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.

10 Dec 2014

FEI Week 15: Tiebreakers

by Brian Fremeau

Much of the debate over the final weekend of the college football regular season revolved around two Big 12 contenders, Baylor and TCU. In every set of College Football Playoff rankings prior to the final release, the selection committee ranked the Horned Frogs ahead of Baylor. TCU was ranked six spots ahead of Baylor in the initial rankings set released on October 28 and maintained a position at least two ranking spots better than the Bears for six straight weeks. In Sunday's final rankings, Baylor slipped one spot ahead of TCU.

Depending on your perspective, the debate between Baylor and TCU was either incredibly simple or incredibly complex. Baylor and TCU played against one another back on October 18 in Waco, Texas. Baylor won that game by a final score of 61-58, scoring 24 unanswered points in the fourth quarter to win. Unlike every other major conference, Big 12 teams play a full round-robin schedule. Baylor and TCU both finished the year 11-1, 8-1 in conference play. Baylor had the head-to-head victory over TCU, commonly used as the primary tiebreaker to determine a conference champion. Baylor ahead of TCU is a no-brainer for many.

The committee didn't see it that way until the final weekend, which vexed many fans who have grown so accustomed to the way in which voted polls have traditionally operated over the years. Baylor and TCU both won on Saturday, and the Horned Frogs did so in dominating fashion. Why did they get passed as a result? In the committee's eyes, TCU's one-loss record was more impressive than Baylor's one-loss record until the final week. What changed in that final week was the conclusion of round-robin play in the Big 12 -- TCU finally played the league's last place team (Iowa State), and Baylor finally played the league's third-best (Kansas State). Until the end of the day Saturday, the teams had not faced comparable conference schedules. Once they did, the committee responded and pushed the head-to-head winner ahead of the head-to-head loser.

I agree with the committee decision, but I also understand the perspective that TCU is better than Baylor and deserving of a higher ranking. For one thing, the numbers agree with that assessment. The final pre-bowl FEI ratings give an edge to TCU in terms of their overall opponent-adjusted efficiency. In terms of degree of difficulty, TCU's one-loss record was more difficult to achieve than Baylor's due to TCU's stronger non-conference opposition. TCU has a stronger DFEI, STE, and FPA rating than Baylor, and on a neutral field, my ratings would give TCU a 55 percent likelihood of defeating the Bears.

But what if we stripped away all the data except for the games contested only in league play? I ran an alternate set of Big 12 FEI ratings using only the 45 games played in conference. Through this lens, Baylor finishes ahead of TCU in overall FEI by a margin almost identical to TCU's margin in the official FEI ratings. Baylor would have a 55 percent likelihood of beating TCU on a neutral field if the only data included in our analysis was Big 12 game data.

TCU is better than Baylor. Baylor is better than TCU. TCU is more deserving of a higher ranking than Baylor. Baylor is more deserving of a higher ranking than TCU. The difference between these statements is razor-thin and depends entirely on your perspective.

The debate was resolved to the satisfaction of the majority of football fans by using a head-to-head result as the tiebreaker. The intensity of the debate was diffused to the satisfaction of the majority of football fans by the committee's selection of Ohio State in the playoff field and keeping both Baylor and TCU on the outside looking in.

My degree of difficulty ratings make the case for Ohio State as well (due to the Buckeyes' having played two more FBS opponents than either TCU or Baylor), but I'm not content to simply nod in agreement with the committee. My primary concern about the College Football Playoff selection process is consistency, and I think this debate is one that will need to be addressed again and again in the coming years. Does every game result matter? Do only a subset of game results matter most? Does head-to-head only come into play when it is otherwise too close to call? Are the best teams selected? Are the most deserving teams selected? Can a balance be struck between these questions and delivered consistently year after year? Time will tell.

This is the final installment of the FEI ratings column in the calendar year 2014. I'll produce a new set of FEI ratings following the New Year's Day semifinals in January.

Final 2014 Degree of Difficulty Rankings

The degree of difficulty rankings (DOD) are based on current FEI ratings, but instead of measuring efficiency against schedule, DOD 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?

My hypothesis is that the inaugural College Football Playoff selection committee is likely to value and reward something akin to DOD through their process and deliberations. The current selection committee rankings are provided in the table for comparison.

Degree of Difficulty: Record Against Schedule To Date
DOD
Rank
Team Record DOD FEI FEI
Rank
SOS
Pvs
Rank CFP
1 Florida State 12-0 .156 .262 4 .156 42 3
2 Alabama 11-1 .336 .315 2 .085 17 1
3 Oregon 11-1 .458 .317 1 .145 38 2
4 Ohio State 12-1 .702 .257 5 .292 62 4
5 TCU 10-1 .717 .229 8 .289 61 6
6 Baylor 10-1 .725 .218 9 .325 69 5
7 Mississippi State 9-2 .754 .218 10 .076 13 7
8 Arizona 10-3 .808 .183 14 .029 2 10
9 UCLA 9-3 .853 .208 12 .079 14 14
10 Georgia Tech 9-3 .885 .266 3 .089 18 12
11 Mississippi 8-3 .892 .242 7 .099 21 9
12 Clemson 8-3 .900 .168 18 .066 11 17
DOD
Rank
Team Record DOD FEI FEI
Rank
SOS
Pvs
Rank CFP
13 Michigan State 9-2 .912 .165 19 .148 39 8
14 Auburn 7-4 .918 .214 11 .026 1 19
15 Georgia 8-3 .923 .256 6 .139 34 13
16 Missouri 9-3 .944 .148 23 .110 24 16
17 Kansas State 8-3 .953 .146 24 .138 33 11
18 Louisville 8-3 .955 .176 15 .188 50 21
19 Arizona State 8-3 .963 .175 16 .165 43 15
20 LSU 7-4 .967 .135 27 .066 10 23
21 Boise State 11-2 .971 .107 36 .391 79 20
22 Colorado State 9-2 .973 .117 32 .437 86 -
23 Utah 7-4 .980 .156 21 .079 15 22
24 USC 8-4 .984 .203 13 .107 23 24
25 Duke 8-3 .984 .140 26 .217 56 -

I think my instincts were on point. The top four teams in the final DOD rankings were selected to participate in the inaugural College Football Playoff. The Group of Five selection, Boise State, is the DOD pick as well, ranked 20th by the selection committee and 21st in DOD. Twenty-three of the 25 teams ranked by the selection committee appear in the DOD rankings. I will continue to use DOD as a tool to anticipate and evaluate the College Football Playoff selection process in the future.

FEI 2014 Week 15 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 games scheduled, including bowl games. 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 11-1 .317 2 .280 2 .085 22 11.4 .7 .718 3 -.415 18 1.758 12 .549 8
2 Alabama 11-1 .315 1 .215 7 .051 7 11.0 .7 .666 5 -.617 5 -.595 91 .487 87
3 Georgia Tech 9-3 .266 3 .143 18 .060 14 10.3 .6 .968 1 -.080 57 1.224 23 .541 15
4 Florida State 12-0 .262 6 .123 22 .062 15 10.4 .3 .649 7 -.419 17 .553 43 .487 85
5 Ohio State 12-1 .257 8 .295 1 .118 35 11.9 .3 .560 9 -.487 12 1.275 20 .576 3
6 Georgia 8-3 .256 4 .213 8 .105 30 9.5 .7 .558 10 -.381 21 1.802 11 .585 1
7 Mississippi 8-3 .242 5 .164 11 .065 17 9.0 .5 .251 34 -.699 3 .715 39 .534 18
8 TCU 10-1 .229 7 .225 4 .183 55 9.8 .5 .324 23 -.607 6 1.837 10 .568 4
9 Baylor 10-1 .218 14 .223 5 .252 65 10.0 .7 .504 12 -.353 23 .913 33 .553 7
10 Mississippi State 9-2 .218 10 .198 9 .043 4 8.6 .4 .333 21 -.518 10 .462 48 .526 24
11 Auburn 7-4 .214 11 .073 33 .021 1 7.7 .7 .722 2 -.208 41 -.340 79 .505 52
12 UCLA 9-3 .208 12 .056 43 .064 16 9.0 .7 .667 4 -.090 54 .450 49 .497 66
Rk Team FBS
Rec
FEI LW GE GE
Rk
SOS Rk
FBS
MW
FBS
RMW
OFEI Rk DFEI Rk STE Rk FPA Rk
13 USC 8-4 .203 13 .135 19 .090 23 9.5 .7 .302 28 -.363 22 .535 44 .516 32
14 Arizona 10-3 .183 9 .069 36 .026 2 9.3 .7 .341 20 -.295 33 -.254 75 .531 22
15 Louisville 8-3 .176 18 .088 28 .112 31 8.2 .3 .129 47 -.636 4 -.498 88 .472 109
16 Arizona State 8-3 .175 16 .063 39 .135 41 8.1 .6 .324 24 -.346 24 .691 41 .532 21
17 Arkansas 5-6 .171 17 .078 31 .048 5 7.3 .9 .309 26 -.388 20 -.385 81 .525 28
18 Clemson 8-3 .168 21 .095 25 .052 10 7.6 .5 -.049 63 -.867 1 -1.263 107 .498 64
19 Michigan State 9-2 .165 22 .216 6 .101 28 8.6 .3 .333 22 -.187 43 1.556 14 .549 9
20 Stanford 6-5 .161 23 .092 26 .052 8 7.4 .8 .117 50 -.512 11 -.488 86 .506 49
21 Utah 7-4 .156 24 .027 60 .068 19 7.1 .6 -.115 74 -.463 14 2.674 5 .537 16
22 Oklahoma 8-4 .155 19 .170 10 .197 58 9.4 .5 .482 13 -.170 46 1.210 24 .522 30
23 Missouri 9-3 .148 20 .078 32 .095 25 8.5 .6 .086 53 -.405 19 .530 45 .490 78
24 Kansas State 8-3 .146 25 .159 14 .097 26 7.8 .3 .440 15 -.027 60 2.750 4 .546 12
25 Virginia Tech 5-6 .144 27 .011 64 .122 36 7.4 .8 -.224 91 -.721 2 -.001 67 .483 97

Posted by: Brian Fremeau on 10 Dec 2014

1 comment, Last at 13 Dec 2014, 12:14pm by DFJinPgh

Comments

1
by DFJinPgh :: Sat, 12/13/2014 - 12:14pm

Didn't the BCS decide this long ago? It was 1998 (I think), and FSU was 2, Mia 3, Wash 4. The major reason for that order (each team had 1 loss, I think Tenn was 1 and undefeated) was that Wash beat Mia, and Mia beat FSU. Best loss decided those rankings.

Based on that, assuming my memory is correct, TCU should definitely be ranked ahead of Baylor. Yes?