Weekly Playoff Probabilities - Week 2

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

Rank	Team	Agg Rank	CFP Prob	Change
1	Clemson	2	83.94%	8.6%
2	Alabama	1	65.73%	3.7%
3	LSU	3	43.10%	4.8%
4	Ohio State	4	36.14%	17.5%
5	Oklahoma	6	29.60%	-5.9%
6	Wisconsin	5	24.97%	15.5%
7	Notre Dame	9	24.75%	8.1%
8	Penn State	7	18.34%	-13.2%
9	Georgia	8	9.70%	-20.1%
10	Utah	18	8.06%	-0.1%
11	Michigan State	14	7.70%	5.6%
12	Oregon	10	6.62%	4.6%
13	Florida	11	6.38%	0.6%
14	Auburn	12	6.10%	1.1%
15	Washington State	17	4.05%	1.2%
16	Michigan	16	3.76%	-16.4%
17	Iowa	19	2.94%	1.1%
18	USC	23	2.74%	2.4%
19	Baylor	22	2.04%	1.4%
20	Maryland	25	1.87%	1.6%
21	North Carolina State	35	1.78%	0.6%
22	Mississippi State	21	1.65%	0.1%
23	Oklahoma State	28	1.09%	0.1%
24	Virginia	30	1.00%	0.1%
25	Texas A&M	13	0.80%	-0.7%
26	Washington	26	0.78%	-19.5%
27	Minnesota	36	0.65%	0.3%
28	North Carolina	39	0.63%	0.5%
29	Texas	20	0.53%	-2.2%
30	Wake Forest	48	0.50%	0.4%
31	Central Florida	15	0.46%	0.3%
32	Kentucky	33	0.38%	0.0%
33	Boise State	34	0.32%	-0.2%
34	Kansas State	31	0.23%	0.2%
35	Texas Tech	32	0.20%	0.1%
36	Indiana	40	0.07%	0.1%
37	Iowa State	37	0.07%	-0.2%
38	Boston College	51	0.06%	-0.1%
39	California	50	0.05%	0.0%
40	Florida State	38	0.04%	-0.2%
41	Miami (FL)	29	0.03%	-0.4%
42	TCU	42	0.03%	0.0%
43	South Carolina	27	0.02%	0.0%
44	Arizona State	49	0.02%	-0.1%
45	Virginia Tech	47	0.02%	0.0%
46	Memphis	43	0.02%	0.0%
47	Pitt	57	0.01%	0.0%
48	Colorado	55	0.01%	0.0%
49	Stanford	45	0.01%	-0.4%
50	Louisville	65	0.00%	0.0%
51	Syracuse	66	0.00%	-0.8%
52	San Diego State	67	0.00%	0.0%
53	Air Force	64	0.00%	0.0%
54	Appalachian State	61	0.00%	0.0%
55	Wyoming	69	0.00%	0.0%
56	Ole Miss	41	0.00%	0.0%
57	SMU	56	0.00%	0.0%
58	Northwestern	62	0.00%	0.0%
59	Nebraska	46	0.00%	-0.1%
60	Illinois	79	0.00%	0.0%
61	Georgia State	101	0.00%	0.0%
62	Troy	86	0.00%	0.0%
63	West Virginia	80	0.00%	0.0%
64	Duke	59	0.00%	0.0%
65	Arkansas	81	0.00%	0.0%
66	Rutgers	100	0.00%	0.0%
67	Kansas	115	0.00%	0.0%
68	Arizona	63	0.00%	0.0%
69	Purdue	73	0.00%	0.0%
70	Georgia Tech	83	0.00%	0.0%
71	UCLA	58	0.00%	0.0%
72	Vanderbilt	87	0.00%	0.0%
73	Tennessee	52	0.00%	0.0%
74	Oregon State	90	0.00%	0.0%
75	Temple	75	0.00%	0.0%
76	UAB	111	0.00%	0.0%
77	Navy	92	0.00%	0.0%
78	New Mexico	112	0.00%	0.0%
79	Hawaii	84	0.00%	0.0%

1.  A lot of the big moves this week make sense.  Ohio State looked great against Cincinnati and climbs way up.  Michigan was barely able to move the ball on Army and takes a big dive.  Washington loses at home to Cal at 4:30 Eastern time and drops precipitously.  This is all fine and logical. 

And then there's Georgia.  The Bulldogs dropped over 20 percentage points after an uneventful win over Murray State.  Why?  Well the easy explanation is that their computer ratings dropped from 3rd to 8th, and if Georgia is merely very good rather than great, it will be tough to get out of the SEC in good enough shape to make the Playoff.  But why the significant drop in the ratings?  It turns out it's all attributable to the Sagarin ratings (25% of the aggregate), where Georgia dropped from 7th to 45th!  If you pore over the Sagarin ratings, you can see that everyone in Georgia's network of opponents (Vandy) and opponents' opponents (Purdue) is doing poorly.  The reason for this is Oregon's thumping of Nevada, who beat Purdue, who beat Vandy, who is Georgia's only opponent of note thus far.  As a result, everyone in this group is dragged down.  While it may be fair to ding Georgia a bit for not really proving anything yet, this is a clearly incorrect outlier and I'm not sure what to do about it.  My hope is that Georgia's Sagarin rating rebounds quickly in the coming weeks and that I don't have to worry about it.  For now you'll just have to bump them up a few spots in the Playoff odds in your mind.

2.  Speaking of Oregon, it might seem odd that they're ahead of Auburn after their head-to-head loss to the Tigers a week ago.  But this one actually makes perfect sense upon further review.  First, the Oregon-Auburn game came down to the last second, so neither team proved themselves to be clearly better and the only tangible benefit Auburn took from the result was not losing.  Second, Oregon did absolutely crush Nevada this week while Auburn was only OK against Tulane.  As a result Oregon is actually slightly higher in the aggregate ratings this week.  Finally, if you cheat and look below, you'll see that Oregon is now a strong favorite to win their division while Auburn is still staring up at Bama and LSU.  This discrepancy in schedule means that on average, Auburn loses almost 2 more games than Oregon over the rest of the season.  So yeah, it pays to be in the Pac 12.

3.  More Group of Five teams lost and were summarily eliminated.  Cincinnati, Marshall, Western Michigan, Ohio, North Texas, Army, Northern Illinois, Buffalo, Louisiana-Monroe, Old Dominion, UTSA, Southern Miss, Bowling Green, Central Michigan, San Jose State, Nevada, UConn, Eastern Michigan, UNLV, UTEP, Tulane, and Charlotte will not be in the Playoff, but you already knew that.

4.  Two weeks after I made fun of ACC writers for picking Virginia to win the Coastal, they are now in the drivers' seat.  You guys win this round.  In other news, Baylor is in the top 25 of the aggregate ratings, which translates to a surprising second-best odds of winning the Big 12.  We have our third Big Ten East favorite in as many weeks, with Ohio State jumping into the lead.  Utah and USC looks like a very balanced race in the Pac 12 (and might be decided next Friday when they play).  And in the yearly tradition of a team from the American coming out of nowhere to be great, congrats to SMU.

Conference	Favorite	Perc		Runner-Up	Perc
ACCA	Clemson	95.6%		North Carolina State	2.8%
ACCC	Virginia	47.5%		Miami (FL)	26.9%
AMEE	Central Florida	89.4%		Cincinnati	7.3%
AMEW	Memphis	45.5%		SMU	27.2%
B10E	Ohio State	48.4%		Penn State	25.7%
B10W	Wisconsin	71.7%		Iowa	20.6%
B12	Oklahoma	80.2%		Baylor	32.7%
CUSAE	Marshall	57.9%		Florida Atlantic	20.2%
CUSAW	Southern Miss	43.6%		North Texas	33.0%
MACE	Ohio	59.7%		Miami (OH)	25.5%
MACW	Toledo	44.0%		Western Michigan	25.4%
MWCW	Fresno State	47.4%		San Diego State	36.6%
MWCM	Boise State	59.3%		Utah State	23.7%
P12N	Oregon	65.5%		Washington State	24.1%
P12S	Utah	51.1%		USC	42.0%
SECE	Georgia	55.9%		Florida	30.1%
SECW	Alabama	69.2%		LSU	24.0%
SUNE	Appalachian State	68.0%		Troy	19.5%
SUNW	Arkansas State	66.0%		Louisiana-Lafayette	17.9%

Week 3 Preview

Home	Away	Home Win Prob	Playoff Teams Lost
UCLA	Oklahoma	10.9%	0.024
Syracuse	Clemson	3.6%	0.022
South Carolina	Alabama	10.1%	0.016
Indiana	Ohio State	14.6%	0.011
Kentucky	Florida	29.9%	0.007
Penn State	Pitt	92.4%	0.006
Washington	Hawaii	90.2%	0.004
Iowa State	Iowa	35.9%	0.003
Michigan State	Arizona State	87.6%	0.002
Georgia	Arkansas State	93.3%	0.002

On one hand this week looks awesome, as four of the very best teams face their first road tests of the year.  On the other hand, none of those home dogs are likely to pull the upset or even make it particularly close.  Which is how you get Gameday going to El Assico (which is awesome, by the way).  UCF also plays Stanford in the most consequential Group of Five opportunity of the year.