NCAA Football · Model Accuracy

Picks Model Accuracy

Weekly comparison of BBMI model vs Vegas — spreads and over/under totals

64.2%

Premium ATS

edge ≥ 6 pts, spread ≤ 14

346+

Picks Per Season

458 total (edge ≥ 5)

63.9%

Edge ≥ 10 pts

205 picks

About the Edge Filter & Performance Stats

Headline stats count only picks where edge ≥ 5 pts and Vegas spread ≤ 21 pts. Football lines are captured at a specific point in time — lines routinely move 1–3 points between open and kickoff, and can vary across different books. A difference smaller than 5 pts is within normal market noise and does not represent a meaningful BBMI disagreement with Vegas. Those games are shown in the table but marked ~ and excluded from the record. All win percentages include 95% confidence intervals.

Minimum Edge (|BBMI Line − Vegas Line|):

Filter by Team:

Tip: The model performs best when edge is highest. Try ≥ 6 points to see picks where BBMI most strongly disagrees with Vegas.

Summary Metrics

Sample Size	% Beats Vegas	Wagered	Won	ROI
669	57.0%	$66,900	$72,771	8.8%
—	95% CI: 53.2%–60.7%	—	—	—

Team Performance Analysis

Show:

Number of Teams:

Rank	Team	Picked	Win %	Wagered	Won	ROI
1	Utah State	6	100.0%	$600	$1,146	91.0%
2	Kennesaw State	6	100.0%	$600	$1,146	91.0%
3	TCU	4	100.0%	$400	$764	91.0%
4	Toledo	3	100.0%	$300	$573	91.0%
5	James Madison	3	100.0%	$300	$573	91.0%
Minimum 2 games required. Based on current edge filter (≥0.0 points).

Historical Results By Week

Team records indicate Win-Loss when BBMI picks that team to beat Vegas.

Weekly Summary

Sample Size	% Beats Vegas	Wagered	Won	ROI
1	0.0%	$100	$0	-100.0%
—	95% CI: 0.0%–79.3%	—	—	—

Weekly Performance Breakdown — All Weeks95% confidence intervals shown

Week	ATS Picks	ATS Win%	ATS ROI	O/U Picks	O/U Win%	O/U ROI
1/17 – 1/23	1	0.0%	-100.0%	0	0.0%	+0.0%
1/10 – 1/16	1	100.0%	+91.0%	1	0.0%	-100.0%
1/3 – 1/9	3	33.3%	-36.3%	2	100.0%	+91.0%
12/27 – 1/2	21	52.4%	+0.0%	19	47.4%	-9.5%
12/20 – 12/26	9	88.9%	+69.8%	10	90.0%	+71.9%
12/13 – 12/19	5	40.0%	-23.6%	8	50.0%	-4.5%
12/6 – 12/12	8	50.0%	-4.5%	8	87.5%	+67.1%
11/29 – 12/5	44	63.6%	+21.5%	41	39.0%	-25.5%
11/22 – 11/28	59	54.2%	+3.6%	51	54.9%	+4.9%
11/15 – 11/21	48	52.1%	-0.5%	45	48.9%	-6.6%
11/8 – 11/14	44	65.9%	+25.9%	43	58.1%	+11.0%
11/1 – 11/7	44	54.5%	+4.2%	41	43.9%	-16.1%
10/25 – 10/31	46	56.5%	+8.0%	43	53.5%	+2.2%
10/18 – 10/24	50	50.0%	-4.5%	48	45.8%	-12.5%
10/11 – 10/17	51	45.1%	-13.9%	47	51.1%	-2.5%
10/4 – 10/10	47	59.6%	+13.8%	43	55.8%	+6.6%
9/27 – 10/3	46	47.8%	-8.7%	42	54.8%	+4.6%
9/20 – 9/26	39	59.0%	+12.6%	35	54.3%	+3.7%
9/13 – 9/19	30	70.0%	+33.7%	27	59.3%	+13.2%
9/6 – 9/12	31	45.2%	-13.7%	30	36.7%	-30.0%
8/30 – 9/5	28	78.6%	+50.1%	25	60.0%	+14.6%
8/23 – 8/29	14	85.7%	+63.7%	13	84.6%	+61.6%
Season Total	669	57.0%	+8.8%	622	52.7%	+0.7%

A wider 95% CI range indicates a smaller sample where week-level results are less conclusive. Season ROI is approximate.

🟡 Gold rows = Edge ≥ 6 pts — premium tier, highest accuracy~ Faded rows = Edge < 5 pts — within normal line movement, excluded from stats

Date⇅	Away⇅	Home⇅	Home Line		Edge▼	Actual Line⇅	Away Sc.⇅	Home Sc.⇅	Bet⇅	Win⇅	Result⇅
Date⇅	Away⇅	Home⇅	Vegas⇅	BBMI⇅	Edge▼	Actual Line⇅	Away Sc.⇅	Home Sc.⇅	Bet⇅	Win⇅	Result⇅
2026-01-20	Miami (FL)5-3	Indiana5-2	-7.5	-11.5	~4.0	-6	21	27	$100	$0	✗

Understanding the Numbers — A Guide for New Visitors

Win % (% Beats Vegas)

The share of picks where BBMI correctly predicted which side of the spread would cover. The break-even point at standard −110 juice is ~52.4%. The 95% confidence interval beneath this number shows the plausible range for the true underlying rate — a wider interval means a smaller sample with less certainty. Only picks with edge ≥ 5 pts and Vegas spread ≤ 21 pts are included.

ROI (Return on Investment)

Simulated return assuming a flat $100 wager per pick at −110 odds. Positive ROI means the model has generated paper profit over the tracked period. Past simulated performance does not guarantee future results, and real-world factors like line movement and juice vary.

Edge Column — including the ~ marker

Shown for every row — always a non-negative number representing how far apart BBMI and Vegas are on a given game. Rows with Edge ≥ 6 are highlighted in gold — the premium tier with historically the strongest accuracy. Rows marked ~ have edge < 5 pts and are excluded from stats. The Edge Filter above the table controls which games appear.

Weekly Breakdown Table

Shows week-by-week results so you can assess consistency rather than just the season headline. Football has fewer games per week than basketball — a single bad week can look dramatic on a small sample. Look for the cluster of weekly results around the season mean, and use the 95% CI column to judge how much weight to put on any single week.

Sample Size & Track Record

This model has been walk-forward validated across two independent seasons (2024 and 2025) with a 0.0-point overfitting gap — meaning calibration on past data transfers perfectly to unseen future data. The premium tier (edge ≥ 6 pts) showed 62.4% ATS on 599 games across both seasons. BBMI's results are tracked transparently each season to maintain an honest long-term record.