If you had to describe the principal “””subdivisions””” of the United States of America, you’d do well to say that there are 50 states, 1 district, and 6 outlying areas. (ISO 3166:US). EB, you might even be able to identify them all on a map some day. But what if you tried the same thing with the United Kingdom of Great Britain and Northern Island? How would that Sporcle quiz look? Well, you’d end up with 32 council areas, 27 two-tier counties, 11 districts, 78 unitary authorities, 36 metropolitan districts, 32 London boroughs, and 1 city corporation—all a part of one province and three countries, two of which are sometimes grouped together as one unit “for completeness”. (ISO 3166:GB). And don’t think that this difference is just a product of the fact that each US state has counties (“parishes” in Louisiana, “boroughs” in Alaska); the UK’s non-London subdivisions similarly have civil parishes (“communities” in Wales). (Wikipedia has a comprehensive map to illustrate the point.)
The administrative structure of the UK compared to the US is much like the structure of college football compared to the NFL. Starting with the familiar, the NFL’s structure today is traceable to a purposeful, thoughtful creation reached after deliberation (arising from a merger of the old-NFL and the AFL), made relatively recently (in 1966), involving a limited number of entities (24 clubs up to 32). College football’s structure today stems from the fact that the sport arose organically, owing back to matches played before the sport of football was recognizable and between students at colleges which were founded before the country was, and involves hundreds upon hundreds of colleges (and military academies). Add on to that the fact that colleges, unlike NFL teams, sponsor multiple sports (more like European soccer clubs) and those other sports can impact how college football works.
What follows is a quick overview of the “structure” of college football today, with a focus on the highest level of college football—Division I FBS football—you’re likely to see on ESPN, FOX, CBS, etc. I considered taking an entirely historical perspective on how we got to the current state of affairs, but I’ll try to limit the backstory here. Instead, I’ll answer 3 questions you are probably wondering about college football: 1) What is the NCAA?, 2) what are conferences?, and 3) how does the college football postseason determine a champion?
What does the NCAA do for college football?
Running any sport league is a collective action problem. Each competitor (team or individual) seeks an advantage over every other competitor. But in some cases, if each competitor engaged in that competitive behavior whenever it could, all competitors would suffer over repeat play. For example, imagine that eye gouging was permitted in the UFC. Hypothetically, some fighter could be the best at eye gouging and benefit from its permissibility in the context of whether he wins or loses each bout. But it is probable that the entire UFC promotion would suffer and bring in less money as a result of this rule because no paying viewer wants to watch an eye-gouging fest, especially after the most exciting athletes avoid fighting in UFC bouts so as not to be blinded. Thus even our top eye gouger would suffer in the long run as less money was poured in to his pocket, despite his perfect win-loss record.
Creating collectively-beneficial rules (like banning eye gouging) is not the end of the collective action problem, however. Those rules must be enforced by someone other than the competitors, because each competitor is strongly disincentivized to police its own violations of the rules. See, e.g., J. S. Russell & Alister Browne, Performance-enhancing drugs as a collective action problem, Journal of the Philosophy of Sport, Vol. 45:2 (2018-02-25), pp.109-127 at 111 (individual athletes); Evan Zepfel, Baseball’s Collective Action Problem, Sports Agent Blog (2013-08-08) (team sport).
The NCAA is the organization that solves, 1) for most schools, 2) in most sports, 3) the most important of these collective action problems. But lets pause to examine each of those “most”s. First, not every college is a member of the NCAA. Many smaller colleges are members of the NAIA (National Association of Intercollegiate Athletics) and many junior colleges (“jucos”) are members of the NJCAA (National Junior College Athletic Association). Second, the NCAA does not govern all sports which colleges may choose to sponsor. For example, college rugby is governed by USA Rugby (the same organization which determines who to send for Team USA Rugby in the Olympics) and college squash by the appropriately named CSA (College Squash Association).
Third, the NCAA does not solve every collective action problem which is why this blog is necessary. The NCAA’s main function is setting rules for player eligibility. This is where it gets roasted for its notion of “amateurism” and “student-athletes,” but it is obvious that without any rules on player eligibility, one school could field a team of students that are actually ringers to the disadvantage of every other school who maintains a semblance of anything other than a minor league for professional sports. Everything else is pretty much outside the NCAA’s wheelhouse, including prosecuting kiddy rape and even setting a common testing protocol for Covid-19 to prevent spread among teams, a classic collective action problem.
For our purposes, there are a few major collective problems to stress that the NCAA does not solve for: allocation of television rights (see NCAA v. Board of Regents of the University of Oklahoma, 468 U.S. 85 (1984)), scheduling (beyond setting a few maxima), or selecting a national champion (only for FBS, the highest division of college football). Instead, these problems are mostly solved by the conferences.
What are conferences?
Congratulations, you were just made the athletic director at Generic State University (Go Vanilla Beans!). You are in charge of determining who GSU’s football team will play next year. You’ve got 10 open slots and about 130 schools in the same position as yours. Let’s start the negotiations between you and each of the other 130 ADs as you try to hammer out your school’s schedule. ABC State just called, which week do you want to play? Whose referees will call the game? Is it at your stadium, or are you travelling? How should you split up the ticket revenue/gate from whoever’s stadium it is? (Oh, and do you get an amount of tickets set-aside just for your fans?). Who gets what percentage of the television revenue from broadcasting the game? What do you do when DEF Tech makes a better offer? Please repeat for all the open dates on your schedule.
Conferences are how to simplify many of the above problems. A conference is a group of around 8–14 schools which enter into an arrangement for scheduling purposes that fill up most of each school’s schedule for a year. For games between schools within the conference, there is a common pool of referees, the television revenue gets shared between all the schools, and the gate and amount of visiting tickets is set, and the schedule is predictable and balanced (at least between road and home games). It is up to each school to fill out its remaining schedule with other schools out of conference (OOC), but that is only 3–4 games a year. At the FBS level, the most wealthy and powerful schools will often pay in cash some poorer, cupcake schools to visit and get whomped and schedule a home-and-home with a roughly equal school where each school hosts the other once in a certain timeframe. (To keep track of all these latest out-of-conference agreements, use FBSchedules.com)
There are currently ten FBS conferences (each division of NCAA football—or JUCO football, etc.—will have different conferences). Five FBS conferences are more relevant and wealthier and make up the “Power 5”: The Big Ten Conference (Big Ten or B1G), the Southeastern Conference (SEC), the Atlantic Coast Conference (ACC), the Pac-12 Conference, and the Big 12 Conference. The other “Group of Five” conferences are the Mid-American Conference (MAC), the Sun Belt Conference, the American Athletic Conference (AAC), the Mountain West Conference (MW), and Conference USA (C-USA). There are still several major schools that are not in any conference and go about all the complications mentioned above, most notably the BYU Cougars and the Notre Dame Fighting Irish. Notre Dame faces fewer difficulties than other independents because it 1) has recently entered into an agreement with the ACC which fills up many games on its schedule each year, 2) has an ongoing home-and-home arrangement with one of its rivals, the Southern California Trojans of the Pac-12, and 3) is one of the “blue bloods” of college football who can negotiate its own television deal with NBC and with whom schools would love to schedule a game.
To the casual college football viewer, only the Power 5 schools and Notre Dame matter because only those schools have any realistic shot of winning the national championship in a given year. But an FBS school really doesn’t “win” the national championship as much as it is “selected” as the winner by the media. Really. Read on.
Who determines the national champion?
If there are any Big Ten teams that shoot for a national championship, they’re damn fools. You play to win the Big Ten championship, and if you win it and go to the Rose Bowl and win it, then you’ve had a great season. If they choose to vote you number one, then you’re the national champion. But a national champion is a mythical national champion, and I think you guys ought to know that. It’s mythical.”
Bo Schembechler (Head Coach of the Michigan Wolverines from 1969 to 1989), July 1989.
We end with a necessary history lesson.
Before there was such thing as a college football post-season, there was just the season. In the early days of college football, each team played all the games on its schedule and that was that. After all, there was no governing body that all the schools signed up to that forced them to play more games somehow. Each school and conference was just doing its thing in its own corner of the country. You had no guarantee that the two best teams could meet each year, or even guarantee they had a common opponent: the best team in the “East” (say, Yale) could go undefeated along with the best team in the South (Sewanee) and the best team in the “West” (Michigan) and no one could settle who was better on the field.
That doesn’t mean that football fans didn’t make ways to argue about which team was better. It’s a human instinct to debate the impossible like whether the Hulk or some shonen would win in a fight. But in the mid-1920s, a nerd named Frank G. Dickinson finally used math to try to determine a formula for who was the best football team at the end of the season based simply on the final scores of games played. See Zach Pekale, Before the AP poll, the Dickinson System ruled college football rankings, NCAA (2020-11-06). The Dickinson System became the first widely accepted “selector”—a system that chooses a national champion. Teams were even given a trophy at the end of the season for being ranked 1st from the Dickinson System. Today, there are several computer-based selectors (like Sagarin‘s Rankings and the Billingsley Report) that are considerably more complicated than the Dickinson System but ultimately depend on statistical results from games.
But stats aren’t the only way to settle the debate of who the best team is. The biggest innovation in determining the best college football team came in 1936 with the Associated Press (AP) Poll, which (unlike the Dickinson System) is still relevant today. The AP Poll is simple: a group of journalists based in different parts of the country rank every week who they think are the best teams in the country (today, each voter ranks their Top 25). Then the votes are aggregated. Each school gets 25 points for a first place vote, 24 points for a second place vote, and so on. Whoever has the most points is #1, and if you are #1 at the end of the season you can go ahead and call your team national champions. When people talk about the number of national championships a school has, or talk about the 1972 USC Trojans as national champions, they usually care about the national champion that the AP Poll selected. The main competitor to the AP Poll is the Coaches Poll, which works the same way but uses current FBS head coaches instead of journalists.
Note that neither statistical rankings nor polls need a postseason to work. If you’re just using stats or opinions to determine “who is the best team,” the only thing another game gives you is an extra data point. And this was a good thing because in the early days of college football there was no post-season.
The first post-season game was on New Year’s Day, 1902 in Pasadena, California. Essentially, the organizers for some parade agreed to pay Michigan and Stanford to bring their football teams to town for everyone’s entertainment. The 10-0 Michigan team crushed the 3-1-2 Stanford team 49-0 (under a different scoring system from today) and the organizers decided to run chariot races instead of football for the next decade or so because it would be more entertaining. But eventually football came back, and the Rose Bowl Game became an annual New Year’s Day fixture. Thus was born the bowl game.
While the Rose Bowl is The Granddaddy of Them All, other bowls eventually followed. The formula for making a new bowl game was simple. Essentially, a bowl committee in some warm-weather city decides that it would make sense to have a football game in January or December to bring tourism to the city, and two football teams agree to take a nice vacation in the winter and get another game in. After having only the Rose Bowl from 1916–1934, by the 1936 season (the year of the first AP Poll), college football had five regular bowl games: Rose (Pasadena, CA); Orange (Miami, FL); Sugar (New Orleans, LA); Sun (El Paso, TX); and Cotton (Dallas, TX).
Of course, not every team played in a bowl game at the end of season. In fact, until 1946—when the Big Ten, the Pac-12’s predecessor and the Rose Bowl signed an agreement whereby both conferences would send one team to Pasadena each year—the Big Ten did not let any of its teams play in bowl games. (The Big Ten would continue for decades to have idiotic rules regarding bowl games, including a Rose-Bowl-or-bust rule and a no-repeat rule, where a team could not play in the Rose Bowl two years in a row). As far as determining a national championship, the AP and Coaches Polls used to declare their national champions before the bowl games were played. This was only fixed after the 1973 season, when Alabama (AP #1/Coaches #1) lost to Notre Dame (#3/#4) in the Sugar Bowl. Because the final Coaches Poll was taken before the bowl games and the final AP Poll was taken after, Alabama and Notre Dame both claim the 1973 national championship.
As you can imagine, the bowls, as a collective, became influential in determining the national champion by virtue of having the ability to invite top teams to face each other at the end of the year. But there was a problem in that no individual bowl had the ability to ensure that it would get the two best teams every year. The best Big Ten and Pac-8 teams were contractually obligated to play in the Rose Bowl, the best SEC team in the Sugar, and the best Big 8 team in the Orange and there was no guarantee that those bowls would settle any national championship debate.
Eventually, all of the major bowls got together into the BCS (Bowl Championship Series) and agreed that they would invite whoever the #1 and #2 team in the country were to play against each other (!!!) in the national championship game and that each bowl would rotate hosting that national championship game as opposed to its normal contractual arrangement. Controversy reigned however in how to determine who the #1 and #2 teams were before bowl season. Over its lifespan the BCS used a mix of statistical rankings (like Sagarin) along with the AP and Coaches poll and weighted each 1/3. People got upset like every year on who was left out. Instead of reverting to an argument-based system like THE LORD intended, in 2014 FBS college football finally got a playoff system: the College Football Playoff (CFP).
Like the BCS, the CFP involves all the major bowls getting together and rotating who hosts the games. These are the “New Year’s Six” bowls: Rose, Sugar, Orange, Cotton, Peach (Atlanta, GA) and Fiesta (Glendale, AZ). The playoff consists currently of 2 semi-final bowl games, the winners of which play in a neutral-site national championship game. So instead of having to pick the #1 and #2 team in the country like the BCS did, the CFP has to determine the #1–4 teams in the country and place them in the semi-final games. Also unlike the BCS system, the CFP does not use statistical rankings or polls to determine who the top teams are. Instead, and I am not joking, a “Selection Committee” sits around in a smoke-filled room in Grapevine, TX and decides who the best teams are based off some set of generic criteria. This Selection Committee releases its first set of rankings about midway through the season and its last set of rankings determines the bowl game match-ups.
For anyone looking to understand the playoff system right now, that’s about all you need to know. Fans of FBS schools today focus on making sure that their team finishes in the Top 4 of the CFP rankings at the end of the year (which probably involves winning their conference but not necessarily) so they can get into a semi-final and have a shot at the “real” national championship. But the CFP has not been without its own controversies so far, particularly its poor rankings of the Group of 5 teams. In 2017, the undefeated UCF Knights were not ranked in the Top 4 by the committee, then won their bowl game against UCF, and subsequently declared themselves national champions. And who’s to say they’re not? It’s part of the beautiful mess that makes college football the best sport to watch.
This was Just My Blog