Monday, September 27, 2021

College Football Playoff Predictor - Week 4

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

Rank Team Agg Rank CFP Prob Change
1  Georgia 1 76.57% 12.5%
2  Alabama 2 67.07% 4.0%
3  Oklahoma 4 43.47% -15.2%
4  Michigan 7 34.97% 5.0%
5  Notre Dame 13 30.31% 22.4%
6  Oregon 12 26.54% -5.0%
7  Penn State 9 22.33% -5.2%
8  Ohio State 3 17.77% 1.0%
9  Ole Miss 8 15.24% 2.7%
10  Florida 6 10.62% 2.0%
11  Iowa 17 8.99% -5.9%
12  Michigan State 24 6.29% 3.2%
13  Arkansas 16 5.24% 4.3%
14  Clemson 5 4.74% -15.4%
15  Wake Forest 37 3.23% 2.9%
16  Baylor 26 3.12% 2.2%
17  UCLA 19 3.04% 2.3%
18  North Carolina State 23 3.02% 2.3%
19  Texas 10 2.89% 1.7%
20  Pitt 18 2.11% 1.8%
21  Maryland 32 2.10% 1.7%
22  Boston College 45 1.73% 1.1%
23  Oklahoma State 36 1.54% 1.2%
24  Texas A&M 15 1.29% -12.8%
25  Kentucky 48 1.27% 1.0%
26  Cincinnati 14 0.61% 0.1%
27  Arizona State 27 0.54% 0.4%
28  Oregon State 47 0.40% 0.4%
29  Rutgers 42 0.35% 0.2%
30  Virginia Tech 41 0.32% -1.3%
31  Coastal Carolina 28 0.26% 0.1%
32  Iowa State 11 0.22% -1.2%
33  North Carolina 21 0.20% -6.7%
34  LSU 25 0.20% 0.1%
35  Wisconsin 22 0.17% -3.4%
36  Auburn 20 0.16% -0.7%
37  SMU 31 0.15% 0.1%
38  Louisville 61 0.12% 0.1%
39  UTSA 67 0.12% 0.1%
40  Miami (FL) 30 0.10% 0.1%
41  BYU 54 0.08% 0.0%
42  Kansas State 51 0.07% -0.5%
43  Washington 35 0.07% 0.0%
44  TCU 40 0.06% -0.7%
45  West Virginia 29 0.04% 0.0%
46  Utah 39 0.04% 0.0%
47  Texas Tech 57 0.04% -0.3%
48  Purdue 52 0.04% 0.0%
49  USC 38 0.04% -1.8%
50  Army 66 0.03% 0.0%
51  San Diego State 69 0.03% 0.0%
52  Virginia 55 0.01% -0.4%
53  Syracuse 74 0.01% 0.0%
54  Mississippi State 33 0.01% -0.1%
55  Wyoming 83 0.01% 0.0%
56  Georgia Tech 50 0.00% 0.0%
57  Tennessee 49 0.00% 0.0%
58  South Carolina 75 0.00% 0.0%
59  Stanford 68 0.00% 0.0%
60  Duke 78 0.00% 0.0%
61  Minnesota 62 0.00% -0.2%
62  Missouri 60 0.00% 0.0%
63  Indiana 58 0.00% 0.0%
64  Northwestern 76 0.00% 0.0%
65  South Alabama 107 0.00% 0.0%

One month in, the quintet of teams atop the original rankings have gone in very different directions.  Bama and Georgia have looked every bit the part of title contenders and have separated from the pack accordingly (#2 Bama is nearly 6 points clear of #3 OSU in the computer ratings).  Oklahoma is still unbeaten, but looks significantly more vulnerable than originally thought.  Conversely, while Ohio State has taken a loss, they might have the pieces to put together a run.  And then there's Clemson, which I will damn with the faintest of praise by noting that they are the most likely team to make the Playoff from the ACC.

Which thing would you have thought more likely at the start of the season: Michigan being in the top four of the Playoff odds by the end of September, or Pitt simultaneously losing to Western Michigan and earning the second-highest computer rating in the ACC?  Trick question it is both.

Conference Favorite Perc   Runner-Up Perc
ACCA Clemson 42.5%   North Carolina State 36.5%
ACCC Pitt 38.4%   Virginia Tech 22.5%
AMER Cincinnati 79.1%   UCF 38.7%
B10E Ohio State 46.1%   Michigan 28.7%
B10W Iowa 59.6%   Wisconsin 23.0%
B12 Oklahoma 80.2%   Texas 47.0%
CUSAE Marshall 47.8%   Western Kentucky 25.2%
CUSAW UAB 46.2%   UTSA 42.2%
MACE Buffalo 49.8%   Kent State 26.1%
MACW Toledo 48.7%   Western Michigan 34.2%
MWCW Fresno State 51.8%   San Diego State 28.7%
MWCM Boise State 72.1%   Wyoming 11.6%
P12N Oregon 58.6%   Washington 23.9%
P12S UCLA 43.3%   Arizona State 27.8%
SECE Georgia 81.1%   Florida 15.9%
SECW Alabama 79.1%   Ole Miss 11.2%
SUNE Coastal Carolina 56.7%   Appalachian State 40.7%
SUNW Louisiana 86.1%   Arkansas State 5.6%

Clemson is still a slight favorite in the ACC, but these are the closest margins we've seen since Florida State was good at football.  Ohio State drops below 50% in what might be the best division in the sport?  UAB-UTSA is starting to look like one of the best under-the-radar races.  And Bama is still the king until otherwise noted...sorry.


Test Simulation of the Week


Week 2 of my new feature is a reminder that reality (or potential reality, at least) is sometimes boring.  Yes, a debate between 11-1 Notre Dame and 11-2 Georgia (and maybe 10-2 Florida?) would be mildly interesting but keep in mind that the prize would be playing Alabama in a semi-final.  Which means the committee should put UTSA in.


Week 5 Preview

Home Away Home Win Prob Playoff Teams Lost
Alabama Ole Miss 82.1% 0.082
Georgia Arkansas 87.6% 0.047
Notre Dame Cincinnati 58.4% 0.046
Wisconsin Michigan 35.5% 0.043
Kansas State Oklahoma 20.0% 0.029
Stanford Oregon 23.6% 0.021
Rutgers Ohio State 21.4% 0.020
Maryland Iowa 44.8% 0.017
Kentucky Florida 22.7% 0.015
Clemson Boston College 85.3% 0.012

There are three ways to look at this weekend, each of which confirm its place as one of the best weekends of the year.  The first is through the Bama-Georgia prism.  These are clearly the two best teams at this point, which means that this season will be defined in part by how they perform the rest of the way.  If one or both is going to lose before Atlanta, this weekend is one of the most likely culprits.  Bama-Ole Miss gets the edge as game of the week because the Rebels appear to have a slightly higher ceiling than Arkansas.

View #2 is that this is a ridiculously good road test weekend for the next level of contenders.  Ohio State, Michigan, Cincinnati, Oregon, Oklahoma, Iowa, and Florida all face a 20% chance or better of losing to an inferior opponent (sorry ND but you know it's true).  The depth and breadth of threats present to this group of teams is unlike anything I can remember in recent history.

Finally, there's just a lot of games, man.  Important games!  The list above doesn't even include a game between 4-0 teams in Oklahoma State-Baylor, a battle for the Pac-12 South in Arizona State-UCLA, a suddenly important (?) Georgia Tech-Pitt matchup, and Oregon State hosting Washington in their bid to steal the North.  Oh and a decent Auburn-LSU game that's somehow only the 20th most-important game of the weekend.  Soak it in folks, October is here.