Guest Post: In Defense of the Bowl Championship Series

Related Post Roulette

41 Responses

  1. Steven Donegal says:

    That’s all very nice, but the current system not only makes the regular season inconsequential for most teams, but makes all the rest of the bowl games inconsequential as well. Seriously, does anyone cares who wins the Orange Bowl or the Rose Bowl (or any other bowl for that matter). The whole point of bowl games at this point is to sell TV ads and give the participating teams an extra few weeks of practice.Report

    • trumwill in reply to Steven Donegal says:

      Cincinnati is going to its first BCS game in history. It matters a whole lot to them. USC is probably staying home from their first BCS bowl in seven years. Mack Brown caught endless grief for his inability to win his conference and go to a BCS game. This all mattered a great deal to the schools involved.Report

  2. ed bowlinger says:

    There simply is no reasonable way to winnow 120 teams down to one champion. It’s just not possible. The FBS ought to be divided into three 40-team divisions, with relegation and promotion mechanisms. Teams may only play out-of-conference within their division. Otherwise the conference structure is maintained, along with the bowl tie-ins for conference champions. Teams with winning records on a divisional bubble play in a promotion/relegation bowl game. (A bowl game with consequences!)Report

    • Trumwill in reply to ed bowlinger says:

      James Joyner proposed something like that. It’s a great (and fair) way to handle things, but it is completely and utterly incompatible with college football’s history and traditions. Even moreso than a playoff system would be.Report

  3. Zach says:

    Speaking of the BCS, a fumbled snap (I assume, watching play by play online… maybe blocked) on a Pittsburgh extra point may very well be the difference between Cincinnati or TCU going to title game if Texas loses today.

    I think you’d satisfy almost everyone with a 10-team playoff; 6 conference winners are in automatically, four at-large seats, seeds are determined by the current BCS computer rankings (though I’d prefer these be adjusted to actually rank team quality rather than popularity & ability to avoid hard nonconference schedules), and the bottom 4 teams have to win an extra game that would be played next week.

    I don’t think it makes sense to compare the possible BCS tournament situation to the NFL; there’s no way that a tournament would admit enough teams to wind up with the 2007 Giants / 2009 Cardinals example. MLB is probably more analogous; teams are left out that could definitely win the World Series in a larger tournament.

    College football isn’t exactly hurting, but what would help the most would be a postseason system that allows for (or encourages) competitive nonconference schedules.

    Lastly, there’s an argument that the national title should go to the team who is the best at the end of the season rather than the team who did the best throughout the season. A playoff is an imperfect way to identify that team.Report

    • trumwill in reply to Zach says:

      A 16-teams tournament would let in 9-3 teams and essentially give them the same chance at a title as a 12-0 team that beat them during the season. There may be an argument for that, but I don’t buy it.

      There is no way you’re going to get a consensus on just 10 teams. Give me ten teams and I will give you an 11th just as worthy 4 years out of 5. What starts at 8 will end up at 32.Report

      • trumwill in reply to trumwill says:

        That should say 16 or 32.Report

      • Zach in reply to trumwill says:

        I agree that you can always argue that the 11th team is just as worthy as the 8th, 9th, or 10th. I can’t think of a college season in which you could make a good argument that teams left out of a 10-team playoff (6 automatic + 4 at-large) would not be huge underdogs against the #1 ranked team. This would be a risk with an 8-team playoff; if you look at the BCS selection rules, Boise State would probably not make an 8-team playoff this year (Florida would probably beat them for the at-large bid, and TCU ranks higher for the non-big-6-but-still-ranked-high automatic bid). A 10-team playoff would essentially guarantee that no undefeated small conference teams or one-loss teams in strong conferences are left out. Penn St. would be on the outside looking in, and I don’t think there’s any case to be made that they should be included.

        Note that this becomes complicated if Notre Dame ever decides to become a top tier team again. 10 would usually be enough, but there’s no obvious reason to go beyond 12. I like having fewer than 16 to give some advantage to the top teams that balances out random luck. 16 is probably the practical maximum — 3 weeks in a row, a week off, and the title game would end the season close to where it is today.

        This season is a good argument for a playoff; Florida and Alabama are likely the best two teams, but we’ll end up seeing Alabama play Texas, TCU, or Cincinnati instead.Report

        • kth in reply to Zach says:

          In my ideal system, at large bids would only be awarded to teams from conferences that didn’t get automatic berths. We don’t necessarily want the 8 best teams, but 8 teams containing every team with a plausible claim to be the best. Florida just lost any title to such a claim, even though they are probably better than teams that (because they are relatively untested; e.g., Boise State) still can make such a claim.

          My scheme would give automatic berths to the 6 BCS conference winners*, and exclude those conferences from consideration for the at-large berths.

          *though, really, the Big East and ACC aren’t power conferences (nor is the Big 10, but at least they used to be), and in a perfect world only the PAC, SEC, Big 12, and (maybe) the Big 10 champs would get automatic berths.Report

          • Zach in reply to kth says:

            The Big East has been better than the Big 10 and ACC of late; ironic recalling how much people fretted about how terrible the conference would be at football when Miami, BC, and Syracuse got poached in 2005. If the Big 10 and Pac 10 had championship games, I could see only taking one team per conference. Those conferences are too big to play every other team (unlike the Big East) and too small to split into two divisions and have championship games w/in the current rules (unlike the SEC, ACC, and Big 12).

            Of course there’s the problem of Nebraska possibly beating Texas in a couple hours. A team with a far worse record and strength of schedule can win a conference based on a single head-to-head game. And out-of-conference games generally don’t count.Report

          • Trumwill in reply to kth says:

            Ditto what Zach said about the Big East.

            Also, like I said in the post, you can’t satisfy both of the two main groups that support a playoff system with 8 teams. Either you satisfy those that want to see the best 8 teams play or you satisfy those that want every team to have a chance at a championship (if they win out).Report

        • trumwill in reply to Zach says:

          Add a playoff of any size and the goalposts move. Penn State need only argue that they are better than some team that got in. Just a couple years ago a two-loss Georgia team made that argument against LSU.Report

        • Trumwill in reply to Zach says:

          We already saw Florida play Alabama, though. I want conference titles to actually mean something. I am not in favor of teams that can’t even be the champion of their own conference turning around and saying that they’re the champions of Division I-A. I realize that the current system allows that for, but it’s much more of a rarity than it would be with a playoff system.

          This season is a good argument for a playoff. Not because it won’t allow for a Florida-Alabama rematch, but because a number of teams did everything they could to win a national championship and will not get the chance. Florida had their chance.

          If there was a way to give Cincinnati and TCU a shot at the title without also giving a chance for 9-3 Virginia Tech to pretend their losses never happened, I’d be in favor of it. But it’s extremely difficult for me to imagine them staying at anything less than 16.

          That’s the reason why I added hard metrics to my playoff proposal. Expressly require that a team wins their conference and 10 games answers the question “How come this team and not that team?”Report

          • Zach in reply to Trumwill says:

            I agree; I’d rather the national championship game be between Alabama and a team that’s not Florida if we’re only allowed a single game. My point is that you seem to want to determine who the best team is outright, and the best way to do that is to have the two best teams play each other. If one were to handicap games between all of the top teams, Florida and Alabama would be favored across the board.

            Using your NFL example, let’s say the Saints win out except for one loss to the Colts, and the Colts win out except for one loss to the Lions (I know this isn’t actually scheduled)… who’s the best team if we can only let one into the championship game?

            How do your preferences pan out in the Big 12 last year, when TT beat TX who beat OK who beat TT, leaving 3 top-ranked, one-loss teams? Focusing on a single game isn’t always sufficient to say who’s best (OK got in by virtue of being the first team to lose). They’re all in the Big 12 South, so there’s on conference title matchup to square things.Report

            • Trumwill in reply to Zach says:

              On the field, though, Florida has not had the second best season. They lost a game by a significant margin. Texas, TCU, and Cincinnati have not lost a game. SEC fans will say “But they only lost because they’re in the SEC!” but I don’t really care. They have the honor of playing in the SEC to begin with. I’m sure the ACC would take Florida in a heartbeat.

              But… you’re right. I draw an exception to my “two best teams” rule when the two have already played one another. If one team has had their chance at proving that they’re better than the #1 team and failed to seal the deal, a team that has not gotten that opportunity ought to get a chance. That was how I felt a few years ago when some were clamoring for a Buckeyes-Wolverines rematch.

              As you’ve pointed out, there are a number of considerations. There are playoff schemes that I believe could balance them. I just don’t believe that they would be accepted by the larger fandom and I believe the result would be a 16-team playoff, which is a cure worse than the disease.Report

  4. Zach says:

    Oh, and it’d be cool if teams making the playoffs (or the round of 8 in my example) would be required to stage a rematch of their opening round game to kick off the subsequent season. An automatic way to promote strong nonconference schedules.Report

  5. kth says:

    Very well said. Playoffs are especially pernicious (and total boob bait as well) in baseball and basketball. In sports where teams play 100 matches (give or take), the championship should simply go to the team with the most wins (assume identical interleague schedules). Have an as-needed 3-game playoff if two teams have exactly the same record, but none otherwise. Obviously this will never happen: it’s less idle to speculate about a VAT tax or a US withdrawal of all forward deployments, things that are likelier than the NBA and MLB abandoning the cash cow of a post-season tournament.

    College football should actually have a playoff, because the schedules don’t overlap sufficiently for valid comparisons to hold. But the number of teams should be the minimum to ensure the inclusion of every team with a legitimate claim to be considered the best in the nation, based on their regular season record alone.

    I submit that eight is that number. (Obviously has to be a power of 2.) Four would probably suffice 9 years out of 10, but 8 would put the matter utterly out of dispute.Report

    • Zach in reply to kth says:

      The biggest problem with almost any playoff proposal will be the same as the BCS: the politics involved will ensure that the big 6 conferences get automatic bids. Between strong big 6 conferences and non-BCS conferences, more than 2 teams amongst them can too often credibly claim to be much better than teams with automatic bids.

      That’s why I think upping it to 10 teams from the 8 that’s normally talked about is a good idea. Forcing the bottom four to win an extra playoff game makes the teams with poorer regular season performance have to do more to prove themselves in the playoffs (similar to the role of seeding in the NCAA basketball tournament).

      Again, it depends on whether you think the title should go to the best team at the end of the season or the team that was consistently the best throughout the season. I think a playoff with a low fraction of teams advancing is a compromise on that score.Report

      • Trumwill in reply to Zach says:

        I think that the cracks in the “Big 6” are starting to show and the mystique is dissipating. Particularly with regards to the ACC, but also (unfairly) the Big East. I think that we’re reaching a point where it’s going to be The Big Four, The Next Three, and The Rest.

        The BCS bowl games this year will be particularly interesting, assuming that Boise State and TCU both get in. I would really love to see TCU and Cincinnati go at it and Georgia Tech and Boise State do the same. I really, really hope that they don’t stuff TCU and Boise State together.Report

        • Zach in reply to Trumwill says:

          “I really, really hope that they don’t stuff TCU and Boise State together.”

          Oh well. Sort of inevitable that Florida would be the first pick by the top ranked bowl. Reading through the arcane BCS rules, this looks like how the BCS selection went:

          1. Florida and TX to title game
          2. Oregon and OSU to Rose Bowl
          3. GT to Orange Bowl (host team)
          4. Sugar Bowl picks Florida (replacement pick for ‘Bama)
          5. Fiesta picks TCU (replacement for TX)
          6. Orange picks Iowa (why?)
          7. Fiesta picks Boise State
          8. Sugar picks Cincinnati

          No clue how this worked out. I guess Iowa’s good for more money than BSU and Cincy and BSU makes sense for the Fiesta bowl geographically (sort of). It’s too bad the non-championship matchups are decided so arbitrarily. They should at least go by BCS ranking instead of letting bowls choose who they want.Report

          • Trumwill in reply to Zach says:

            Huh. How about that. I really thought that TCU and Boise got put together because the PTB wanted to deny them of a “real” BCS victory. Or so they could point to the empty seats (hard to get excited about playing the team you played in a bowl last year) as proof of why they don’t invite non-BCS schools more often. Looks like it’s just sort of one of those things that just happened.

            My guess is that the Orange Bowl wanted someone other than Cincy because last year the Orange Bowl was a forgettable BE/ACC affair. I wonder if it were Boise State’s first time around if they might have chosen that for the publicity.

            Blah. Very disappointing.Report

            • Zach in reply to Trumwill says:

              Cincy has a stronger BCS computer ranking than Texas as well. I don’t exactly know how tight the bonds are between the different bowls’ selection committees, but I wouldn’t rule out collusion to advance their collective interests. The final draw was about as close to ideal as possible in terms of profitability. Ideally you’d have Florida in the Orange Bowl by that score. Iowa only makes sense there because there’s a big Orange Bowl parade and Iowa has a better marching band than the other options.Report

              • Trumwill in reply to Zach says:

                Good point about the Orange Bowl Parade.

                Regarding Cincinnati vs Texas, that’s the problem with polls. The BCS is in a no-win situation. After the USC-LSU-OU debacle, they were all but forced to emphasize the polls over the computers. I always viewed that as a mistake.Report

              • Zach in reply to Trumwill says:

                I think they should just hire Nate Silver to do a regression analysis that’s reweighted every season (so that you can’t game it in scheduling) to rank teams based on their win probability against the top 25 teams. More voodoo than the BCS score, but less arbitrary.

                I thought I was joking about the parade 🙁Report

  6. Dino says:

    So what if the “best” team does not win the championship?

    The Cinderella stories provide compelling drama. Look at the basketball tourney. Numerous upsets occur in the opening rounds, and once in a great while an Indiana State or a George Mason reaches the Final Four. Every other NCAA sport determines their champion on the field.

    The current system exists to perpetuate the cash flow to the established powers. The networks collude in the system because brand name teams draw more viewers. This is shortsighted. The championship pairing often results in boring routs. The other BCS bowl games are of little import. Personally, I will watch anything involving a ball but don’t usually watch the championship or many of the bowl games, unless an alma mater is participating.

    The talent in college football is dispersing. Players attend smaller schools in search of playing time. Every NFL roster contains players from unknown schools. New powers emerge and decline, e.g. Miami, Louisville. A more inclusive and meritocratic system is the American Way.

    TCU and Boise State may not be as good as Alabama but it should be settled on the field, not by a vote. The traditional powers will not play them. Boise State offered to play on the road next year without the promise of a return date. So far, no takers. Alabama’s non conference schedule consisted of Florida International, North Texas State, Tennessee Chattanooga, and the respectable Virginia Tech. In fairness to Bama, there is not an advantage to playing quality out of conference teams — other than the money from made for TV matchups. You can’t duck opponents in the NFL, hence that appearance of wild cards in the Superbowl.

    Thirty two teams, cut the regular season by a game or two. Institute a ranking program to discourage games against Cupcake State. Please don’t complain about the players missing classes.Report

    • Trumwill in reply to Dino says:

      Every other NCAA sport determines their champion on the field.

      Because to me, that’s the purpose of having champions. You could create all sorts of drama, if you wanted, simply by handicapping better teams.

      The purpose of a playoff system should be to identify the best team. Notably, few playoff systems limit themselves to that goal. Instead, they expand to a sort of second season (This time… relevant!). I don’t know of any way for college basketball to end except with a tournament of some sort. But college football already has a system in place. A system that, unlike the next division down, doesn’t crown 3-loss teams champions.Report

  7. Zach says:

    Texas getting into the title game by the skin of its teeth (kickoff penalty, horsecollar personal foul, luckily dodging clock management mistake) by winning its mediocre conference is probably as good a counterargument as any. Nothing in their record distinguishes them from TCU and Cincy.Report

  8. Trumwill says:

    Yeah, it’s a shame that we can’t pit those four against one another. Of course, then Boise State would point out that they never lost a game. Then Florida would point out that they’re better than Boise State. Penn State would point out that unlike Florida they at least won their conference…

    Seriously, it is a shame that we can’t have a four team tournament. But it’s nearly impossible to open that door without everything breaking loose.Report

  9. Publius says:

    Trumwill- I posted a challenge to you at http://www.thefourthbranch.com. Take up the debate!

    Thanks for your comments.Report

  10. Zach says:

    Last thing on this: an 8-team playoff with two non-big-6 at-large bids would work very well if the Big 10 picked up an additional team (Notre Dame ideally, but I guess Miami (OH) or something since that would never happen) and the Pac 10 grabbed two (Boise St. & Utah), and both conferences as well as the Big East added a title game.

    I guess a Pac10 title game isn’t strictly necessary, but it’s pretty silly to only have 3 nonconference games because you have to play a billion teams in your conference every year. I think that puts the Pac10 at a disadvantage in the human rankings.Report

  11. Trumwill says:

    Zach, the problem is that you need to come up with some sort of selection system that is at least ostensibly neutral. The NCAA can’t overtly play favorites. Taking the top 8 conference winners is one possible path and would include the BCS conferences almost every year except for the Oklahoma-KState problem, which would strongly disincentivize conference playoffs.

    Utah has had a couple breakout seasons, but BYU arguably has the better program overall. You have three teams for two slots in the Pac-10. You also need to get the Washington and Oregon states to agree not to play all the California schools every year. Since they do so much of their recruiting in California, that’s a tough sell. There’s also the Pac-10 snobbery of only wanting certain kinds of schools that don’t include the likes of Boise State and (less fairly) BYU.

    Regarding the Big 10, Notre Dame is and always will be the most likely candidate. But they could take Missouri (which has allegedly always wanted in to the conference) or Iowa State with the Big 12 taking Colorado State, Air Force, or Arkansas. If they take Arkansas, the SEC can take Louisville. The Big East can take East Carolina or Central Florida. And so on.

    One thing that I pondered when I was trying to figure out a way out of this mess (that did not include making a bigger mess) was allowing interconference matchups at the end of the season for 1-division conferences. Let the Pac-10 play the Big 10 and the Big East play the Mountain West conference while the other conferences are playing their championship game. It could strongly disincentivize conference championship games, though.

    It’s also worth pointing out that conference championships themselves are imperfect crowners of champions. Not just because inferior teams sometimes win, but because teams that have a strong case to play in the conference championship are not allowed. Last year, Texas and Texas Tech had the same record as Oklahoma but didn’t get to play for the conference championship.

    That doesn’t mean that I don’t favor conference championship games, just that it’s not a cure-all. I think that you have to just accept that divisional ties will exist and have good rules to determining who gets the nod in the event of one.

    I think a lot of the debate comes down to the belief among many that there is a system that will always fairly produce an indisputable champion. I don’t think such a system exists, but for college football in particular, I think that the current system is better than what we’re likely to get with a tournament. There are some tournament structures that I could see as being superior to the current arrangement (including my own and maybe yours, Zach), but everybody has a different idea of what constitutes “fair” and the only way to please everybody is to (in my opinion) significantly diminish the importance of the regular season and everything not directly related to the playoffs.Report

  12. Sam M says:

    One thing nobody has mentioned: What about the sheer number of games?

    If you play in a 12-team conference, that’s an 11 game schedule. Then, to win a 16-team playoff, you need to win four games. That’s 15 games.

    I know there are that many weeks in a season. But in terms of the health of the players (I know, like anybody cares…) that seems like an awful lot of games.

    But OK. We have fifteen games to determine a champion. I think that basically destroys the idea of any nonconference games in the regular seasons. That doesn’t matter for crowning a champion, but there are a lot of sweet interconference games that I like to watch.

    I guess you could split the 12 teams in the conference into two six-team divisions, of course…

    Still,. adding a sixt-team playoff adds a lot of games.Report

    • ed bowlinger in reply to Sam M says:

      This is right. Any playoff system large enough to have some semblance of fairness would add too many games to the college football schedule. The only way to compensate would be to take away non-conference games from the early season. Then you’re confronted with the problem of 104 other teams with only 9 or 10 games on their schedules.

      As far as I can see, any “solution” has to require that top teams play other top teams in the non-conference schedule. A quasi-relegation scheme which required that the top 40 teams play only amongst themselves in the non-conference schedule would go along way to making regular-season records reflect actual team performance.

      Preserve conferences, preserve rivalries, preserve the bowls, but *eliminate* the scheduling of out-of-conference mismatches.Report

    • Trumwill in reply to Sam M says:

      The lower divisions seem to have figured it out for a 16-team playoff. It does mean fewer OOC games, though, or fewer complete conference games. There’s also the matter of conference championship games, which lower divisions do not (AFAIK) have.

      Doing away with divisions doesn’t mean that they play every other team in their conference every year, though, as the Big Ten demonstrates. On the other hand, the 12-team conferences grouped together in part because of the championship games. Get rid of those and you might start seeing some significant realignment.

      Anyway, with a 16-team playoff, it’s workable. Once you’re getting to 32 teams, though, it becomes a little bit tougher. It’s hard to reduce the number of regular-season games. Some have suggested that the reason they added the 12th game was to make a playoff system less possible. Could be.Report

  13. CaptBackslap says:

    ed bowlinger: That’s pretty much what I’m thinking, too. The NCAA needs to have and use veto power over teams’ nonconference schedules. Someone needs to be able to tell football-factory coaches, “No, you can’t play this Alcorn State, Montana, and…OK, the School of Hard Knocks isn’t even a real school, it’s just an expression. Try again.”Report

    • Trumwill in reply to CaptBackslap says:

      If the NCAA simply banned games against I-AA schools, it would go a long way towards alleviating it. There are only so many Sun Belt and MAC teams to go around. Throw in an enforced minimum home-game requirement and it gets even more expensive.

      Of course, the press that complains about it so vociferously could put an end to it tomorrow. All they have to do is stop rewarding teams when they do it.Report

  14. carlos the dwarf says:

    Theoretically, the best team in the country is the team that can beat every other team in the country. Why not have a tournament at the end of the season for just the undefeated teams? Have a random draw (Champions League-style) to determine who plays who (and, with a 5-team pool, who has a round off), and keep going until every team but one has lost. Voila, your champion. The regular season still matters, the post-season is small and short, and there can be no doubt who deserves to be national champion.Report

    • I’ve thought of that. The problem it would present is that it would encourage even weaker scheduling than we have now. Though I don’t like multi-loss teams having a shot at the title, allowing for a loss or two does have its merits. Balancing these objectives is one of the real challenges to the post-season question for college football.Report

      • carlos the dwarf in reply to Trumwill says:

        Perhaps the answer then (or perhaps part of the answer in any case) is to have NCAA-mandated minimum strength of schedule requirements for FBS teams, based on the previous season’s win-loss record. For argument’s sake, what happens if we require every team to have a season schedule who’s combined win percentage the previous year was at least .450. Considering as the FBS-wide winning percentage would be .500, you’d have pretty much every team’s schedule in a narrow band between .450 and .550. Teams from better conferences would be able to get away with weaker non-conference opponents, teams from weaker conferences would have to play stronger non-conference opponents. With every team’s schedule at a similar level of difficulty, there’s almost no argument over strength of schedule. This really wouldn’t change that much for the teams from the best conferences, who could still get away with fairly weak out of conference schedules, but it would shut up the detractors of teams like Boise State and TCU, assuming they continued to be undefeated. It would give us a relatively equal basis on which to compare teams with the same record who haven’t played each other–or, at the very least, a more equal one than we have now.Report

  15. Sam M says:

    “Anyway, with a 16-team playoff, it’s workable.”

    Not so sure.

    Take a look at, saym the Big 10. Let’s say Michigan wins the conference by winning 9 games and a conference championship. Then it wins all four to sweep the playoffs. Thats 14 games. Which amounts to a ton of tickets sold and a ton of TV revenue, etc.

    But you are left with Ohio State, Penn State, Iowa, and all the rest stuck with just 9 games on their schedule. Ten for the team that lost to Michigan in the championship. That’s about three games short of what they normally play, which amounts to a ton of revenue for the schools in terms of tickets and TV. Not to mention the money they lost by not being in the crappy bowl game.

    Sure, D-II and D-III have figured this out. But schools like Carnegie Mellon don’t lose tens of millions of dollars by switching from a 12-game schedule to a 9-game schedule.

    So we have here a situation in which a playoff system means too many games for the eventual champ, and too few who are in it for the money.

    Given the money factor, I suspect that any playoff system would amount to at least 15-games, and probably 16, for the eventual champ.

    Over a four-year college career, this amounts to a full season of extra games if you were to win the championship all four years. It would amount to a full season of FEWER games if you were never to win it.Report

    • Trumwill in reply to Sam M says:

      The FCS manages to play 11-game seasons with a 16-team playoff. Some teams only play 10, but I think that’s just because of scheduling difficulties at that level (or maybe they can’t afford it). None of them have conference championship games except the SWAC, which chooses not to participate in the tournament.

      Would it be worth getting rid of conference championship games for a tournament? I don’t think it would and it would be a pretty bum deal for the conferences that expanded to twelve just so they could get that game in. However, most playoff proponents would disagree. Some of them, ironically, were against them in the first place.Report