Statalysts.
Statistical modelling · College sport

Machine Learning College Predictions.
Each game predicted.

Statalysts fits a single power rating to every college football and basketball team from the season's scoring margins, anchored to a preseason projection and pulled toward each team's own results as the evidence accumulates. Those ratings produce a projected margin and win probability for every upcoming game. Want to create your own matchup? Open the custom predictor and see how the model rates any two teams.

Statalysts
Play of the day
Football Widest disagreement
Tennessee Tech 2.4%
at
Mississippi State 97.6%

The model makes Mississippi State a 26.0-point favorite, which is 9.5 points clear of the market's line. This is the largest gap between the model and the market on this week's board.

Model line 26.0 pts
Mississippi State
Market line Mississippi State -35.5
Edge 9.5 pts
Win probability 97.6%
One game a day · the widest gap on the board See the full football board
This week

Week 12 recommended games

The games where the model is at least five points away from the posted market line. A wide gap is a measure of disagreement. Click any game for the full breakdown.

64 games this week
Matchup Model line Market line Edge Win prob
Tennessee Tech 9-2
Mississippi State 4-6
Sat, Nov 21 · 1:00 PM ET
Mississippi State −26.0 Mississippi State -35.5 9.5 97.6%
2.4% Tennessee Tech
WIN PROB
97.6% Mississippi State
Projected margin Mississippi State by 26.0
Disagreement with market 9.5 pts strong
Venue Davis Wade Stadium — Starkville, MS
New Mexico 8-2
Wyoming 7-3
Fri, Nov 20 · 9:00 PM ET
New Mexico −16.6 New Mexico −7.5 9.1 91.5%
91.5% New Mexico
WIN PROB
8.5% Wyoming
Projected margin New Mexico by 16.6
Disagreement with market 9.1 pts strong
Venue War Memorial Stadium — Laramie, WY
Virginia Tech 6-4
5 Miami 9-1
Sat, Nov 21 · 7:00 PM ET
Miami −29.2 Miami -20.5 8.7 98.5%
1.5% Virginia Tech
WIN PROB
98.5% 5 Miami
Projected margin Miami by 29.2
Disagreement with market 8.7 pts strong
Venue Hard Rock Stadium — Miami Gardens, FL
Georgia State 5-5
Marshall 5-5
Sat, Nov 21 · 4:00 PM ET
Georgia State −5.7 Marshall -2.5 8.2 69.3%
69.3% Georgia State
WIN PROB
30.7% Marshall
Projected margin Georgia State by 5.7
Disagreement with market 8.2 pts strong
Venue Joan C. Edwards Stadium — Huntington, WV
Northwestern 6-4
Minnesota 5-5
Sat, Nov 21 · 12:00 PM ET
Northwestern −9.7 Northwestern −2.5 7.2 79.9%
79.9% Northwestern
WIN PROB
20.1% Minnesota
Projected margin Northwestern by 9.7
Disagreement with market 7.2 pts strong
Venue Huntington Bank Stadium — Minneapolis, MN
North Texas 7-3
Tulane 5-5
Sat, Nov 21 · 12:00 PM ET
North Texas −12.2 North Texas −6.5 5.7 85.0%
85.0% North Texas
WIN PROB
15.0% Tulane
Projected margin North Texas by 12.2
Disagreement with market 5.7 pts strong
Venue Yulman Stadium — New Orleans, LA

Showing the 6 widest gaps of 64 games

The full board can be filtered by conference and week, and sorted by edge or kickoff.

Last week

How the recommended games finished

Every recommended game, a 5+ point edge over the market line, from the most recently completed week, graded against the spread and on the outright winner. Click any game for the full breakdown.

Last week

Recommended games 8
Against the spread 5–3 62.5%
Straight up 6–2 75.0%

2026 season

Recommended games 139
Against the spread 72–63–4 53.3%
Straight up 105–34 75.5%

Pushes are listed last in the spread record and don't count toward the percentage.

Date Final score Model called Recommended Outcome
2026-11-15
North Dakota State 21
Hawai'i 13
North Dakota State by 13.988% win prob North Dakota State 7.4 pt edge
Winner Covered
87.9% North Dakota State
WIN PROB
12.1% Hawai'i

What the model gave each side before kickoff.

Model projected North Dakota State by 13.9
Actual result North Dakota State by 8
Market line North Dakota State −6.5
2026-11-15
Ole Miss 17
Oklahoma 24
Ole Miss by 3.663% win prob Ole Miss 6.1 pt edge
Winner No cover
62.6% Ole Miss
WIN PROB
37.4% Oklahoma

What the model gave each side before kickoff.

Model projected Ole Miss by 3.6
Actual result Oklahoma by 7
Market line Oklahoma -2.5
2026-11-14
Oregon State 41
Boise State 46
Boise State by 9.780% win prob Boise State 5.2 pt edge
Winner Covered
20.1% Oregon State
WIN PROB
79.9% Boise State

What the model gave each side before kickoff.

Model projected Boise State by 9.7
Actual result Boise State by 5
Market line Boise State -4.5
2026-11-14
East Carolina 17
Charlotte 16
East Carolina by 17.292% win prob East Carolina 5.7 pt edge
Winner No cover
92.2% East Carolina
WIN PROB
7.8% Charlotte

What the model gave each side before kickoff.

Model projected East Carolina by 17.2
Actual result East Carolina by 1
Market line East Carolina −11.5
2026-11-14
Michigan 7
Oregon 31
Oregon by 15.790% win prob Oregon 6.2 pt edge
Winner Covered
9.6% Michigan
WIN PROB
90.4% Oregon

What the model gave each side before kickoff.

Model projected Oregon by 15.7
Actual result Oregon by 24
Market line Oregon -9.5
2026-11-14
Tulane 20
Rice 17
Tulane by 2.458% win prob Tulane 5.4 pt edge
Winner Covered
58.3% Tulane
WIN PROB
41.7% Rice

What the model gave each side before kickoff.

Model projected Tulane by 2.4
Actual result Tulane by 3
Market line Rice -3.0
2026-11-14
James Madison 22
UConn 46
James Madison by 10.682% win prob James Madison 7.1 pt edge
Winner No cover
81.9% James Madison
WIN PROB
18.1% UConn

What the model gave each side before kickoff.

Model projected James Madison by 10.6
Actual result UConn by 24
Market line James Madison −3.5
2026-11-13
Louisiana 42
UL Monroe 21
Louisiana by 18.593% win prob Louisiana 7.0 pt edge
Winner Covered
93.4% Louisiana
WIN PROB
6.6% UL Monroe

What the model gave each side before kickoff.

Model projected Louisiana by 18.5
Actual result Louisiana by 21
Market line Louisiana −11.5
Power ratings

Top ten, both sports

A rating is a team's strength in points. The difference between two ratings is the model's projected margin on a neutral floor.

#TeamRating spreadRating
1 Georgia 29.6
2 Notre Dame 28.5
3 Alabama 26.5
4 Miami 26.4
5 Ohio State 24.7
6 Wake Forest 23.6
7 Texas 23.5
8 Indiana 23.1
9 Utah 22.6
10 Oregon 22.3

Full top 25 for both sports

Every rated team, ordered by rating, updated as results come in.

Method

How a rating becomes a prediction

01

Collect the season

Every completed game is pulled from the college football and basketball data APIs: final scores, schedule, venue and, where posted, the market line. Football rates FBS and FCS teams on the same scale, because a large share of FBS games are played against FCS opponents and those results only mean something if both sides carry a real rating.

02

Fit a rating to each team

A ridge regression fits one rating per team to the season's scoring margins, so a team's number reflects who it played as well as by how much it won. Early in the season there is little to fit on, so the regression is pulled toward a preseason projection built from the team's own final rating last season and a published preseason rating.

As games accumulate, the weight on this season's results grows and the prior fades.

03

Project the game

The difference between two ratings, plus home advantage where it applies, is the projected margin. That margin converts to a win probability for each side.

Where a sportsbook has posted a line, the model's number is compared against it. The gap between the two is the edge — not a claim that the model is right, but a measure of where it disagrees most with the market.