2024 College Football Rankings - 11/16/2024 edition

The Week that Was, and Comparing Bubbles

Well, Georgia defeated Tennessee in a key SEC match and put themselves in pretty good shape to still win the conference. Tulane ended any hope of an Army-Navy AAC championship game pretty authoritatively, and late at night Kansas - arguably the unluckiest team in the country this season - went to Provo and knocked BYU from the ranks of the unbeaten. Meanwhile, Oregon survived an ugly and perilous trip to Camp Randall in Madison, where many a good team has gone down, and Clemson won late at Pitt keeping their hopes alive in the ACC.

The rankings will be below next, but one of the things I was curious about was to look at the current actual College Football Playoff rankings and breaking it down into bubble terms. The rankings themselves are Big Ten heavy at the top which caused a lot of consternation from SEC fandom, but what does it really mean?

Turning the Committee Rankings into a bracket looks like this:

Automatics: Oregon (Big Ten), Texas (SEC), Miami-FL (ACC), Boise State (G5), BYU (Big 12)

At Large: Ohio State (2), Penn State (4), Indiana (5), Notre Dame (6), Alabama (7), Ole Miss (9), Georgia (10)

Bubble (last 3 in): Alabama, Ole Miss, Georgia

Bubble (first 3 out): Tennessee (11), SMU (13), Texas A&M (15)

First round matchups: BYU at Ohio State, Georgia at Penn State, Ole Miss at Indiana, Alabama at Notre Dame

To me, the big surprise is Notre Dame - a good team that keeps improving - at #6. Like, did the Northern Illinois loss matter at all? Northern Illinois is a 6-4 MAC team! That has to be a stiffer penalty than this. They don't deserve quite that high a seed to me. The other surprise to me is SMU, who IS the only unbeaten left in ACC play and should have gotten more automatic bid consideration.

Let's take my rankings from this week:

Automatics: Oregon (Big Ten, #1), Texas (SEC, #2), Boise State (G5, #4), SMU (ACC, #8)

At Large: Ohio State (3), Indiana (5), Penn State (6), Georgia (7), Alabama (10), Notre Dame (11), Miami-FL (12)

Bubble (3 in): Alabama, Notre Dame, Miami-FL

Bubble (3 out): Army (13), Ole Miss (14), South Carolina (15)

Personally, the top of the rankings look reasonably similar to the committee, but again, the committee seems to be ignoring early losses in terms of who deserves consideration in a way that doesn't make sense. When you have such a small sample of games, they all have to count. For my rankings, Ole Miss' losses to Kentucky and LSU have looked worse over time. Meanwhile South Carolina has been gaining steam to the point that their resume next to Tennessee is pretty close. Army is still an unbeaten conference champ. The committee's ranking of Army will be very interesting if they beat Notre Dame this weekend.

The Rankings

As always, the methodology is here:

2024-25 Men's College Basketball Rankings #1 - Thru 11/14/24

Since I am ranking and writing a bit more, we will try to do the Field of 68 this season. This is more to see blossom over the season as more games get into the model - and more good teams play each other. College basketball teams end up playing somewhere between 50 and 65 percent of their games in conference compared to more like 70% with college football - so there are just more games to use for a pairwise model.

Of course the season is only 2 weeks old - so while there is enough data for a Bradley-Terry sort of model to converge, the error terms are pretty wild. At the same time, who cares? It is interesting to see how the season evolves. The teams that are high now here will drop off pretty quickly (I'm looking at you George Mason) - and losses to good teams are actually helpful.

The Method

Basically this is the same as what I am doing for football. The components.

1. Bradley-Terry model using a small margin of victory component. This is more or less a logistic regression. A team gets 1 point for having a better rating than a competitor.

2. Head to Head record. A team gets as many points as net wins over a competitor. (no negative points for a team with a losing record)

3. Record vs Common Opponents. A pairwin here is worth 1 point.

4. Regular season leader. During the season we will use the highest rated team with the fewest conference losses. A pairwin here is worth 0.5 points. Based on how the committee seeded teams last year, 0.5 points fit the data better than 0 or 1.

   
   

For each component, a team is compared to the other 365 teams (all non D1 are treated as a single opponent) and points are awarded. Whichever head to head has more points counts a pairwin.

For example: Team A has a better rating than Team B, but they had no head to head, Team B did better against common opponents, and Team A was the regular season champ of its conference. In this example Team A has 1.5 points (1 for rating, 0.5 for reg season champ). Team B has 1 point for common record. Team A would get the overall pairwin.

Half pairwins are used for ties. The team with the most pairwins is the #1 overall team. If there is a tie in pairwins, the Bradley-Terry rating is used as a tiebreaker.

The Field of 68

Through the games of 11/14/2024. With so few games, this is genuinely hilarious. The SoS is based on the perspective of a .500 team (using the Bradley-Terry rankings as weights)

The Bubble

The first 4 teams are in the field with byes to the Round of 64. Teams 5-8 are the play-in teams. The others are out in order. As you can see, the model is pretty dumb right now, but it is fun to watch it get smarter.

2024 College Football Rankings - 11/9/2024 Edition

Well, the SEC's circular firing squad continues apace with Georgia getting hammered by Ole Miss over the weekend. The SEC's parity has been arguably THE story of the college football season - we could very easily have a 2-loss conference champ. Now, this complicates any good attempt to model, especially contrasting it with a sport and committee which has historically reward the whole winning games things. In some systems (foreshadowing!) it impacts the rankings in a way that is probably unfair. I understand how the real life College Football Playoff Committee is keeping an SEC team in the Top 4.

But before, we discuss that, we need to talk about my alma mater ...

Miami falls from the ranks of the unbeaten, and not just that - but they lose their grip on the ACC's championship berth, at least with my rankings.

In 2003, I was on campus when another fanstorming happened on the Flats:

Just like 2024, then we took down the goalposts and carried them to the university president's residence. Georgia Tech has been up and down as a program, but the ups with a big college football win have always been special.

The Rankings

(through games of 11/9/2024 -rankings method here)

Notes:

• As a matter of course, I would probably have Texas ahead of Boise State, despite my own numbers disagreeing. We know Boise State won't end up with a Top 4 seed in real life. That said, their resume is surprisingly strong especially with the almost win at Oregon.

• At the moment, Tennessee, Texas A&M and Texas are the 3 one-loss SEC teams - and Texas gets the SEC's champion's bonus as a result. But Tennessee gets Georgia this weekend which could upend this whole thing even more.

• Alabama kept its season alive with loser leaves town blowout of LSU. At #20, it looks like getting back into the Top 12 is too big an ask.