2025 College Football Rankings — Sun Belt

Rec CollegeFootballData

Wins–losses. Sorting by it uses win percentage, so 3-0 outranks 4-3.

Elo CollegeFootballData

A results-based rating: teams trade points for every game, and beating a strong team is worth more than beating a weak one. It knows who won and by how much, and nothing else — so it needs games before it means anything.

SP+ Bill Connelly, via CollegeFootballData

An efficiency rating built from play-by-play rather than results, expressed as a predicted scoring margin against an average team. It is published before kickoff, which is why this board falls back to it early in a season.

FPI ESPN, via CollegeFootballData

ESPN's Football Power Index — its own predicted margin against an average opponent.

SRS CollegeFootballData

Simple Rating System: average scoring margin, adjusted for the strength of who you played.

Schedule Gridpex, from the season's results

How hard the season has been: the average rating of the opponents a team has actually played, ranked against every other team — #1 is the toughest schedule in the country. The count beneath it is how many rated opponents that average is built from, which in September is often one. Who has it easy from here is in the road lists below the table.

AP · Δ vs AP Associated Press

The human poll of sportswriters. Δ is how far this board's order differs from theirs — positive means we have the team higher than the voters do.

Off / Def (CORE) Gridpex — ours, published nowhere else

Opponent- and situation-adjusted efficiency, in points per 100 qualifying plays, above average. Offense is better when high, defense when low. It is the one rating here that publishes how sure it is: /ratings shows the range each team could plausibly occupy, and most of the table is wider than the gap between neighbouring ranks.

Every rating here is bought in except CORE, which we build. How we build it →