Thursday, November 29, 2007


Yesterday morning, I wrote about how pollsters are exhibiting a total lack of critical thought this year in blithely placing West Virginia, Missouri, and Ohio State in the top three spots in the polls simply because they have one loss and certain teams chasing them have two, even though several of the two-loss teams have played markedly tougher schedules. As if on cue, our old amigo Stewart Mandel arrived on the scene to illustrate what uncritical thinking looks like:

Not only have I been anti-playoff, but I've also been pro-BCS since it's inception (I can, in my mind, justify every title game participant). If Ohio State backdoors its way into the title game, then it's off to the playoff bandwagon I go. Please tell me you agree that putting Ohio State in the title game (and not just because of last year's performance versus Florida) is a bad, bad idea. A loss to Illinois isn't bad ... but beating absolutely no one of real significance is.
--Jared, Lawton, Okla.

First of all, that's very brave of you, Jared, to admit your fondness for the BCS in a public setting. I believe in certain states you can be fined and imprisoned for such a thing.

While I'm certainly not crazy about the idea of a team advancing to the national-title game while sitting on the couch for two weeks, I fail to see any grave injustice were the Buckeyes to make it. A month ago, when we were talking about then-undefeated Ohio State's credentials versus more accomplished one-loss teams like LSU and Oregon, I was right there with you. But I've said it before and I'll say it again: I don't care what conference you're in; if you can't make it through your schedule with less than two losses, you're just not national championship material.

Has Georgia played a tougher slate than the Buckeyes? Absolutely. USC? Yep. But by no rational measure are either more deserving than Ohio State. The Dawgs lost at home to 6-6 South Carolina and by three touchdowns at Tennessee. That's the equivalent of OSU losing to Michigan State and getting destroyed by Illinois (rather than losing by a touchdown). USC lost to Stanford. Can you imagine if the Buckeyes lost to Minnesota? I think we can all agree that in neither scenario would OSU even remotely enter the national-title discussion, yet because of the backlash that still lingers from last year's Florida game, there are no shortage of people who would rather see 10-2 Georgia or 10-2 USC in the title game instead of the 11-1 Buckeyes. That's just silly.

I hate to break it to the masses, but guess what? Ohio State? That's not a bad football team. I don't necessarily think it's the best team in the country, but it's up there. Last we saw them, the Buckeyes went to Ann Arbor and held Michigan to 91 total yards. No, it was not a great Michigan team, but holding any team to less than 100 yards of offense is pretty darn impressive. LSU did not do that this season. Neither did Oklahoma, nor USC, nor ...

I think OSU is in the exact right spot in the pecking order as of now -- last of the remaining one-loss teams, but ahead of any two-loss teams.

I'll agree that there would be no "grave injustice" if Ohio State made the title game in place of a series of two-loss teams. However, the question is whether a good case can be made for those two-loss teams ahead of the Bucks. The pollsters seem to think that the answer is no, as they uniformly put the one-loss teams ahead of the two-loss teams without a care in the world as to the schedules played by those teams.

Mandel claims that "no rational measure" could put Georgia or USC ahead of Ohio State. None of us should be lectured about rational thought by Mandel, especially since the centerpiece of his argument is completely irrational. He claims that he was willing to consider ranking one-loss teams with better schedules ahead of an unbeaten Ohio State team that hadn't played anyone, but he's not willing to rank two-loss teams with better schedules ahead of a one-loss Ohio State team. In what universe does this make sense? The analysis is exactly the same: can a one-game disparity in record be overcome by a significant different in schedule strength and if so, how big does the schedule gap have to be? But don't take my word for it that a two-loss team can be ranked ahead of a one-loss team on the basis of having faced a more difficult schedule; Mandel himself employs this reasoning by ranking one-loss Kansas behind two-loss Georgia, LSU, USC, and Oklahoma.

Leaving aside the irrationality of a hard and fast "two-loss teams cannot play for the national title," Mandel's arguments regarding those two-loss teams are weak as hell. I'm going to leave USC aside because their strength of schedule isn't much better than that of Ohio State. Georgia's SOS is much better than that of the Bucks, but Mandel thinks that Georgia's losses were worse. That's probably true, although deriding a loss to a 6-6 South Carolina team is very misleading, as the South Carolina team that Georgia played in September had Captain Munnerlyn and Jasper Brinkley in it. This was the South Carolina team that started the year 6-1 and beat Kentucky handily, rather than the South Carolina team that collapsed after blowing the Tennessee game in ludicrously unlucky fashion. On the other hand, Ohio State doesn't have a win that comes close to a 12-point win over top ten Florida at a neutral site or a 25-point win over a top 20 Auburn. Should we focus on the quality of wins or the quality of losses? Seems like an interesting question to me, but Mandel and the voters are unwilling to consider it.

Comparing Georgia and Ohio State is instructive because it illustrates the importance of scheduling. Georgia and Ohio State both came into the season with inexperienced players all over their rosters. Georgia's young players were thrown into the fire with games against Oklahoma State, South Carolina, Alabama, and Tennessee in the first half of the season. Ohio State didn't play a bowl team until its sixth game of the year. It didn't play a team that would finish better than 7-5 until the 9th game of the year. They played one team that finished with fewer than four losses all year and lost that game. At home. Could that breeze of a schedule in the first half of the year possibly explain a one-game gap in record?

Even if Mandel is right that Ohio State deserves to be ahead of Georgia because Georgia's losses are bad, he's omitting another two-loss team that played a markedly more difficult schedule than Ohio State and also happens to have better losses: LSU. The Tigers' two losses were both to quality teams in three overtimes. LSU's wins over Florida, Auburn, and Virginia Tech are better than anything on Ohio State's resume. If LSU beats Tennessee this weekend, then the Tigers will have an additional quality win and their strength of schedule will go even farther past that of Ohio State. Is there "no rational measure" that would put the Tigers in the title game?

Again, for the record, I have no problem with Ohio State playing for the national title. They have an excellent defense and their schedule, while weak, isn't Hawaii weak. I do have a problem with the fact that pollsters don't seem to be critically evaluating their resume against that of LSU and Georgia. College football's silly post-season structure requires hair-splitting between teams, but if Mandel is any guide, the media are either refusing to split hairs or they are doing a crappy job in that task.

1 comment:

Anonymous said...

I think many in sportsmedia lack the ability to split hairs so finely. I've said it before and I'll say it again, but most of these guys didn't choose sportswriting over med school.