Case file · college football
The Rankings Board
Who beat whom, and how hard the schedules were.
Compare file
result (winner to loser)
marker: lost to a team outside the Power Four
hypothesis
Wins over teams outside the Power Four don't count.
Hypothesis desk
Pick winners for upcoming games.
Enter your rankings
Drag the top 30 into the order you think is right. The boards and the playoff field follow your order.
Outcome file
Knock-on effects on the top 25
Schedule strength by week
Rankings as a plain list
How the ratings work
- Results are opponent-adjusted: beating a good team on the road counts for more than beating a weak one at home, and blowouts count with diminishing returns.
- Games between top-group teams (the four Power Four conferences plus Notre Dame) count more than other games. A top-group team's win over a team outside that group is ignored; a loss to one counts in full.
- The score blends results, forward strength and the strength of the schedule played so far. Schedule strength is the average rating of past opponents, with road opponents counted as stronger.
- The auto-bid race simulates the rest of each Group of Six conference's schedule from the same ratings. A conference's champion is the team with the most conference wins, and the best-rated of the six champions gets the bid. The circles mark teams with a real shot at it.
- The playoff field follows the 2026 format: the four Power Four conference champions and the best Group of Six champion get automatic bids, Notre Dame gets one if it finishes in the top 12, and the rest go to the highest-ranked teams. One spot is held for the Group of Six champion since only the Power Four and Notre Dame are rated here. In the field built from your rankings, the best-ranked team in each conference stands in for its champion; in the simulations, the champion is the team with the most conference wins.
- Three strikes: a team with three losses is out of the playoff, conference champions included. A one-loss team can't afford to lose two more.
- Injuries are not modeled yet. Settings were fitted to how well the ratings predict the next week's games.
- Data: CollegeFootballData.com, refreshed by the author. The page reads a static data file; no API key or private service is involved. Team logos belong to their schools and are shown only to identify teams.