Sunday, September 23, 2018

Weekly Playoff Probabilities - Week 4

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

Rank Team Agg Rank POFF Prob Change
1 Alabama 1 80.63% 1.4%
2 Clemson 4 72.33% 12.0%
3 Georgia 3 42.82% 6.8%
4 Penn State 5 36.48% 3.1%
5 Ohio State 2 34.27% -8.3%
6 Oklahoma 7 27.13% -7.8%
7 Notre Dame 11 25.53% 10.4%
8 Washington 8 18.45% -3.0%
9 Stanford 14 18.06% 8.5%
10 West Virginia 16 5.69% 3.4%
11 LSU 18 5.64% 1.1%
12 Michigan 6 5.51% 3.5%
13 North Carolina State 29 5.34% 3.6%
14 Duke 25 3.76% 0.6%
15 Miami (FL) 17 2.92% 0.3%
16 Kentucky 30 2.45% 2.3%
17 Oklahoma State 19 1.69% -16.1%
18 Auburn 9 1.55% 0.3%
19 South Carolina 23 1.49% 1.2%
20 Mississippi State 12 1.00% -14.4%
21 Texas Tech 24 0.98% 1.0%
22 Texas A&M 15 0.69% -0.1%
23 Michigan State 13 0.68% 0.2%
24 Syracuse 43 0.66% 0.3%
25 Wisconsin 10 0.60% 0.4%
26 Iowa 26 0.59% -1.3%
27 Florida 21 0.57% 0.5%
28 Missouri 28 0.40% 0.0%
29 Texas 22 0.39% 0.3%
30 Washington State 32 0.39% -1.0%
31 California 47 0.29% 0.1%
32 Central Florida 36 0.22% 0.1%
33 Boston College 39 0.16% -4.3%
34 Utah 35 0.14% -0.1%
35 North Texas 46 0.10% 0.1%
36 Oregon 37 0.10% -0.4%
37 Virginia Tech 40 0.09% -4.3%
38 USC 31 0.05% 0.0%
39 TCU 27 0.03% -0.2%
40 Colorado 55 0.02% 0.0%
41 Buffalo 69 0.02% 0.0%
42 Virginia 50 0.02% 0.0%
43 Indiana 53 0.01% -0.1%
44 South Florida 61 0.01% 0.0%
45 Cincinnati 65 0.01% 0.0%
46 Maryland 49 0.00% 0.0%
47 Arizona State 42 0.00% 0.0%
48 Baylor 51 0.00% 0.0%
49 Minnesota 60 0.00% -0.1%
50 Iowa State 48 0.00% 0.0%
51 Arizona 56 0.00% 0.0%
52 Vanderbilt 58 0.00% 0.0%
53 Wake Forest 66 0.00% 0.0%
54 Louisville 72 0.00% 0.0%
55 Kansas State 73 0.00% 0.0%
56 Illinois 94 0.00% 0.0%
57 Pitt 70 0.00% 0.0%
58 Tennessee 67 0.00% 0.0%
59 Kansas 83 0.00% 0.0%
60 Florida State 62 0.00% 0.0%
61 Northwestern 54 0.00% 0.0%
62 North Carolina 68 0.00% 0.0%

1.  A wild week leads to relatively straightforward results in the playoff odds.  There were six teams that have never won a title in the top 15 last week, and four of them (VT, OSU, Miss St, and BC) lost.  So we're left with a clearly-delineated top nine that contains 8 traditional powers and Stanford.  On one hand, this is boring, but on the other hand there is a lot of season left, which leaves a lot of time for upsets.

If you're reading this, you might be an ND fan, and you might be wondering why ND is only slightly higher than their preseason Playoff odds (25% vs 20%).  After all, the Irish's computer ratings are roughly the same as they were then and they've got four wins in the bank.  The main problem though is that an SOS that looked formidable a month ago is now well below average.  This means that an 11-1 ND is going to have trouble getting in, especially if enough other top teams finish 11-1 or better.

One notable oddity is that a 2-2 Texas A&M team is still in the top 25.  This is stunning given that a third loss eliminates you from contention in my model, which means that their entire case is dependent on winning out against a schedule that includes LSU, Mississippi State, Kentucky, South Carolina, and Auburn.  The Aggies are clearly a very good team so it's not that weird to dream on an unlikely run.  But it's still pretty weird.

2.  Eight more teams were eliminated this week, leaving us less than half of FBS.  Goodbye to Marshall, Louisiana Tech, Akron, Georgia Tech, Arkansas, Nebraska, Rutgers, and Oregon State.  Brief note on the Huskers: Yeah, it sucks to be this bad, but at least they're getting the stupidly hard schedule out of the way now.  2019 features Wisconsin, Iowa, and Ohio State at home, and they miss out on all the other powers from the East.

3.  We finally have a mild amount of intrigue in the conference-level playoff odds, as the ACC leaps into second place.  Just as importantly, the CUSA jumps above 0 for the first time this year.  Woohoo.

Conference Exp Playoff Teams
SEC 1.373
ACC 0.853
B10 0.782
P12 0.375
B12 0.359
IND 0.255
AMER 0.002
CUSA 0.001
MAC 0.000

4.  An interesting week in the conference title grid.  Georgia passes a mild test, but actually loses a little ground to an insurgent Kentucky.  Utah State keeps creeping up on Boise.  And Miami and USC, both somewhat written off by the media, are both at even odds to win their divisions.  Finally, something that may interest you just off the grid: Texas Tech is the third-most likely team to reach the Big 12 title game.

Conference Favorite Perc   Runner-Up Perc
ACCA Clemson 87.6%   North Carolina State 6.4%
ACCC Miami (FL) 51.9%   Duke 22.9%
AMEE Central Florida 56.3%   South Florida 19.6%
AMEW Memphis 51.1%   Houston 43.0%
B10E Ohio State 48.1%   Penn State 29.2%
B10W Wisconsin 81.7%   Iowa 9.2%
B12 Oklahoma 74.1%   West Virginia 36.7%
CUSAE Marshall 38.2%   Florida International 36.9%
CUSAW North Texas 74.7%   Louisiana Tech 13.6%
MACE Buffalo 65.8%   Ohio 14.3%
MACW Toledo 48.0%   Western Michigan 27.6%
MWCW Fresno State 77.9%   San Diego State 16.1%
MWCM Boise State 69.3%   Utah State 29.5%
P12N Washington 68.0%   Stanford 27.6%
P12S USC 49.6%   Arizona 15.2%
SECE Georgia 86.6%   Kentucky 8.3%
SECW Alabama 95.1%   LSU 2.8%
SUNE Appalachian State 81.0%   Troy 10.5%
SUNW Arkansas State 80.7%   Louisiana-Monroe 8.9%


Week 5 Preview

Home Away Home Win Prob Playoff Teams Lost
Penn State Ohio State 46.0% 0.118
Notre Dame Stanford 59.3% 0.070
Clemson Syracuse 92.1% 0.021
Texas Tech West Virginia 46.7% 0.011
Oklahoma Baylor 90.3% 0.009
Washington BYU 91.5% 0.008
Kentucky South Carolina 49.4% 0.008
North Carolina State Virginia 71.8% 0.005
Georgia Tennessee 96.5% 0.005
LSU Ole Miss 76.4% 0.004

Week 5 is now the third-most important weekend of the year, largely because of two even games involving teams in the top ten.  Other than that, the week features mostly solid home favorites, so there's not a lot else to recommend, other than a suddenly very important West Virginia-Texas Tech game and a chance for Kentucky to start 3-0 in the SEC.  Unless you're a true degenerate like me, I recommend enjoying the outdoors on the last weekend of September before taking in the two big night games.

No comments:

Post a Comment