2025 College Football Week 2 Model Backtest
Quick answer: In this retrospective 2025
Week 2 FBS-vs-FBS backtest, the frozen model went 40/50 (80.0%) with mean Brier
0.1519. Full 2025: 550/762 (72.2%). These were not live picks.
Ratings use final 2023-2024 FBS-vs-FBS games only. They were frozen before any
2025 result was graded. Historical pregame spreads may be blended, but their probability
mapping was calibrated on 2024 outcomes—not on the 2025 test results.
Week 2 scorecard
| Metric | Value |
|---|---|
| Games | 50 |
| Decided | 50 |
| Correct picks | 40/50 (80.0%) |
| Mean Brier | 0.1519 |
| Full-season accuracy | 550/762 (72.2%) |
Full-season method comparison
The headline is the blended ensemble. This table separates the ratings model from
the historical spread so market information is not misreported as pure model performance.
| Method | Games | Correct | Accuracy | Mean Brier |
|---|---|---|---|---|
| Elo model only | 762 | 495 | 65.0% | 0.221 |
| Market spread only | 762 | 564 | 74.0% | 0.1733 |
| 40/60 ensemble | 762 | 550 | 72.2% | 0.1804 |
Every Week 2 FBS-vs-FBS game: pick vs actual
| Date | Away | Home | Home WP | Pick | Actual | Result | Source |
|---|---|---|---|---|---|---|---|
| 2025-09-05 | James Madison 14 | Louisville 28 | 75.7% | Louisville | Louisville | HIT | ensemble |
| 2025-09-05 | Northern Illinois 9 | Maryland 20 | 65.6% | Maryland | Maryland | HIT | ensemble |
| 2025-09-06 | Illinois 45 | Duke 19 | 52.0% | Duke | Illinois | MISS | ensemble |
| 2025-09-06 | Kennesaw State 9 | Indiana 56 | 96.3% | Indiana | Indiana | HIT | ensemble |
| 2025-09-06 | Iowa 13 | Iowa State 16 | 55.5% | Iowa State | Iowa State | HIT | ensemble |
| 2025-09-06 | Liberty 24 | Jacksonville State 34 | 40.0% | Liberty | Jacksonville State | MISS | ensemble |
| 2025-09-06 | Virginia 31 | NC State 35 | 65.3% | NC State | NC State | HIT | ensemble |
| 2025-09-06 | Florida International 0 | Penn State 34 | 97.2% | Penn State | Penn State | HIT | ensemble |
| 2025-09-06 | Central Michigan 17 | Pittsburgh 45 | 85.6% | Pittsburgh | Pittsburgh | HIT | ensemble |
| 2025-09-06 | Baylor 48 | SMU 45 | 67.4% | SMU | Baylor | MISS | ensemble |
| 2025-09-06 | UConn 20 | Syracuse 27 | 71.5% | Syracuse | Syracuse | HIT | ensemble |
| 2025-09-06 | San José State 7 | Texas 38 | 94.8% | Texas | Texas | HIT | ensemble |
| 2025-09-06 | Kent State 14 | Texas Tech 62 | 97.8% | Texas Tech | Texas Tech | HIT | ensemble |
| 2025-09-06 | Utah State 22 | Texas A&M 44 | 92.8% | Texas A&M | Texas A&M | HIT | ensemble |
| 2025-09-06 | Bowling Green 20 | Cincinnati 34 | 68.9% | Cincinnati | Cincinnati | HIT | ensemble |
| 2025-09-06 | Troy 16 | Clemson 27 | 90.7% | Clemson | Clemson | HIT | ensemble |
| 2025-09-06 | Delaware 7 | Colorado 31 | 89.4% | Colorado | Colorado | HIT | ensemble |
| 2025-09-06 | Ole Miss 30 | Kentucky 23 | 25.2% | Ole Miss | Ole Miss | HIT | ensemble |
| 2025-09-06 | Kansas 31 | Missouri 42 | 68.4% | Missouri | Missouri | HIT | ensemble |
| 2025-09-06 | UAB 24 | Navy 38 | 90.7% | Navy | Navy | HIT | ensemble |
| 2025-09-06 | Oklahoma State 3 | Oregon 69 | 96.5% | Oregon | Oregon | HIT | ensemble |
| 2025-09-06 | Fresno State 36 | Oregon State 27 | 53.2% | Oregon State | Fresno State | MISS | ensemble |
| 2025-09-06 | Miami (OH) 17 | Rutgers 45 | 66.2% | Rutgers | Rutgers | HIT | ensemble |
| 2025-09-06 | Texas State 43 | UTSA 36 | 60.2% | UTSA | Texas State | MISS | ensemble |
| 2025-09-06 | North Texas 33 | Western Michigan 30 | 35.2% | North Texas | North Texas | HIT | ensemble |
| 2025-09-06 | West Virginia 10 | Ohio 17 | 56.9% | Ohio | Ohio | HIT | ensemble |
| 2025-09-06 | Middle Tennessee 10 | Wisconsin 42 | 90.8% | Wisconsin | Wisconsin | HIT | ensemble |
| 2025-09-06 | South Florida 18 | Florida 16 | 85.0% | Florida | South Florida | MISS | ensemble |
| 2025-09-06 | Arkansas State 14 | Arkansas 56 | 76.4% | Arkansas | Arkansas | HIT | ensemble |
| 2025-09-06 | Missouri State 21 | Marshall 20 | 76.8% | Marshall | Missouri State | MISS | ensemble |
| 2025-09-06 | North Carolina 20 | Charlotte 3 | 21.0% | North Carolina | North Carolina | HIT | ensemble |
| 2025-09-06 | Memphis 38 | Georgia State 16 | 14.4% | Memphis | Memphis | HIT | ensemble |
| 2025-09-06 | Army 24 | Kansas State 21 | 75.1% | Kansas State | Army | MISS | ensemble |
| 2025-09-06 | Houston 35 | Rice 9 | 36.3% | Houston | Houston | HIT | ensemble |
| 2025-09-06 | Tulane 33 | South Alabama 31 | 27.5% | Tulane | Tulane | HIT | ensemble |
| 2025-09-06 | Western Kentucky 21 | Toledo 45 | 61.8% | Toledo | Toledo | HIT | ensemble |
| 2025-09-06 | Ball State 3 | Auburn 42 | 91.8% | Auburn | Auburn | HIT | ensemble |
| 2025-09-06 | Louisiana Tech 7 | LSU 23 | 94.4% | LSU | LSU | HIT | ensemble |
| 2025-09-06 | Boston College 40 | Michigan State 42 | 50.2% | Michigan State | Michigan State | HIT | ensemble |
| 2025-09-06 | Arizona State 20 | Mississippi State 24 | 25.7% | Arizona State | Mississippi State | MISS | ensemble |
| 2025-09-06 | Akron 0 | Nebraska 68 | 90.6% | Nebraska | Nebraska | HIT | ensemble |
| 2025-09-06 | Michigan 13 | Oklahoma 24 | 51.6% | Oklahoma | Oklahoma | HIT | ensemble |
| 2025-09-06 | Georgia Southern 20 | USC 59 | 82.1% | USC | USC | HIT | ensemble |
| 2025-09-06 | Vanderbilt 44 | Virginia Tech 20 | 65.6% | Virginia Tech | Vanderbilt | MISS | ensemble |
| 2025-09-06 | UL Monroe 0 | Alabama 73 | 97.1% | Alabama | Alabama | HIT | ensemble |
| 2025-09-07 | UCLA 23 | UNLV 30 | 57.6% | UNLV | UNLV | HIT | ensemble |
| 2025-09-07 | Tulsa 14 | New Mexico State 21 | 58.1% | New Mexico State | New Mexico State | HIT | ensemble |
| 2025-09-07 | Stanford 3 | BYU 27 | 89.7% | BYU | BYU | HIT | ensemble |
| 2025-09-07 | San Diego State 13 | Washington State 36 | 68.7% | Washington State | Washington State | HIT | ensemble |
| 2025-09-07 | Sam Houston 20 | Hawai’i 37 | 54.9% | Hawai’i | Hawai’i | HIT | ensemble |
Highest-confidence misses
Limited to the 25 most confident misses so the table
stays readable; the complete game-level record is preserved in the JSON artifact.
| Game | Pick | Confidence | Actual | Margin |
|---|---|---|---|---|
| South Florida at Florida | Florida | 85.0% | South Florida | 2 |
| Missouri State at Marshall | Marshall | 76.8% | Missouri State | 1 |
| Army at Kansas State | Kansas State | 75.1% | Army | 3 |
| Arizona State at Mississippi State | Arizona State | 74.3% | Mississippi State | 4 |
| Baylor at SMU | SMU | 67.4% | Baylor | 3 |
| Vanderbilt at Virginia Tech | Virginia Tech | 65.6% | Vanderbilt | 24 |
| Texas State at UTSA | UTSA | 60.2% | Texas State | 7 |
| Liberty at Jacksonville State | Liberty | 60.1% | Jacksonville State | 10 |
| Fresno State at Oregon State | Oregon State | 53.2% | Fresno State | 9 |
| Illinois at Duke | Duke | 52.0% | Illinois | 26 |
Methodology and interpretation
- Train: 2023 and 2024 final regular-season FBS-vs-FBS games.
- Test: 2025 final regular-season FBS-vs-FBS games, with ratings frozen.
- Model: opponent-adjusted Elo with score-margin adjustment and empirical home field.
- Ensemble rows: 40% Elo plus 60% historical spread-derived win probability.
- Brier measures probability error; lower is better and 0.25 is the coin-flip reference.
FAQ
Did this model exist in the 2025 season?
No. This is a retrospective reconstruction, clearly labeled as such. It does not claim
these picks were published before kickoff.
Why only FBS-vs-FBS?
It gives one comparable
opponent pool and avoids pretending schedules across NCAA divisions are interchangeable.
Is this betting advice?
No. It is a model audit and historical scorecard.
Related: 2026 CFB predictions
and the NFL Facts Stats Hub.
Sources
Retrospective backtest rebuilt 2026-09-06.
No mock data and no 2025 result used to fit the frozen ratings.