Week 11 was probably the quietest one so far this season. There were no shocking on-field outcomes, especially among the select few teams still in the hunt for a Playoff berth. West Virginia continued to look strong in a blowout over TCU, Notre Dame predictably rolled over a collapsing FSU, Georgia took care of business against Auburn, Michigan stomped Rutgers, Clemson handily defeated Boston College, and Alabama continued to look invincible after blanking Mississippi State 24 – 0. Even Ohio State, who has struggled in recent weeks with issues on and off the field, posted a 20-point victory over a quality Michigan State team, in what I can only describe as the ugliest rock fight that I have watched in years. While that game was miserable to watch (especially for a neutral observer like me), it’s difficult to discount the value of a win over a quality divisional rival, even if Buckeyes fans did not come away from that game feeling fully reassured of their team’s future. The closest we came to a major upset was Oklahoma’s scare from Mike Gundy and the Cowboys. While the Sooners eventually pulled out a 48 – 47 victory, they were one play away from seeing their Playoff hopes completely dashed. After the Cowboys scored a touchdown on the final drive of the game, Mike Gundy elected to go for two instead of playing for overtime. Oklahoma State quarterback Taylor Cornelius, who up to that point had put on a prodigious performance, threw a very poor incomplete pass to an open receiver, sealing the win for the Sooners.
Due to the lack of surprising outcomes, I expected the results of this week’s MaxDiff Poll to be less interesting than usual, so I decided to change things up this week. Instead of viewing sets of four teams and choosing the best and worst team in each set, respondents instead viewed pairs of teams.
In each pair, they chose the team they felt would be most likely to win on a neutral field if the two teams were to play right now. I then used the usual Hierarchical Bayes estimation procedure to calculate the rankings. Notice that respondents weren’t choosing the team that was “most deserving” or had the best resume. They simply chose the team they thought would win head-to-head with home field advantage removed from the equation. In a sense, this is similar to how the College Football Playoff Committee ranks teams. The CFP Committee has said on many occasions that while they consider a variety of factors, the four Playoff spots ultimately go to who they think are the best four teams. That’s not exactly the most objective method, but it is the system we have to live with for now. Another benefit of having respondents evaluate pairs of teams is that we have data on who fans think would win in particular head-to-head matchups, which you can view in more detail below.
I usually field this poll on Monday mornings, with the results published on Tuesday mornings. Obviously, I’m running behind this week, but that gives me the ability to compare against the CFP Committee rankings, which are not published until Tuesday evening each week. For the remaining few weeks of this season, I plan to publish the MaxDiff Poll results on Wednesday mornings so that we can continue to compare against the CFP Committee rankings. What do you think about this plan? Let me know in this post’s comments on Reddit CFB.
Alright, let’s get to the results!
This Week’s Poll Specs
When was this fielded?: November 11, 2018
Number of respondents: 1,224
Research technique: Online quantitative survey using the MaxDiff Market Research Technique.
Interpreting the Scores: The MaxDiff scores for each team are indexed against the average score of 100. For example, Washington State’s MaxDiff score of 130 means that team was ranked 30% higher than the average team in the exercise. It is important to note that these scores should not be interpreted as “votes.” See my explanation of MaxDiff if you would like more information.
FanJuicer’s “Big Six” Takeaways From This Week’s Poll
Michigan enters Tier I.
Few teams have looked as good in recent weeks as the Michigan Wolverines. With an elite defense and an offense that has improved greatly since last season under the leadership of Shea Patterson, the Wolverines are solidly in the Playoff mix. One would think that if the Wolverines can overcome the Ohio State Buckeyes in a monster showdown in two weeks, that they would be guaranteed a Playoff spot. However, consider if Georgia were to beat Alabama. We’d be looking at Georgia, Alabama, Clemson and who else? Michigan or Notre Dame? A major source of discussion this season has been the comparison between Notre Dame and Michigan. While Notre Dame did defeat the Michigan in the first week of this season, it seems that the consensus among both fans and the media is that Michigan has looked like the more impressive team since then. Don’t believe me? Keep scrolling down this article and check out the head-to-head data.
Fans think the CFP Committee is really overrating the Florida Gators.
The Gators took a very disappointing loss a couple of weeks ago to Mizzou, and nudged past South Carolina last Saturday with a 35 – 31 win at home in The Swamp. These performances are likely contributing to fans’ dwindling perceptions of the Gators. Throughout this season, fans have been very high on the SEC teams, particularly the rank-and-file (i.e. every team other than Alabama and Georgia). In many MaxDiff Polls this season fans have voted the SEC rank-and-file among the most underrated in the AP. That trend has cooled substantially in recent weeks because. . .
Fans think the Committee is overrating Kentucky too.
In the first half of this season, the SEC East was surprisingly beginning to look like one of the top three divisions in college football. A major contributor to this was the unsuspected, meteoric rise of the Kentucky Wildcats. However, Kentucky has fallen back to earth in recent weeks, particularly this past Saturday. The Wildcats took a major hit as they were upset by Jeremy Pruitt and the Tennessee Vols. I suspect that the Committee may be putting more stock into the Wildcats entire body of work this season than fans, who tend to be somewhat more reactive.
Notre Dame is still seen as overrated by most fans.
Despite being undefeated with a victory over Michigan, Notre Dame has consistently been seen as overrated in recent weeks by most fans. The Committee may currently view the Fighting Irish as one of the top three teams in the nation, but the data suggests most college football fans do not.
Texas is now seen as underrated by fans.
The Longhorns bounced back from their heartbreaking loss to West Virginia by defeating Texas Tech in a road shootout. The Longhorns have another chance to impress this upcoming Saturday as they host Iowa State, a team held in high regards by both fans and the Committee.
Fans and the Committee agree on UCF.
Both groups place the Knights right outside the top ten at #11. This upcoming Saturday, the Knights face what will easily be their most challenging opponent on the biggest stage yet as they host the Cincinnati Bearcats. ESPN GameDay will be in attendance, giving the Knights perhaps the most exposure in the entire history of their football program for a regular season matchup. Fans expect UCF to win that game by more than a touchdown. If that comes to fruition, I’m not convinced it will put UCF in the Playoff conversation for either fans or the Committee. We’ll have to wait to find out.
The HEAD-TO-HEAD Data
How Do Fans Feel About This Week’s ESPN GameDay Showdown Between UCF and Cincinnati?
- Last week, fans were very close in their prediction of Clemson’s margin of victory over Boston College, but they overestimated how many total points would be on the board.
- Fans gave the LSU offense too much credit two weeks ago, but correctly predicted a Bama win.
- Fans’ predicted score was relatively close to the actual score of the Georgia – Florida game three weeks ago
- Fans were way off of the predicted score between Oregon and Washington State four weeks ago.
- Fans correctly predicted the winner of the game five weeks ago, but underestimated by how much Michigan would beat Wisconsin.
- Fans were way off with their prediction six weeks ago in the game between Texas and Oklahoma.
- Fans correctly predicted the winner seven weeks ago, but they overshot by how many Ohio State would beat Penn State.
- Eight weeks ago, fans were pretty close to dead on in their prediction of Stanford versus Oregon.
- Fans were also extremely close to the actual score nine weeks ago.
In case you are interested, here are the FanJuicer MaxDiff Poll rankings versus The AP.
FOLLOW FANJUICER ON TWITTER IF YOU WOULD LIKE TO RECEIVE UPDATES ABOUT WHEN THIS POLL IS AVAILABLE FOR YOU TO PARTICIPATE.