Model Performance

2026 Season · 3,584 graded predictions

2026 2025 2024 2023 2022 All

In-sample data: 948 of 3,584 predictions were generated after games completed. Only the 2,636 prospective predictions are reliable for evaluation. Metrics will become trustworthy as the current season progresses and daily ETL generates pre-game predictions.

66%

Overall Accuracy

74%

Last 50 Games

0.2125

Brier Score (↓ better)

0.6144

Log Loss (↓ better)

0.6924

AUC (↑ better)

3,584

Predictions Graded

Calibration

When the model says X%, does the favored team win X% of the time? The vertical bar shows the predicted probability; the filled bar shows actual.

50–55%567 games51% actual(-1pp)

55–60%611 games57% actual

60–65%546 games60% actual(-3pp)

65–70%531 games70% actual(+3pp)

70–75%355 games69% actual(-4pp)

75–80%279 games80% actual(+3pp)

80–100%695 games82% actual(-5pp)

Actual win rate Predicted

Score Projection Accuracy

How close were the Monte Carlo score projections to actual results?

4.9

Avg margin error (runs)

+0.5

Margin bias (under-projects)

53%

Own-line cover rate

Within 2 runs of projected margin28%

Within 5 runs of projected margin60%

Based on 3584 games with score projections.

D1 Diamond Top 25 — 2026 Season

Composite ranking of all D1 teams with ≥5 games played.

#	Team	Record	Elo	wOBA	FIP	Run Diff	Score
1	UCLAUCLA Bruins	37–3	1845	0.424	3.63	+5.4	91
2	GTGeorgia Tech Yellow Jackets	33–7	1812	0.478	4.26	+6.3	90
3	TEXTexas Longhorns	30–8	1785	0.419	2.89	+4.4	86
4	UNCNorth Carolina Tar Heels	35–7	1847	0.406	3.93	+4.6	86
5	TA&MTexas A&M Aggies	31–7	1793	0.450	4.72	+5.3	84
6	MSSTMississippi State Bulldogs	31–10	1755	0.423	3.26	+4.8	83
7	UGAGeorgia Bulldogs	32–10	1769	0.465	4.07	+4.5	83
8	ORSTOregon State Beavers	30–8	1756	0.384	2.97	+3.1	80
9	AUBAuburn Tigers	28–12	1750	0.386	3.17	+3.2	77
10	FSUFlorida State Seminoles	29–11	1761	0.393	3.58	+3.1	77
11	MISSOle Miss Rebels	30–12	1766	0.386	3.67	+3.0	77
12	JVSTJacksonville State Gamecocks	34–8	1692	0.399	3.61	+3.6	75
13	NEBNebraska Cornhuskers	31–10	1737	0.413	4.26	+2.8	75
14	MIAMiami Hurricanes	29–11	1703	0.418	4.16	+4.3	74
15	KUKansas Jayhawks	30–11	1753	0.402	4.43	+1.7	74
16	OUOklahoma Sooners	28–12	1735	0.390	4.07	+2.4	73
17	OREOregon Ducks	28–11	1687	0.426	4.18	+3.3	73
18	WVUWest Virginia Mountaineers	27–10	1695	0.404	3.96	+3.1	73
19	FLAFlorida Gators	28–14	1707	0.371	3.09	+2.1	72
20	ASUArizona State Sun Devils	28–13	1661	0.436	4.14	+2.9	72
21	BCBoston College Eagles	32–12	1720	0.381	4.06	+2.9	72
22	USMSouthern Miss Golden Eagles	29–12	1692	0.378	3.29	+1.7	71
23	WAKEWake Forest Demon Deacons	28–14	1695	0.418	4.40	+3.6	71
24	LIBLiberty Flames	29–11	1674	0.392	3.46	+2.5	71
25	TENNTennessee Volunteers	27–14	1720	0.378	3.98	+2.5	70

Minimum 5 games played. wOBA/FIP from FanGraphs.

By Confidence Tier

Accuracy when the model is highly confident in its pick.

Confidence	Games	Accuracy
≥ 60%	2406	72%
≥ 65%	1860	75%
≥ 70%	1329	78%
≥ 75%	974	81%

Top Teams by Elo

Rolling Elo rating (1500 = average). Updated after every game result.

Rank	Team	Games	Elo
1	North Carolina Tar Heels	291	1847
2	UCLA Bruins	275	1845
3	Georgia Tech Yellow Jackets	275	1812
4	Texas A&M Aggies	291	1793
5	Texas Longhorns	289	1785
6	Georgia Bulldogs	277	1769
7	Ole Miss Rebels	281	1766
8	Florida State Seminoles	277	1761
9	Oregon State Beavers	291	1756
10	Mississippi State Bulldogs	272	1755

How to Read This

Overall Accuracy — % of games where the model's favored team won.

Brier Score — measures probability calibration quality. Lower is better.

Log Loss — penalises confident wrong predictions more harshly than Brier score. Lower is better.

AUC — probability that the model ranks a random home win above a random home loss. 0.5 = no skill; 1.0 = perfect.

Calibration chart — a perfectly calibrated model's bar would exactly touch the line in every bucket.

Elo ratings — self-correcting power ratings that update after every game. 1500 is average.