal = well calibrated.

Brier score by quarter: ours vs ESPN

Same 300 games. We're slightly better in Q1–Q2, even in Q3–Q4, and worse in overtime (only 95 plays).

The B1G+ take: The win-probability model clearly beats a spread-only baseline (0.177) and cfbfastR's built-in model on the same plays. Against ESPN it's a statistical tie: the 95% interval for the Brier gap spans zero. Overtime is weak because there's so little OT data. The pregame projection record is only 5 games, which proves nothing yet. We'll keep adding every graded game here.