Sunday, November 22, 2015

Weekly Playoff Probabilities - Week 12

As always, explanation here, rankings below, thoughts after.

Rank Team FPI Rank CHAMP Prob POFF Prob Change
1 Clemson 6 20.62% 77.03% 7.2%
2 Iowa 28 4.33% 53.55% -1.3%
3 Oklahoma 1 18.53% 41.73% 11.0%
4 Alabama 3 13.33% 36.02% 4.6%
5 Baylor 2 12.89% 32.37% 12.1%
6 Michigan State 17 4.55% 31.60% 22.6%
7 Notre Dame 8 6.64% 29.75% 1.5%
8 Oklahoma State 14 4.30% 24.78% -15.0%
9 Ohio State 4 7.37% 23.84% -30.1%
10 North Carolina 15 3.01% 18.60% 7.1%
11 Florida 19 1.80% 12.86% -7.1%
12 Northwestern 55 0.11% 4.79% 4.0%
13 Texas Christian 5 1.16% 4.08% -6.6%
14 Florida State 13 0.72% 4.01% 0.9%
15 Stanford 9 0.61% 2.86% 1.0%
16 Michigan 16 0.35% 2.15% 0.9%

1. Even at this late date, there's still a good deal of carnage at the bottom of these rankings.  We say goodbye to Utah, LSU, Houston, and Wisconsin.  Because of two awful calls, the Badgers really shouldn't be in this group, but they probably should have lost to Nebraska last month, so it all evens out.  Northwestern remains easily the worst remaining "contender."  This isn't just an FPI oddity either, as Sagarin and S&P+ both have the Wildcats in the 40s.  One might find reason begrudge Northwestern for their success, but after a couple of hard-luck seasons for Pat Fitzgerald they've earned some good luck.  Just don't use the L-word around their fans.

2. The below chart appears to indicate that the final spot will come down to the SEC champ or Notre Dame, but there's a couple of issues with that.  One, I know my model is underestimating Bama's chances, so the SEC is probably closer to equal footing with the other three conferences than it appears.  Two, the Big Ten still has a lot of chances to knock itself around over the next two weeks.  11-1 Iowa vs. 10-2 Michigan for the conference crown has a reasonable chance of happening, and slightly less crazy things have an even better chance of coming true.

Conference Exp Playoff Teams
Big 10 1.159
Big 12 1.030
ACC 0.996
SEC 0.489
Independent 0.297
Pac 12 0.029

3. TCU lost a game and fell less than a team that won (Florida).  I'm not saying I disagree with that happening, as it's no crime to lose by one on the road against what might be the best team in the country.  It's just interesting that my system seems to recognize that without explicitly knowing that that happened. 

4. Our New Champ Probability went up a touch this past week, to 25.2%.  Baylor makes up over half of this percentage, but CFP standings may dictate that Iowa will represent the best chance for this to actually happen.  Iowa vs. Baylor is actually one of the most interesting possible debates for the final playoff spot.  Both teams are similar in that they're schedules leave something to be desired.  Baylor has a bunch of blowout wins as their argument, while Iowa has the lack of losses to lean on.  If both can win their conferences, this could become an interesting case study for better understanding the committee's way of thinking.


Nebraska Bowl Watch

Nebraska did not play this weekend, and Iowa beat Purdue by the expected number of points, so this doesn't change a whole lot.  Nebraska sits at a 45.9% chance of making a bowl.  There has been some chatter about scenarios where not enough teams make it to 6 wins, which would make a 5-7 Nebraska an attractive option*, but let's not talk about that for now.

*I did it, I used "attractive" and "5-7" in the same sentence.


Week 13 Preview

Home Away Home Win Prob Playoff Teams Lost
Oklahoma State Oklahoma 33.07% 0.152
TCU Baylor 47.94% 0.099
Stanford Notre Dame 56.88% 0.097
Nebraska Iowa 45.85% 0.082
Michigan Ohio State 42.61% 0.063
Florida Florida State 52.73% 0.051
Auburn Alabama 22.38% 0.040
South Carolina Clemson 14.25% 0.036
NC State North Carolina 36.88% 0.034
Michigan State Penn State 79.09% 0.033
Illinois Northwestern 42.93% 0.020

This is an amazing week, no matter which angle you take.  Bedlam is not just the most impactful game of the week, but of the year.  The next five games are all virtual coin flips and will all affect one or more serious playoff contenders.  The next two games see the top two in the actual CFP rankings with stronger than you might think odds of losing (although FPI is a little down on both of those teams compared with the average computer ranking, so its exaggerating the upset odds in all likelihood).  And there are a number of amazing games between non-contenders that will either shape conference races (USC-UCLA, WKU-Marshall, Houston-Navy) or just be a whole lot of fun (Ole Miss-MSU).  Don't interact with your friends or family this weekend, just watch football.  Or watch football with them, I guess...you decide.

No comments:

Post a Comment