Writers of Pro Football Prospectus 2008

Most Recent FO Features

JohnsonKer18.jpg

» Seventh Day Adventure: Week 13

The biggest game this week is the Iron Bowl, where the playoff hopes of Alabama, Auburn, and Georgia hang in the balance.

22 Oct 2014

FEI Week 8: What's With A&M?

by Brian Fremeau

The FEI ratings often challenge conventional wisdom, but few seasons have provided more peculiar results than the 2014 season to date. The top five teams in the current Associated Press poll -- Mississippi State, Florida State, Mississippi, Alabama, and Auburn -- rank 18th, seventh, sixth, 14th, and 11th in this week's FEI ratings. The top five teams according to FEI -- Oregon, Georgia, USC, Arizona, and TCU -- rank sixth, ninth, 20th, 15th, and 10th in the Associated Press poll. I'm not sure about the wisdom part, but FEI is exceptionally unconventional this year.

Consider Texas A&M. Three weeks ago, the Aggies were ranked fifth in FEI. Three weeks later, following losses to three of the top four teams in the country according to conventional wisdom, Texas A&M has fallen to 84th in FEI, behind the likes of Ball State, FIU, Hawaii, and Purdue. That can't be right, can it? There are more than 100 other rating systems in the Massey Consensus and only one other system ranks Texas A&M lower than 36th this week. What in the world is FEI measuring that those systems are not?

On October 4, Texas A&M lost 48-31 to Mississippi State. The score was 48-17 in non-garbage time possessions before the Aggies added two late touchdowns to pull to within 17 points. On October 11, A&M lost 35-20 to Ole Miss. The score was 35-7 in non-garbage possessions before the Aggies added two late touchdowns to pull to within 15 points. On October 18, A&M lost 59-0 to Alabama. The score was 45-0 at the half and 52-0 after one possession in the second half, the end of non-garbage time according to FEI. In total, the Aggies lost three games by a combined final score of 142-51. In non-garbage time, the combined final score of those three games was 135-24.

There have been a total of 391 FBS games played through the first eight weeks of the season, a total of 782 single-game performances (one for each team in each game). Each of Texas A&M's last three games ranks among the bottom 100 single-game unadjusted performances of the year to date according to game efficiency. The list of teams with three or more bottom-100 GE performances to date is not a good one: SMU, Appalachian State, Eastern Michigan, Fresno State, North Texas, Troy, and Texas A&M. The Aggies ranked ninth in overall game efficiency three weeks ago. They rank 79th in overall game efficiency today.

Texas A&M has three losses of 28 points or more in non-garbage time on the books. Only SMU (five), Fresno State (three), and Troy (three) also share that distinction so far in 2014. Last year, 24 teams finished the year with three or more non-garbage time losses of at least 28 points each. The average end-of-year FEI rank of those 24 teams was 105th. Only two -- Washington State (56th) and Iowa State (72nd) -- finished the year ranked among the top 80 teams in the nation.

Yeah, but Texas A&M just played three of the top teams in the country. Losing badly isn't anything to brag about, but shouldn't opponent adjustments soften that blow? From the perspective of FEI, there is not yet enough data to confidently say so. Texas A&M has seven FBS games in the books, and only its opening weekend rout of South Carolina measures as a particularly strong result. The loss to Mississippi State is A&M's second best GFEI of the year, and only the 265th best performance nationally to date.

The Aggies aren't doing any favors for their SEC brethren either. Alabama's win was exceptionally dominant (No. 1 in game efficiency, No. 3 in opponent-adjusted game efficiency for the Crimson Tide), but Mississippi State's win over A&M is only its fourth best GFEI. Ole Miss has had four better GFEI performances than its win over A&M. Strength of schedule in the SEC has taken a hit as well, particularly for those teams that don't face the SEC West gauntlet. Missouri has a grand total of one opponent this year currently ranked in the FEI top 40: Georgia, to whom the Tigers lost at home on October 11 by a final score of 34-0. Missouri has the nation's 81st-ranked schedule according to FEI, ranked alongside the likes of Colorado State (5-1) in terms of overall difficulty.

I introduced the concept of "degree of difficulty" before the season kicked off, a measure by which arguments might be settled over which team records are most deserving of consideration for one of the coveted playoff spots. The degree of difficulty metric is based on the FEI ratings, but instead of measuring efficiency against schedule, it simply measures record against schedule. How difficult would it be for an elite team (two standard deviations better than average) to play a given team's schedule to date and achieve that team's record?

Degree of Difficulty: Record Against Schedule To Date
DOD
Rank
Team Record DOD FEI FEI
Rank
SOS
Pvs
Rank SOS
Fut
Rank Win
Out
1 Florida State 6-0 .529 .244 7 .529 77 .433 40 .300
2 Mississippi 7-0 .581 .249 6 .581 87 .466 43 .337
3 Mississippi State 6-0 .610 .201 18 .610 90 .304 25 .089
4 Arizona 5-1 .658 .254 4 .179 7 .174 7 .103
5 Oregon 5-1 .674 .306 1 .236 14 .313 27 .373
6 Arizona State 4-1 .788 .208 16 .293 24 .192 9 .036
7 Auburn 5-1 .807 .228 11 .360 39 .148 4 .043
8 Georgia 6-1 .811 .262 2 .386 43 .583 61 .513
9 Michigan State 5-1 .813 .168 25 .255 18 .518 52 .150
10 UCLA 5-2 .816 .239 9 .127 1 .315 28 .146
11 Colorado State 5-1 .827 .176 23 .407 49 .890 99 .697
12 Alabama 6-1 .845 .213 14 .372 40 .422 37 .176
13 Kansas State 4-1 .845 .200 19 .379 42 .225 15 .054
14 Utah 4-1 .849 .241 8 .401 46 .100 2 .034
15 Baylor 5-1 .860 .202 17 .443 61 .424 38 .165
DOD
Rank
Team Record DOD FEI FEI
Rank
SOS
Pvs
Rank SOS
Fut
Rank Win
Out
16 TCU 4-1 .867 .253 5 .424 55 .536 54 .436
17 Notre Dame 6-1 .869 .182 21 .406 48 .252 20 .040
18 USC 5-2 .884 .260 3 .182 8 .185 8 .123
19 Duke 5-1 .886 .170 24 .468 68 .506 50 .149
20 Clemson 4-2 .891 .224 12 .156 4 .602 67 .447
21 Marshall 6-0 .913 .162 26 .913 127 .925 116 .752
22 East Carolina 4-1 .918 .147 31 .481 72 .903 104 .591
23 Nebraska 5-1 .931 .196 20 .557 81 .674 77 .385
24 Minnesota 5-1 .944 .056 51 .464 65 .401 36 .005
25 Ohio State 5-1 .946 .213 13 .583 88 .500 49 .287
26 West Virginia 4-2 .949 .141 32 .313 30 .434 41 .048
27 Oklahoma 5-2 .951 .229 10 .269 19 .680 79 .501
28 Boise State 5-2 .962 .082 47 .341 34 .849 92 .212
29 Washington 4-2 .973 .087 45 .232 13 .209 13 .001
30 Georgia Tech 4-2 .983 .177 22 .413 50 .232 17 .030
DOD
Rank
Team Record DOD FEI FEI
Rank
SOS
Pvs
Rank SOS
Fut
Rank Win
Out
31 LSU 5-2 .988 .057 50 .397 45 .481 46 .015
32 Louisville 5-2 .989 .154 29 .373 41 .325 30 .045
33 Oklahoma State 4-2 .990 .044 52 .305 27 .193 10 .000
34 Maryland 4-2 .991 .101 38 .563 83 .628 70 .075
35 Oregon State 3-2 .991 .088 43 .336 33 .203 12 .001
36 Rutgers 4-2 .993 -.011 66 .435 58 .342 33 .000
37 Missouri 4-2 .994 .137 34 .498 75 .775 88 .366
38 Central Florida 3-2 .995 .035 55 .577 86 .675 78 .089
39 Georgia Southern 4-2 .997 .010 60 .623 93 .907 105 .364
40 Iowa 4-2 .998 .114 36 .536 79 .653 73 .114
41 Arkansas State 3-2 .998 -.008 63 .637 97 .913 111 .509
42 Air Force 4-2 .999 .000 62 .772 111 .691 80 .022
43 Kentucky 4-2 .999 -.095 89 .759 109 .231 16 .000
44 Northern Illinois 4-2 .999 -.143 101 .781 112 .883 96 .004
45 Penn State 4-2 1.000 .066 48 .824 119 .564 59 .019
DOD
Rank
Team Record DOD FEI FEI
Rank
SOS
Pvs
Rank SOS
Fut
Rank Win
Out
46 Wisconsin 3-2 1.000 .024 58 .805 116 .515 51 .002
47 Temple 3-2 1.000 -.061 80 .863 123 .619 69 .000
48 South Alabama 4-2 1.000 -.140 100 .725 104 .746 85 .001
DOD - Likelihood that an elite team would have the given team's record against the given team's schedule
SOS Pvs - Strength of Schedule of all games played to date
SOS Fut - Strength of Schedule of all games remaining to be played
Win Out - Likelihood given team will win all of its remaining games

Degree of difficulty is not a substitute for FEI, but it certainly seems more likely that the playoff selection committee will reward "record against schedule" rather than "efficiency against schedule" since the loss column means so much in traditional college football rankings. To date, no team record has been more impressive than Florida State's. Along with the Seminoles, Oregon is the only other team with two victories against current FEI top 25 opponents. Oregon tops FSU (and all other teams) in FEI because those two victories were so dominant whereas Florida State had close calls, but the Seminoles remain undefeated. That is likely to matter more to the committee, and we'll find out next week whether that assumption is true. I'll continue to post the degree of difficulty table from this point forward as part of the weekly FEI column as the number of teams with two or fewer losses continues to dwindle.

FEI 2014 Week 8 Ratings

The Fremeau Efficiency Index (FEI) is a college football rating system based on opponent-adjusted drive efficiency. Nearly 20,000 possessions are contested annually in FBS vs. FBS games. First-half clock-kills and end-of-game garbage drives and scores are filtered out. Game Efficiency (GE) is a function of the starting field position and outcome of non-garbage possessions. Opponent adjustments are calculated with special emphasis placed on quality performances against good teams, win or lose.

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 all regular season games scheduled. A multifaceted approach to measuring schedule strength is available here and visualized here.

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. Remaining Mean Wins (FBS RMW) represent the average total games a team with the given FEI rating should expect to win against the remaining opponents on its schedule.

Offensive FEI (OFEI) is opponent-adjusted offensive efficiency. Defensive FEI (DFEI) is opponent-adjusted defensive efficiency. Special Teams Efficiency (STE) is the composite efficiency of field goal kickoff, punt, and return units. Field Position Advantage (FPA) is the average share of field position value generated by offense, defense, and special teams efficiency.

These ratings are exclusively produced from 2014 game data and are not influenced by preseason projections. Complete ratings and ratings splits for all 128 FBS teams are available here. Supplemental data including points per drive, game splits, and game factors are available as well.

Rk Team FBS
Rec
FEI LW GE GE
Rk
SOS Rk
FBS
MW
FBS
RMW
OFEI Rk DFEI Rk STE Rk FPA Rk
1 Oregon 5-1 .306 1 .199 11 .074 18 9.1 4.2 .761 3 -.368 27 -.518 77 .533 25
2 Georgia 6-1 .262 7 .279 4 .225 53 9.4 3.4 .588 10 -.260 38 2.801 10 .608 1
3 USC 5-2 .260 3 .152 18 .034 5 8.8 3.4 .584 11 -.270 36 -.548 80 .498 62
4 Arizona 5-1 .254 5 .089 28 .031 4 8.7 4.3 .537 14 -.292 33 -.370 74 .548 16
5 TCU 4-1 .253 4 .206 9 .227 54 9.4 5.3 .475 18 -.572 11 2.204 15 .562 9
6 Mississippi 7-0 .249 2 .301 3 .270 64 9.3 3.1 .400 23 -.700 5 1.677 22 .563 8
7 Florida State 6-0 .244 12 .163 16 .229 55 9.1 4.0 .577 12 -.463 20 .547 48 .487 79
8 Utah 4-1 .241 10 .126 23 .040 7 7.5 3.5 -.180 83 -.668 8 4.222 1 .572 4
9 UCLA 5-2 .239 14 .046 46 .040 6 8.2 3.5 .797 2 -.211 43 .195 63 .495 67
10 Oklahoma 5-2 .229 11 .195 12 .183 47 9.8 4.4 .664 6 -.273 35 2.818 9 .543 20
11 Auburn 5-1 .228 13 .192 13 .053 10 7.8 3.0 .515 16 -.373 26 1.400 25 .542 21
12 Clemson 4-2 .224 20 .074 30 .094 22 8.3 4.4 .335 27 -.679 6 -1.118 96 .509 45
Rk Team FBS
Rec
FEI LW GE GE
Rk
SOS Rk
FBS
MW
FBS
RMW
OFEI Rk DFEI Rk STE Rk FPA Rk
13 Ohio State 5-1 .213 25 .309 2 .291 72 10.1 5.0 .447 20 -.352 28 2.888 8 .598 2
14 Alabama 6-1 .213 15 .262 5 .157 43 8.3 2.6 .402 22 -.586 10 -1.338 98 .472 96
15 Virginia Tech 3-3 .209 19 .058 37 .128 34 8.0 3.8 -.222 88 -.846 1 1.690 21 .508 47
16 Arizona State 4-1 .208 24 .056 39 .056 11 6.9 3.6 .650 7 -.295 32 .677 43 .484 85
17 Baylor 5-1 .202 8 .204 10 .188 50 8.4 3.8 .330 28 -.446 21 .453 52 .551 15
18 Mississippi State 6-0 .201 6 .245 7 .186 48 8.7 3.5 .265 33 -.564 12 .213 61 .535 22
19 Kansas State 4-1 .200 42 .173 14 .086 20 7.7 4.1 .269 32 -.532 15 .553 47 .545 18
20 Nebraska 5-1 .196 26 .245 6 .375 80 9.2 4.2 .206 39 -.481 18 2.140 16 .563 7
21 Notre Dame 6-1 .182 17 .168 15 .102 26 8.5 3.0 .256 34 -.463 19 1.088 28 .519 38
22 Georgia Tech 4-2 .177 16 .057 38 .096 24 7.0 2.7 1.066 1 .401 100 2.106 17 .526 30
23 Colorado State 5-1 .176 18 .066 33 .362 79 9.0 4.7 .445 21 -.057 65 1.786 19 .496 65
24 Duke 5-1 .170 28 .157 17 .237 58 8.2 3.6 .055 57 -.676 7 2.948 6 .558 11
25 Michigan State 5-1 .168 32 .210 8 .132 36 7.8 3.5 .152 45 -.425 23 .399 56 .547 17

Posted by: Brian Fremeau on 22 Oct 2014

3 comments, Last at 22 Oct 2014, 3:47pm by skunkfunk

Comments

1
by skunkfunk :: Wed, 10/22/2014 - 12:22pm

How about doing some playoff odds based on this? Is there any way to do some kind of estimate given the human nature of the selection?

2
by Brian Fremeau :: Wed, 10/22/2014 - 12:57pm

There's always a way! We have produced playoff likelihood by attempting to weight measurable factors such as likelihood of going undefeated, 1-loss, strength of schedule, conference power, etc. The more I think about it though, I'm betting the committee will settle on something more like degree of difficulty. Not this specific version, of course, but I'm betting it will boil down to comparing record (not dominance, not efficiency, not best win, not worst loss. RECORD) against schedule. Since they aren't using a single SOS measure, that part will be subjective, but I think the number in the loss column will drive the discussion.

3
by skunkfunk :: Wed, 10/22/2014 - 3:47pm

Well, I for instance see that Oklahoma and Ohio State have a very high FBS MW, but that doesn't say much for their playoff odds. It would be a huge surprise to see both of them in the playoff, even given that their relatively average schedules should be no hindrance.