Are the best defenses against play action the best against regular passes too? How much impact does play action really have in an NFL game, and does it correlate from year to year?
13 Sep 2013
by Bill Connelly
As you probably know by now, the Week 2 F/+ rankings are now online. For each of the first seven weeks of the season, there are still preseason projections factored into the numbers. Plus, opponent adjustments take on greater weight with each passing week. This means that you'll see teams moving up and down not only because of the events of the previous Saturday. It's just the way it is with early-season data.
You also get some fun volatility from week to week as the numbers try to figure out what the hell is going on with some teams.
Biggest Jumps from Week 1
65. Illinois (up 32 spots)
85. Washington State (up 18)
23. Georgia Tech (up 12)
30. UCF (up 12)
11. Louisville (up 11)
17. Baylor (up 11)
74. Marshall (up 11)
20. Arizona (up 10)
57. Boston College (up 10)
Biggest Tumbles from Week 1
92. San Diego State (down 31 spots)
114. Buffalo (down 19)
43. Cincinnati (down 19)
93. Kent State (down 15)
63. Arkansas State (down 15)
21. Texas (down 12)
117. UNLV (down 11)
Most of these moves, in both directions, are pretty understandable. Raw data is worth more now than it will be later in the year, so if you got blown out by a good team or blew out a bad team, your numbers are probably a little skewed, but for the most part the new rankings make more sense to the eyeballs than the old ones.
Biggest disagreements between FEI and S&P+ (S&P+ more favorable)
61. Maryland (42nd S&P+, 83rd FEI)
101. Ohio (82nd S&P+, 115th FEI)
14. Michigan State (third S&P+, 34th FEI)
39. Tennessee (24th S&P+, 53rd FEI)
71. Tulsa (59th S&P+, 86th FEI)
74. Marshall (61st S&P+, 87th FEI)
51. Arkansas (38th S&P+, 64th FEI)
70. Duke (60th S&P+, 84th FEI)
102. Southern Miss (87th S&P+, 111 FEI)
(Yes, Michigan State's third in S&P+ despite an absolutely wretched offense. I don't like it either. The Spartans' defense has been so good thus far that it's skewed the numbers dramatically. They are first in Def. S&P+ and 125th in Off. S&P+, but third overall. Boo. That will change over time. Obviously.)
Biggest disagreements between FEI and S&P+ (FEI more favorable)
68. Iowa State (55th FEI, 90th S&P+)
58. Syracuse (52nd FEI, 79th S&P+)
81. Navy (67th FEI, 94th S&P+)
73. Rice (63rd FEI, 88th S&P+)
38. Northwestern (26th FEI, 48th S&P+)
41. Kansas State (30th FEI, 51st S&P+)
22. Clemson (12th FEI, 32nd S&P+)
42. Northern Illinois (33rd FEI, 53rd S&P+)
85. Washington State (73rd FEI, 93rd S&P+)
88. Wyoming (78th FEI, 97th S&P+)
As you probably know by now, Brian and I treat FCS data differently. His system ignores FBS vs. FCS games, and he has statistical justification for going that route; I really want to use that data, but my methods have always been a little makeshift: break FCS teams into six tiers (based on Sagarin ratings) and treat games accordingly. So Kansas State lost to "FCS Tier 1" instead of North Dakota State, Mississippi State killed "FCS Tier 6" instead of Alcorn State, etc. This isn't great from a styles perspective, but it helps to nail down the quality of the opponent and, therefore, the quality of the FBS team's performance against it.
(And yeah, the quality of the FCS opponent matters: Tier 1-2 teams are 10-19 after two weeks; Tier 3-6 teams are 1-34.)
With so little data, you can explain some of the differences above by thinking about how small the data set currently is and how one game can skew things. Iowa State is 90th in S&P+ in part because it lost to Northern Iowa. Kansas State lost to North Dakota State. Of course, Arkansas thought about losing to Samford and is 38th in S&P+, so it doesn't explain everything.
It's almost as if the first two weeks of the college football season were the de facto preseason, and the real work starts this weekend. There is quite a fight card for us on Saturday, and while we know what the main event is (Alabama at Texas A&M), and we know about the main undercard matches (UCLA-Nebraska, Ole Miss-Texas, Ohio State-California, etc.), here are some of the more intriguing under-undercard games we can look forward to.
No. 7 Louisville at Kentucky (ESPN, 12:00 p.m. ET). This is Mark Stoops' first chance to make a statement as Kentucky's head coach. His Wildcats already lost to Western Kentucky (the Hilltoppers didn't have the courtesy to turn the ball over five times in six plays like they did against Tennessee) and probably won't beat Charlie Strong's rising Cardinals, but they'll be throwing some haymakers early on, and the crowd should be lively.
Bowling Green at Indiana (ESPNU, 12:00 p.m. ET). This might seem like an odd choice, especially after Indiana's loss to Navy last week, but as a college football nerd I'm interested in seeing both a) how good Bowling Green is overall (there's a chance the Falcons are the class of the MAC this year), and b) how Indiana's offense handles a certifiably solid defense. Hoosier head coach Kevin Wilson's play-calling prowess was a big piece of my book, Study Hall, but he's got his work cut out for him on Saturday.
No. 16 UCLA at No. 23 Nebraska (ABC, 12:00 p.m. ET). I've already written about this one.
Fresno State at Colorado (Pac-12, 2:00 p.m. ET). Setting aside the fact that Boulder is flooding at the moment, this game is a lovely litmus test for both schools. Colorado is 2-0 for the first time since 2008; even if the Buffaloes have just cleaned house on two cupcakes, that's not something they've been able to accomplish lately. Fresno State is easily the best team the Buffs have faced, and this is the first road trip for FSU.
Nevada at No. 10 Florida State (ESPN, 3:30 p.m. ET). Granted, nobody will be watching because of A&M-Bama, but what does Jameis Winston do for an encore after a spectacular debut?
Tennessee at No. 2 Oregon (ABC, 3:30 p.m. ET). Oregon's going to win, and probably by a lot, but how long Tennessee keeps up and puts up a fight will tell us a little bit about what to expect from the Vols from this point forward.
No. 19 Washington vs. Illinois (BTN, 6:00 p.m. ET). Washington looked spectacular in disposing of Boise State in the second half of their Week 1 battle, and a week after barely surviving Southern Illinois, Illinois looked great in blowing out Cincinnati last Saturday. This is another expectations game, as in it will sculpt our expectations of both teams for the remaining months of the season.
Mississippi State at Auburn (ESPN2, 7:00 p.m. ET). How seriously should we be taking either of these teams?
Marshall at Ohio (ESPN News, 8:00 p.m. ET). How seriously should we be taking Marshall, which has pretty much wreaked havoc (against bad teams) in the first two weeks?
Oregon State at Utah (FS1, 10:00 p.m. ET). Utah has scored 100 points in two weeks, and new offensive coordinator Dennis Erickson's tweaks seem to have taken hold to some degree. Can the Utes take care of business against an Oregon State team that has looked quite disappointing so far?
No. 20 Wisconsin at Arizona State (ESPN, 10:30 p.m. ET). This might be the most exciting game of the week from a pure aesthetics standpoint. Both of these teams could be incredibly underrated at the moment, and there's just enough of a contrast in styles to put on a wonderful show as a good portion of the country is going to bed. Watch this one! Or at least DVR it!
The preseason is over. Let the real games begin.
Because of last weekend's project -- a Big Ten road trip that occupied me through Monday (I will be publishing a longform about the trip next week, probably Wednesday) -- it was an abbreviated week of writing for me at SBN.