Writers of Pro Football Prospectus 2008

Most Recent FO Features

TomlniMik07.jpg

» Clutch Encounters: Week 4

Blowout week, but not for the Steelers. Do they play down to the competition? Also: bad Foles, Bridgewater's debut, and did J.J. Watt just end EJ Manuel's career in Buffalo?

29 Aug 2012

FEI Preseason Primer 2012

by Brian Fremeau

Another college football season is upon us, and it is finally time to roll out the FEI projections for 2012. I look forward to sharing these every year, knowing that the data will be received as both a source of insight and befuddlement.

Projecting team efficiency ratings and game outcomes for 124 teams is relatively easy at the macro level. For the most part, the top teams consistently dominate college football and are easy to identify through the projection factors I value most: five-year program success, recruiting success, returning starters, and quarterback reliance. By this methodology, the correlation of FEI projections to FEI final ratings at the end of the year is .785.

There will be misses, of course, and spectacular ones at that. Teams like Baylor and Kansas State last season would never have been identified as conference contenders by a system that significantly weights five-year efficiency. Similarly, teams like Florida and Auburn last year were trumped up in the preseason FEI projections and fell flat in a hurry due to factors not included in my data set.

Half of all teams finished within 10 ranking spots of their preseason FEI projection last year. 26 percent finished within five ranking spots. 20 percent finished within two ranking spots. The big misses draw much more attention, mine and yours. But I know that a mathematical projection system has its limitations as well, and I'm comfortable riding the wave again this year.

The newest data included in the FEI projections this year is quarterback reliance, a data point I explored in an offseason piece for ESPN Insider. It isn't particularly sophisticated, but it is useful. Teams that ran a high percentage of their offense through the quarterback position in terms of total offense production were given an extra projection boost if that starting quarterback was coming back to school this fall, and received an extra penalty if he was not. Denard Robinson is back for Michigan? That's a good thing. Cam Newton wasn't back for Auburn last year? Uh oh.

Of programs debuting new starters in 2012, teams like Oregon, LSU, Wisconsin, and Stanford were low on the quarterback reliance scale last year and their projections weren't impacted very dramatically. Teams like Boise State, Michigan State, Texas A&M, and Oklahoma State relied on the quarterback a bit more and their projections were downshifted accordingly.

This is a small impact compared to Program FEI (PFEI), the foundational component and most significant factor in my annual projections. PFEI alone has a .752 correlation with next-year FEI, and a top-20 threshold in PFEI has been a consistent benchmark for identifying potential championship teams each year. A lofty PFEI rating is why Florida is getting so much preseason love.

Florida is a good example of why it's important to review these projections in the proper context, however. The Gators might be ranked No. 5, but they rank 14th in terms of total projected wins. Notre Dame is ranked 18th in FEI, and ranked 54th in projected wins. Schedule strength matters, and even if we can't be certain that we know how tough every game on a schedule will be, the overall SOS ratings are pretty predictable. Oregon is the season's best bet to run the table, but only the eighth best team overall according to FEI. It's going to be a fun year.

As always, FEI ratings and supplemental stats will be posted here at Football Outsiders, with the 2012 stat page going up next week. Weekly game projections will be posted on my BCF Toys site.

FEI Preseason Projections

The Fremeau Efficiency Index (FEI) rewards playing well against good teams, win or lose, and punishes losing to poor teams more harshly than it rewards defeating poor teams. FEI is drive-based, not play-by-play based, and it is specifically engineered to measure the college game.

FEI is the opponent-adjusted value of Game Efficiency (GE), a measurement of the success rate of a team scoring and preventing opponent scoring throughout the non-garbage-time possessions of a game. FEI represents a team's efficiency value over average. Strength of Schedule (SOS) is calculated as the likelihood that an "elite team" (two standard deviations above average) would win every game on the given team's schedule. SOS listed here includes future games scheduled.

Mean Wins (FBS MW) represent the average total games a team with the given FEI rating should expect to win against its complete schedule of FBS opponents.

Preseason projections are a function of Program FEI ratings, returning starters, recruiting success, and quarterback reliance. The weight given to projected data will be reduced each week until Week 7, when it will be eliminated entirely. Offensive and defensive FEI ratings will also debut in Week 7.

Rank Team FBS
W-L
FEI SOS SOS
Rk
FBS
MW
1 Alabama 0-0 .286 .123 18 9.3
2 Oklahoma 0-0 .277 .143 26 9.2
3 LSU 0-0 .267 .107 12 8.9
4 USC 0-0 .223 .282 55 10.1
5 Florida 0-0 .218 .119 16 8.1
6 Florida State 0-0 .215 .185 33 7.4
7 Ohio State 0-0 .210 .318 63 10.0
8 Oregon 0-0 .206 .338 68 9.3
9 Oklahoma State 0-0 .204 .151 28 8.3
10 Georgia 0-0 .202 .193 36 8.4
11 Stanford 0-0 .201 .255 50 9.8
12 West Virginia 0-0 .193 .135 23 8.0
13 Texas 0-0 .180 .115 14 8.5
14 Virginia Tech 0-0 .177 .157 30 7.5
15 Arkansas 0-0 .177 .123 19 7.5
Rank Team FBS
W-L
FEI SOS SOS
Rk
FBS
MW
16 TCU 0-0 .168 .117 15 7.3
17 South Carolina 0-0 .161 .091 7 6.7
18 Notre Dame 0-0 .152 .079 2 7.2
19 Clemson 0-0 .152 .187 34 7.1
20 Wisconsin 0-0 .148 .360 73 8.2
21 South Florida 0-0 .147 .312 61 7.7
22 Auburn 0-0 .147 .094 9 6.6
23 Michigan 0-0 .145 .121 17 8.0
24 Texas Tech 0-0 .138 .128 21 6.8
25 Georgia Tech 0-0 .128 .179 32 6.8
26 Michigan State 0-0 .123 .241 47 7.7
27 Pittsburgh 0-0 .122 .237 46 6.1
28 Missouri 0-0 .122 .104 11 6.1
29 Nebraska 0-0 .119 .237 45 6.9
30 Boston College 0-0 .118 .194 37 6.4
Rank Team FBS
W-L
FEI SOS SOS
Rk
FBS
MW
31 Connecticut 0-0 .113 .339 70 8.2
32 Boise State 0-0 .110 .573 97 9.9
33 Iowa 0-0 .108 .427 84 7.7
34 Kansas State 0-0 .105 .093 8 5.7
35 North Carolina 0-0 .105 .378 76 7.2
36 Tennessee 0-0 .104 .110 13 5.8
37 Rutgers 0-0 .102 .253 49 7.0
38 Utah 0-0 .098 .530 94 7.9
39 Texas A&M 0-0 .093 .089 6 4.5
40 Miami 0-0 .091 .142 25 4.7
41 North Carolina State 0-0 .090 .223 42 6.0
42 Penn State 0-0 .088 .307 60 7.4
43 Louisville 0-0 .088 .366 75 6.4
44 BYU 0-0 .084 .319 65 7.8
45 Cincinnati 0-0 .079 .364 74 5.6
Rank Team FBS
W-L
FEI SOS SOS
Rk
FBS
MW
46 Central Florida 0-0 .046 .451 86 8.5
47 Mississippi State 0-0 .045 .104 10 5.3
48 Vanderbilt 0-0 .044 .200 38 4.8
49 Illinois 0-0 .034 .262 51 5.8
50 Southern Mississippi 0-0 .031 .529 93 8.1
51 Navy 0-0 .031 .568 96 7.6
52 California 0-0 .024 .150 27 4.8
53 Houston 0-0 .022 .813 123 9.5
54 Oregon State 0-0 .018 .279 54 4.9
55 Duke 0-0 .015 .129 22 4.3
56 Baylor 0-0 .006 .081 3 4.0
57 Mississippi 0-0 .005 .051 1 3.6
58 Kentucky 0-0 .001 .126 20 3.7
59 Purdue 0-0 .000 .211 41 4.6
60 Wake Forest 0-0 -.003 .223 43 4.0
Rank Team FBS
W-L
FEI SOS SOS
Rk
FBS
MW
61 Western Michigan 0-0 -.003 .708 117 7.1
62 Tulsa 0-0 -.006 .531 95 6.8
63 Virginia 0-0 -.007 .208 39 3.7
64 Louisiana Tech 0-0 -.009 .582 99 8.4
65 Arizona State 0-0 -.011 .251 48 4.4
66 Washington 0-0 -.016 .136 24 4.1
67 UCLA 0-0 -.019 .389 78 5.3
68 Central Michigan 0-0 -.022 .619 105 6.3
69 Ohio 0-0 -.023 .711 118 7.5
70 Fresno State 0-0 -.030 .410 82 6.5
71 Arizona 0-0 -.030 .163 31 3.9
72 Maryland 0-0 -.033 .154 29 2.5
73 Nevada 0-0 -.034 .642 107 6.7
74 Northwestern 0-0 -.036 .319 64 3.6
75 Temple 0-0 -.037 .305 59 3.2
Rank Team FBS
W-L
FEI SOS SOS
Rk
FBS
MW
76 Bowling Green 0-0 -.039 .343 71 6.5
77 Iowa State 0-0 -.044 .086 5 2.3
78 Northern Illinois 0-0 -.044 .716 119 6.7
79 Florida International 0-0 -.051 .679 114 7.2
80 Miami (OH) 0-0 -.052 .355 72 5.2
81 Indiana 0-0 -.065 .338 67 3.1
82 Troy 0-0 -.067 .666 113 6.9
83 East Carolina 0-0 -.069 .406 81 4.8
84 Syracuse 0-0 -.069 .188 35 2.3
85 Air Force 0-0 -.071 .614 104 6.2
86 Ball State 0-0 -.079 .523 92 4.7
87 Toledo 0-0 -.090 .768 121 4.9
88 Minnesota 0-0 -.094 .295 57 3.0
89 Arkansas State 0-0 -.094 .421 83 5.6
90 Marshall 0-0 -.095 .477 88 4.1
Rank Team FBS
W-L
FEI SOS SOS
Rk
FBS
MW
91 Louisiana Monroe 0-0 -.106 .513 90 5.7
92 Western Kentucky 0-0 -.110 .339 69 4.7
93 San Diego State 0-0 -.110 .653 110 5.4
94 SMU 0-0 -.111 .577 98 3.9
95 Kansas 0-0 -.111 .083 4 1.8
96 Utah State 0-0 -.120 .485 89 5.4
97 Kent State 0-0 -.121 .661 112 4.0
98 Colorado State 0-0 -.128 .681 115 5.6
99 Louisiana Lafayette 0-0 -.131 .322 66 4.6
100 UTSA 0-0 -.135 .859 124 4.9
101 Eastern Michigan 0-0 -.138 .600 103 3.2
102 UTEP 0-0 -.141 .303 58 3.7
103 Buffalo 0-0 -.141 .385 77 2.9
104 Army 0-0 -.143 .591 101 3.1
105 Colorado 0-0 -.145 .229 44 2.2
Rank Team FBS
W-L
FEI SOS SOS
Rk
FBS
MW
106 Washington State 0-0 -.153 .270 52 2.3
107 Hawaii 0-0 -.153 .391 79 3.8
108 Wyoming 0-0 -.158 .521 91 4.0
109 Middle Tennessee 0-0 -.166 .596 102 3.9
110 Tulane 0-0 -.168 .704 116 4.1
111 UAB 0-0 -.178 .293 56 2.5
112 Florida Atlantic 0-0 -.180 .209 40 3.2
113 Rice 0-0 -.188 .766 120 3.2
114 North Texas 0-0 -.193 .318 62 3.0
115 Akron 0-0 -.204 .654 111 2.1
116 San Jose State 0-0 -.204 .476 87 3.3
117 Memphis 0-0 -.205 .778 122 2.7
118 Massachusetts 0-0 -.216 .432 85 1.9
119 Idaho 0-0 -.228 .279 53 2.8
120 South Alabama 0-0 -.231 .638 106 2.7
Rank Team FBS
W-L
FEI SOS SOS
Rk
FBS
MW
121 Texas State 0-0 -.239 .651 109 2.7
122 New Mexico 0-0 -.242 .394 80 2.8
123 UNLV 0-0 -.257 .650 108 2.4
124 New Mexico State 0-0 -.314 .587 100 1.6

Posted by: Brian Fremeau on 29 Aug 2012

4 comments, Last at 30 Aug 2012, 4:18pm by Dad2sca

Comments

1
by cbirkemeier :: Wed, 08/29/2012 - 12:22pm

I find it interesting that Stanford is considered low on the quarterback reliance scale even though they no longer have Andrew Luck.

2
by Jeremy Billones :: Wed, 08/29/2012 - 3:21pm

They no longer have Andrew Luck, so their FEI will be reduced. However, the offense at Stanford wasn't as QB-dependent as other programs by his measure (over 2700 yds team rushing) so they won't be reduced as much as they might otherwise be.

3
by ClemsonMatt (not verified) :: Wed, 08/29/2012 - 3:34pm

I'll take the over on Wake Forest.

4
by Dad2sca :: Thu, 08/30/2012 - 4:18pm

THANKS - I have been waiting on these - feels like Christmas for grown ups.