Tip Top 25 in helmets, smaller

Comments on the AP College Football Poll: December 4, 2017

1) Clemson 12-1
2) Oklahoma 12-1
3) Georgia 12-1
4) Alabama 11-1
5) Ohio State 11-2
6) Wisconsin 12-1
7) Auburn 10-3
8) Southern Cal 11-2
9) Penn State 10-2
10) Central Florida 12-0
11) Miami (Florida) 10-2
12) Washington 10-2
13) Texas Christian 10-3
14) Notre Dame 9-3
15) Stanford 9-4
16) Louisiana State 9-3
17) Oklahoma State 9-3
18) Michigan State 9-3
19) Memphis 10-2
20) Northwestern 9-3
21) Washington State 9-3
22) Virginia Tech 9-3
23) South Florida 9-2
24) Mississippi State 8-4
25) Boise State 10-3

Others Receiving Votes
North Carolina State 8-4
San Diego State 10-2
Toledo 11-2
Louisville 8-4
Michigan 8-4
Florida Atlantic 10-3
Fresno State 9-4
Missouri 8-4
Troy 10-2
Kansas State 7-5
If you've been wondering why I haven't been fixing the AP college football poll every week, it's because I don't bother to fix the AP poll's top 25 until after the final edition is released in January. When there are no more games to be played, there is no longer any wiggle room for ignoring head-to-head results and season-long performance. But before that, there can be plenty of room for wiggling.

For example, I think that 10-3 Auburn (#7) should be rated ahead of 11-1 Alabama (#4). Details on this issue lie below. But if Alabama beats Clemson in the playoff, then the AP poll will have been proven right. On the other hand, if they lose, and if Auburn wins their bowl game, Auburn will move past Alabama in the final AP poll anyway. So I'll just wait to see how it all plays out.

Obviously, while I will not fix the AP poll until January, it is late enough in the season that I now feel comfortable commenting on the weekly poll. I will go ahead and point out problems with the poll that would be outright errors if the season had ended this last weekend, and no more games were to be played. This can also give voters (in the AP or in any other poll) an idea of potential errors to avoid going forward.

Playoff Committee Rankings

I will also add some comments on the playoff committee's top 25 rankings to the bottom of this article, and I have a comment to make about this year's top 4 playoff field as well, just below. As we have seen most years, the differences between the playoff committee rankings and the AP poll's top 25 have been minor this year. Unfortunately, the committee largely suffers from the same logic problems the AP poll does-- they rely far too much on teams' straight records when ranking them, and they place far too much emphasis on the last game played for each team.

Because the playoff committee will not be ranking teams in a final post-bowl top 25, I will not be fully fixing any of their rankings like I will the final AP poll top 25 in January.

Now let's get to commenting on the problems with the latest AP poll...

Auburn and Alabama

As I noted above, 10-3 Auburn (#7) should be rated ahead of 11-1 Alabama (#4). I know that it is impossible for your average sportswriter (or playoff committee moron) to get their heads around a 3-loss team being more deserving than a 1-loss team from the same conference, but this is an example of it. Yes, Auburn has 3 losses, but 2 of those losses came to #1 Clemson and #3 Georgia, both rated higher than Alabama. Those losses are irrelevant when comparing Auburn to Alabama, since Alabama did not themselves defeat anyone nearly as strong as those teams, and Alabama lost by 2 touchdowns in their one game against a comparable team, that being Auburn.

So it's as simple as this: Auburn took an upset loss to LSU, whom Alabama defeated, but Auburn beat Alabama 26-14 head-to-head. The teams therefore have the same relevant record, but Auburn wins the head-to-head tie-breaker and should therefore be rated higher. Auburn is being punished for playing a vastly tougher schedule than Alabama did, and Alabama is being rewarded for playing by far the most ridiculous schedule ever played by a playoff team. Speaking of which...

Alabama in the Playoff is a Travesty

The college football playoff is broken. It's 2011 all over again. The whole point of creating this playoff, and using a playoff committee instead of a formula, and making conference titles a written criterion for selection, was to avoid a repeat of 2011. With 4 spots and 5 major conferences, it was a given that one conference would be shut out every year, along with all of the teams that aren't in a major conference. What was not a given is that 2 conferences would be shut out, all so one conference could have half the playoff spots. This is supposed to be a national playoff, but it isn't. It's a regional playoff, and this year more than half the teams in the FBS are completely shut out of it.

And for what? Alabama lost by 2 touchdowns to the only top 10 team they played this year, and it was their last game. Their best win came over 9-3 LSU (#16), who lost to Troy. Their next best win came over 8-4 Mississippi State (#24), who was trounced by Georgia and Auburn, and lost to 6-6 Ole Miss. And both games were close! So Alabama gets in despite accomplishing pretty much nothing all year? Because the committee just thinks they look better?

Ohio State took 2 losses because they played a far tougher schedule. They defeated 2 top 10 teams, giving one of them their only loss, and they defeated more rated teams overall. They defeated 3 teams that are ranked higher, according to the committee's own rankings, than anyone Alabama defeated. Oh, and they won their conference, which was supposed to matter, but apparently doesn't make any difference at all.

But the real point is that the playoff is made less national by this decision, less representative, and more parts of the country are shut out of it. The only way a 2nd team should come from the same conference is when there is absolutely no doubt. Meaning an Alabama that beat Ohio State during the regular season. Or even just one top 10 team.

Furthermore, this decision rewards weak scheduling, which is terrible for college football. If USC had not played Notre Dame, and Ohio State had not played Oklahoma, either would be in the playoff over Alabama right now. But Alabama is rewarded for playing the weakest schedule among contenders.

And it's not just Ohio State and Southern Cal. Central Florida won more games over rated opponents than Alabama did, and they took zero losses! And finally, as noted above, if a 2nd SEC team was going to get into the playoff, that team should have been Auburn, not Alabama. But again, this shutting out of more than half the country so that one conference could field half the playoff-- despite doing nothing to earn it at all-- is the biggest problem here.

Miami (Florida), Notre Dame, and Southern Cal

The AP poll punished 10-2 Miami (Florida) hard for losing to #1 Clemson, dropping them from 4 places ahead of 11-2 Southern Cal to 3 places behind USC. However, Miami has 1 upset loss, while USC has 2 upset losses, so Miami has the better relevant record. Furthermore, Miami beat Notre Dame 41-8, and Notre Dame beat Southern Cal 49-14. So it goes without saying that these 3 teams should be rated like so: Miami > Notre Dame > Southern Cal.

Now, it is true that Miami took an upset loss (to Pitt), and Notre Dame did as well (to Stanford), but so did USC (to Washington State), so there is no advantage to USC there. USC has 2 wins over rated teams, but so do Miami and Notre Dame. There is no good reason to ignore the head-to-head chain here.

Stanford and Washington

I covered this issue last week. 9-4 Stanford (#15) should be rated ahead of 10-2 Washington (#12). Washington is 10-2 because they played a ridiculously easy schedule. Stanford has 2 more losses because they played #8 Southern Cal twice, and Washington did not play them at all.

Stanford beat Washington 30-22, and Washington took an upset loss to unrated Arizona State as well. Stanford took 2 upset losses, so the teams have the same relevant record, but Stanford owns the head-to-head edge and should therefore be rated higher. For the same reason, Stanford won the division and Washington did not.

In addition, Notre Dame should be rated ahead of both teams, as covered above, giving Stanford an upset win, and that makes them a slam-dunk to be rated ahead of Washington.

Mississippi State and Louisiana State

Also covered last week, so I'll just repeat what I said then: It is just lazy to simply order teams by their straight records, and AP poll voters need to get over this habit. Louisiana State is rated #16 and Mississippi State #24 because LSU is 9-3 and MSU is 8-4. But MSU is 8-4 because they played 12-1 Georgia from the other division, while LSU played 4-7 Florida. The AP poll is really just punishing MSU for playing a tougher schedule.

Here is the only fact that matters: Mississippi State destroyed LSU 37-7. Therefore they should be rated higher. Now, LSU did beat Auburn (who beat Mississippi State), and MSU did take an upset loss to Ole Miss (whom LSU defeated), and that would put LSU ahead of MSU by one game. But the problem is that LSU also took an upset loss, to unrated Troy. That makes these teams even, with MSU holding the head-to-head tiebreaker. And it was a very decisive outcome.

Now, if MSU takes an upset loss to unrated Louisville in their bowl game, or if LSU beats Notre Dame in theirs, this issue will work itself out. Otherwise, it will need fixing when the final poll comes out in January.

Michigan State

Here is an issue I overlooked last week. 9-3 Michigan State is underrated, and should be above Oklahoma State and Louisiana State-- and probably Stanford, Washington, and Texas Christian as well. Michigan State took one upset loss, in overtime at #20 Northwestern, but they made up for it with a huge win over #9 Penn State. Compare that to these teams rated above them:

9-3 Oklahoma State: Upset loss to unrated Kansas State at home, beat no rated or even near-rated team.
9-3 Louisiana State: Two upset losses, 37-7 at #24 Mississippi State and to unrated Troy at home, made up for only one of those with a big upset win over #7 Auburn.
10-3 Texas Christian: Upset loss at unranked Iowa State, best win over #17 Oklahoma State (Michigan State lost to a better team and beat a better team).
9-4 Stanford: Two upset losses.
10-2 Washington: Two upset losses.

Note that even if Washington is dropped behind Stanford, where they belong, they would still have an upset loss, to unranked 7-5 Arizona State, and their best win is over #21 Washington State, so Michigan State is still clearly better in both regards.

South Florida

As covered last week: 9-2 South Florida is #23, but they should not be ranked at all. Sure, they played a great game in their finale, falling to #10 Central Florida by only a touchdown. But that is by far their most impressive result of the season: a loss. Guess how many teams with a winning record USF has beaten. I'll give you a hint: it's the same number Pee Wee Herman has beaten this season. Zero. They are 0-2 against winning teams, one of whom is not even close to being rated.

USF has accomplished absolutely nothing this season. They lost to 7-4 Houston at home, and frankly they are lucky to be rated higher than Houston, who has beaten 4 winning teams. Furthermore, even if South Florida wins their bowl game over 6-6 Texas Tech, they still will have proven nothing, and will still have no win at all over a winning team.

North Carolina State and Louisville

As covered last week: North Carolina State and Louisville are both 8-4, and both are sitting just outside the top 25 (#26 and #29). Another thing they have in common is that both are overrated. These teams are poised to finish ranked if they win their bowl games, but they should not be ranked even in that case.

8-4 NC State lost to 8-4 South Carolina, and I don't see why they should be rated higher than the Gamecocks. They also lost to Wake Forest, so that makes 2 losses to unrated teams. Beating unrated 7-5 Arizona State in their bowl game will do nothing at all to make up for that fact. Needless to say, if 8-4 Michigan (#30) beats South Carolina in their bowl game, they should be ranked ahead of NC State. But since all 4 of Michigan's losses have come to teams in the top 18, they are already effectively 2 games better than NC State on the season, and should already be ranked higher.

8-4 Louisville lost to NC State, so obviously they belong rated behind the Wolfpack. Like NC State, Louisville also lost to Wake Forest, and they took a 3rd loss to an unrated opponent against Boston College. Louisville is not even close to being a legitimate top 25 team. If they beat #24 Mississippi State in their bowl game, that should knock MSU out of the ratings, but it should do nothing to help Louisville's cause here.

Fresno State and San Diego State

9-4 Fresno State (#32) won 27-3 at 10-2 San Diego State (#27), and they should therefore be rated higher. They were rated higher last week, but dropped a lot after losing the conference championship game 17-14 at #25 Boise State. Boise State beat San Diego State 31-14, so it doesn't make any sense at all to drop Fresno State behind San Diego State based on their conference championship game result.


11-2 Toledo now sits at #28, also poised to get into the final top 25 with a bowl win. They would perhaps be a better choice than #23 South Florida, since they have actually beaten some winning teams, but they haven't beaten anyone who is close to rated, they have taken an upset loss 38-10 to an unrated team, and therefore Toledo should not finish ranked. Their bowl game, against 8-4 Appalachian State, is completely irrelevant to this fact.

Iowa and Michigan

I covered the Iowa part of this last week: I'm not saying that 7-5 Iowa should be ranked in the top 25, but they would be a legitimate choice, and a far better choice than NC State and Louisville. Iowa took one upset loss, to unrated 6-6 Purdue, but they also routed #5 Ohio State 55-24 to make up for it. Their other 4 losses all came to teams that are ranked in the top 20. Iowa also defeated 7-5 Iowa State, who beat #2 Oklahoma and #13 Texas Christian. However, if Iowa loses their bowl game to unrated 7-5 Boston College, the AP poll will be proven right in not ranking them.

Michigan hasn't beaten anyone of value, but they have taken no upset losses, which puts them effectively 2 games better than NC State and Louisville. If they beat unranked 8-4 South Carolina, who beat NC State, they should be rated ahead of both. Of course, if South Carolina wins, they should be rated ahead of NC State, so there's no winning for NC State here.

Comments on the Playoff Committee's Top 25

Prior to this year, the playoff committee's rankings had gotten worse, relative to the AP poll, each year, but this year they are really about the same. If I were to give an edge to one over the other this week, I would give that edge to the playoff committee's top 25. I see 7 differences between the playoff committee's top 25 and the AP poll top 25, and in my opinion, the playoff committee made the better choice 4 times to 3.

The Playoff Committee's 4 Better Choices

The playoff committee was correct, in my opinion, with these choices that differed from the AP poll:
  • Michigan State rated higher than Louisiana State and Oklahoma State (as I covered in detail above)
  • Washington State rated higher than Oklahoma State, Memphis, and Northwestern (not necessary, but a better choice)
  • Texas Christian rated behind Stanford and Notre Dame
  • No South Florida (as covered above)

The AP Poll's 3 Better Choices

The AP poll was correct, in my opinion, with these choices:
  • Central Florida rated ahead of Miami (Florida) and Washington
  • Notre Dame rated higher than Stanford (should be rated ahead of USC as well, as covered in detail above)
  • Boise State rated higher than North Carolina State