clock menu more-arrow no yes

Filed under:

Stop saying that college football polls don't matter. My column:

New, 6 comments

Yes, I'm aware that we have a playoff selection committee. That doesn't turn the AP Poll into the point system from 'Whose Line Is It Anyway?'.

Tommy Gilligan-USA TODAY Sports

'My column' is a surprisingly not ironic college sports, well, column. Ideas, takes, complaints, Roll Tides? Share in the comments below, or email the author at matt AT sbnation dot com

The first weekend of college football is in the books, which gives way to the first revised bout of the AP and Coaches Polls, and subsequently, the first real bout of debate about them. Ohio State, despite covering the spread and beating Navy on the road, fell to No. 7 in the Coaches Poll, and No. 8 in the AP Poll, leading many Buckeye fans to loudly complain.

Perhaps equally loud was the contingent proclaiming that acknowledging these polls is foolish, because they no longer matter, a group that included many a media member. After all, unlike in previous years, the AP and Coaches Poll do not play a direct role in selecting the sport's national championship game participants. Being ranked No. 4 in the AP doesn't guarantee a playoff spot, after all, so who cares?

If you want to complain that the AP Poll is a lousy metric for assessing the quality of a program, you'll get no argument from me. If you want to tell me that the system creates a conflict of interest, I'll agree with you. If you want to tell me that getting upset over a particular team's placement in a poll in Week 2 is dumb, I'll probably concur (especially in the case of Ohio State). If you want to remind me that yes, the polls do not, in fact, explicitly place a team in the playoffs, I'll nod. All of that is correct.

But to say that the polls don't matter at all? That's wrong. Of course the polls matter.

For one thing, these polls have enough psychic value that schools are willing to pay coaches bonuses for hitting poll related benchmarks. Arizona State's Todd Graham gets a $23,000 bonus if the team finishes the year in the Top 25 of the Coaches Poll. If the Sun Devils crack the Top 10, it becomes $46,000. If Gus Malzahn at Auburn cracked the Top 5 of the Coaches Poll, he got a $100,000 bonus. Jimbo Fisher had a similar clause in his contract. George O'Leary at UCF gets a whopping $100,000 if the team finishes in the AP or Coaches top 25. There are a lot more examples.

Now, if you're getting a little uneasy about reporters who depend on relationships with coaching staffs having the ability to determine if they get bonuses of 50K or more, join the club. But remember, the only kinds of conflicts of interest that matter are with bloggers, who admit they're fans of a team.

These figures came from 2013, when yes, the polls were technically a part of the BCS, but the principle is still the same. Being ranked at No. 16, or No. 19, or No. 24, practically speaking, has nothing to do with the BCS, and nothing to do with the postseason chances of any of these teams. Arizona State could, theoretically, go to the same bowl ranked No. 14 or No. 21. The only rational reason that a school would pay somebody a bonus for hitting these benchmarks is if the program derived some sort of benefit, either in prestige, recruiting, or otherwise from the rankings. Fans aren't the only ones who care. Recruits care, coaches care, administrators care, and TV cares.

The polls may not explicitly determine who makes the playoffs, but that doesn't mean they can't implicitly. We won't get a poll from the playoff committee for more than a month, which means that one of the main frames of reference we (as fans, media members, anybody) have to determine who is "good" and who isn't comes from the rankings.

Who had the most successful week one? Many people who say either Georgia or Texas A&M, who picked up wins over quality teams. How do we know that beating Clemson or South Carolina is a big deal? Well, those teams were ranked highly. The narrative right now is that Florida State and Alabama struggled a little with lesser quality opponents in the first week. How do we know that West Virginia and Oklahoma State aren't as good? We don't know anything, but those squads weren't ranked. The entire preseason narrative around strength of schedules centers, in part, around where teams are ranked. As a sport, we aren't very good at reassessing those narratives several weeks down the line once the data changes.

There are 128 teams competing in FBS this season, and with so many of them playing at the same time, everybody has to take some intellectual shortcuts. We consult advanced statistics. We read breakdowns of games we didn't watch, and we try to extrapolate big picture results from our limited sample sizes. Beat writers can't watch every game, or even close to every game -- they're at the stadium, they're in the locker room, and they're on deadline, writing after the final snap. The athletic directors and committee members could potentially have even less. It's difficult to imagine a scenario where the rankings, and the role they play in coverage in major media outlets, doesn't have some sort of subtle, framing impact on where the committee assesses teams.

It's human nature to give some sort of attention to the previous establishment of credibility, even if it is just in the back of their minds.

Look, if the polls 100% truly didn't matter, we wouldn't write about them, and voters wouldn't vote in them. I don't see anybody voluntarily ditching their poll voting privileges. And as long as appetite for this level of coverage remains high, I, along with every other writer worth his salt, will write about them. A preseason Outland Trophy watch list? That doesn't matter. The Top 25 Poll is almost as deeply woven into college football's lore as a marching band at this point.

We can, and should, discuss the limitations of using polls. We can, and should, tell people not to freak out and claim some sort of global media conspiracy because we don't like the poll results after a single game. In a perfect world, maybe we all pledge allegiance to tempo-free advanced statistics.

But we don't live in that world, and these polls still matter. Let's not pretend otherwise.