About The Davis Rankings
Who runs this
The Davis Rankings (TDR) is built and run by Davis, a college football fan who got tired of rankings that move on reputation instead of results. It's an independent, one-person project with no ties to any conference, network, or school. Questions, corrections, and arguments are welcome on the contact page.
Polls are opinions. Rankings are data.
The AP Poll and Coaches Poll are surveys - voters rank teams by feel, reputation, and what they watched on TV that week. The CFP rankings come from a committee voting behind closed doors, with no published formula and no requirement to explain a single placement. All three are opinions, dressed up to look like results.
The Davis Rankings works differently. Every team's score comes from the same published formula applied the same way to every team: wins and losses first, then recency, road wins, opponent quality, and a real penalty for losing to an FCS team. No committee, no ballot, no brand-name benefit of the doubt. See the exact weights on the Methodology page, and click "Show calculation" on any team to see the math behind its score.
Why this exists
College football's postseason is gatekept by exactly the people voting teams in and out - the same conferences that dominate the committee's trust also dominate the committee's results. In practice, that's meant the Big Ten and SEC treated as the sport's haves, and the Group of 5 treated as the have-nots: good enough to be scheduled as a cupcake, rarely good enough to be ranked ahead of a Power Conference team with a worse record. The ACC and Big 12 sit in the middle of that system - not haves, not have-nots, just given enough playoff access and TV relevance to keep buying into a structure that was never built to let them win it. That's a popularity contest, not a ranking. The Davis Rankings scores every team by the same math regardless of conference, logo, or TV contract - if a Group of 5 team earns a better record against good opponents than a Power Conference team, it ranks higher. That will look controversial to anyone used to the polls. It's supposed to.
Good teams win games. Period.
Strip away the preseason hype, the brand name, and the TV contract, and a season is just a record and a schedule. That's what the formula scores. It doesn't hand out a head start for conference membership, and it doesn't hand out a handicap for it either. Win, and the rank moves up. Lose, and it moves down. Nothing about who you play for changes that math.
Historically weak teams get read differently depending on who they are. A Power Conference team that stumbles gets "that conference is just tough top to bottom." A Group of 5 team that wins gets "that conference must be down this year" - the same result, read as proof of opposite things, depending on the jersey. The Davis Rankings doesn't read results that way. A win is a win. A loss is a loss.
Good teams win games. Great teams don't lose.
Undefeated has no worst loss - because there isn't one
The argument goes: "Sure, that Group of 5 team is undefeated, but my one-loss Power Conference team's only loss was to a really good team, so we should be ranked ahead of them." That argument treats a loss like a neutral data point to be explained away, instead of what it actually is: a loss. An undefeated team has no worst loss to defend, because it has no losses, period. A "quality" loss doesn't cancel the loss - it's still one more loss than a team with zero. Win your games, and you don't have to argue about who you lost to.
There is no mulligan
Baseball plays 162 games. The NBA and NHL play 82. A bad night gets buried under a hundred more chances to prove it was a fluke. Football plays around 12. There is no long season to hide a bad loss in and no sample size big enough to average it away - every single game is roughly 8% of the entire résumé. That's exactly why the formula weighs results as heavily as it does: in a 12-game sport, "the better team lost" isn't an excuse the schedule has room for. You lost. It counts.
If you don't like it, win your games
The formula doesn't look at history, conference reputation, or preseason polls - only this season's results. If a team wants to rank higher, the only lever that moves it is winning, not pedigree, not a fan base, not a name. A team that leans on any of that instead of its own record isn't looking for a fair ranking. It's looking for a handout.
Rankings reset weekly. Polls change little.
The Davis Rankings recomputes from scratch every week off that week's actual results, so a team's rank is fluid - it can move several spots in either direction on a single Saturday. The AP and Coaches polls are built to protect a voter's earlier ballot - preseason rank tends to anchor the whole season, and a team can lose and barely fall, or win and barely rise, because the poll is voting on its own prior week more than it's voting on the game just played.
Most of our score is two things: the record (55%) and the latest game (15%). So a loss shows up right away. It is a measure of how good a team looks right now, not how good it looked in August. That is most of the model; location and opponent strength make up the rest. In 2025, a top-25 team moved 3.9 spots a week in our rankings and 2.2 in the AP, and ours brought in 6.9 new teams a week to the AP's 2.7. See the numbers for every season. If a team doesn't like the hit it takes for a loss, the fix is simple: don't lose.
Conference is not a choice
Teams don't get to pick their conference - they're where realignment and history put them, and neither one is about who's actually good on the field. Most programs would take a Big Ten or SEC invite tomorrow, but not because of who's already in those conferences. It's the money: bigger media contracts, bigger payouts, bigger exposure. That's a business decision, not a football one - which is exactly why conference has no place in a ranking that's only trying to measure football.
How the site is built
The Davis Rankings is an independent project. It is a static website: there is no database and no server running behind it. A scheduled job pulls the latest results from the College Football Data API several times a week during the season, recomputes every team's score with the formula described on the Methodology page, and publishes the results as plain data files that your browser reads directly. Past seasons stay browsable with the Season and Week selectors on the home page.
What you'll find on the site
- Rankings - every FBS team, each week, with the AP, Coaches, and CFP positions alongside for comparison, and a conference filter and search.
- Weekly summary, biggest winners and losers - who moved the most since the previous week.
- Weekly awards - recognitions computed from that week's results, such as the Coin Flip of the Week.
- Upcoming games - the next week's matchups between ranked teams, ordered by how big the game is.
- Conference rankings - conferences ranked by their combined non-conference record.
- Team pages - each team's schedule, results, and the full factor-by-factor math behind its score.
Limits worth knowing about
A results-only formula is honest about what it measures, and it is also blunt. It ignores injuries, margin of victory, and how a game was played, and in the first few weeks of a season a small number of games makes scores swing a lot. The Davis Rankings is a way to look at the season through results alone. It is not a prediction of who will win next Saturday.
FAQ
How is this different from the AP, Coaches, or CFP
rankings?
Those are opinions - people voting on who they think is good. The
Davis Rankings is a fixed formula run against real results. Same
inputs, same weights, every team, every week.
Why do you show the AP/Coaches/CFP polls next to your
ranking if they're "just opinions"?
For comparison. Seeing where the polls rank a team next to where the
data ranks it is the whole point - the gap between the two is the
story.
Does a Group of 5 team get scored the same way as a Power
Conference team?
Yes. Conference is not a factor in the formula at all. The
opponent-quality factor already accounts for schedule strength, so a
team that plays and beats good opponents is credited for it
regardless of which conference those opponents belong to.
Why does one loss to an FCS team hurt so much?
An FBS team losing to an FCS team is a result that almost never
should happen, and it says more about that team than almost any
other single game on the schedule. The formula treats it that way on
purpose. See Methodology for the
exact weight.
What data powers this?
Game results, schedules, and poll data come from the
College Football Data
API. The ranking math is ours.
How often is it updated?
Every week of the season. The rankings update on Sunday morning, once Saturday's games are in, and again on Sunday afternoon
when the AP and Coaches polls are out. Runs through Wednesday catch late games, the CFP committee rankings and corrections.
Thursday, Friday and Saturday morning refresh kickoff times and TV listings and add the results of weekday games to team schedules, without
changing a ranking. Details are on the methodology page.
Is this affiliated with the NCAA, a conference, or a
network?
No. The Davis Rankings is independent and has no affiliation with
the NCAA, any conference, or any broadcaster.
Questions or feedback?
See the contact page, or write to
justwin@thedavisrankings.fyi.