Model Performance

2026 Season · 4,750 graded predictions

2026 2025 2024 2023 2022 All

In-sample data: 957 of 4,750 predictions were generated after games completed. Only the 3,793 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

56%

Last 50 Games

0.2147

Brier Score (↓ better)

0.6189

Log Loss (↓ better)

0.6907

AUC (↑ better)

4,750

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%822 games51% actual(-1pp)

55–60%852 games59% actual(+2pp)

60–65%738 games61% actual(-2pp)

65–70%704 games68% actual(+1pp)

70–75%464 games69% actual(-3pp)

75–80%368 games79% actual(+2pp)

80–100%802 games82% actual(-5pp)

Actual win rate Predicted

Score Projection Accuracy

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

4.8

Avg margin error (runs)

+0.4

Margin bias (under-projects)

54%

Own-line cover rate

Within 2 runs of projected margin28%

Within 5 runs of projected margin61%

Based on 4750 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	GTGeorgia Tech Yellow Jackets	45–9	1841	0.477	4.59	+6.0	92
2	UCLAUCLA Bruins	48–6	1830	0.417	3.85	+4.5	89
3	UNCNorth Carolina Tar Heels	44–10	1839	0.409	3.97	+4.6	87
4	UGAGeorgia Bulldogs	43–12	1826	0.458	4.39	+4.3	87
5	TEXTexas Longhorns	40–12	1804	0.410	3.26	+3.7	86
6	MSSTMississippi State Bulldogs	39–16	1756	0.418	3.69	+4.0	81
7	ORSTOregon State Beavers	43–12	1737	0.383	3.20	+3.0	80
8	TA&MTexas A&M Aggies	38–14	1765	0.434	4.87	+3.7	79
9	AUBAuburn Tigers	36–18	1771	0.384	3.48	+3.1	78
10	FSUFlorida State Seminoles	38–16	1758	0.387	3.53	+2.9	78
11	FLAFlorida Gators	37–18	1756	0.375	3.26	+2.4	77
12	USMSouthern Miss Golden Eagles	40–14	1714	0.385	3.17	+2.5	77
13	MISSOle Miss Rebels	37–19	1759	0.386	3.95	+2.3	76
14	WVUWest Virginia Mountaineers	37–13	1752	0.394	4.16	+3.2	76
15	WAKEWake Forest Demon Deacons	38–18	1732	0.416	4.35	+3.2	75
16	OREOregon Ducks	38–15	1725	0.401	4.14	+2.5	75
17	CBUCalifornia Baptist Lancers	19–3	1621	0.411	3.72	+3.9	75
18	JVSTJacksonville State Gamecocks	42–13	1679	0.395	3.85	+3.0	74
19	KUKansas Jayhawks	39–16	1730	0.401	4.40	+1.8	74
20	NEBNebraska Cornhuskers	41–14	1739	0.406	4.66	+2.5	74
21	USCUSC Trojans	41–14	1698	0.383	3.75	+3.0	74
22	TENNTennessee Volunteers	37–19	1746	0.389	4.22	+2.4	73
23	CAMCampbell Fighting Camels	39–15	1666	0.417	4.04	+2.9	73
24	ARKArkansas Razorbacks	36–19	1750	0.382	4.48	+2.3	72
25	ALAAlabama Crimson Tide	37–18	1754	0.364	4.15	+1.6	72

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%	3076	72%
≥ 65%	2338	75%
≥ 70%	1634	78%
≥ 75%	1170	81%

Top Teams by Elo

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

Rank	Team	Games	Elo
1	Georgia Tech Yellow Jackets	289	1841
2	North Carolina Tar Heels	303	1839
3	UCLA Bruins	289	1830
4	Georgia Bulldogs	290	1826
5	Texas Longhorns	303	1804
6	Auburn Tigers	291	1771
7	Texas A&M Aggies	305	1765
8	Ole Miss Rebels	295	1759
9	Florida State Seminoles	291	1758
10	Mississippi State Bulldogs	286	1756

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.