Writers of Pro Football Prospectus 2008

Most Recent FO Features

Harris-Brady.jpg

» Week 12 DVOA Ratings

Denver remains No. 1 in the Football Outsiders DVOA ratings, but New England moves up to No. 2 and has taken over as our Super Bowl favorite.

13 Nov 2013

FEI Week 11: The Games That Matter

by Brian Fremeau

Last weekend had been earmarked as one of the most important of the college football calendar as soon as team schedules were released in the off-season. The battle for SEC West supremacy between Alabama and LSU has had major BCS championship implications in each of the last five seasons. Oregon and Stanford have been perennial top-10 programs since 2010, and their Pac 12 North rivalry has factored heavily into the national championship race as well.

Both games featured top-10 teams according to last week’s FEI ratings. And the victories by Alabama and Stanford in those games were important factors in the Crimson Tide and Cardinal vaulting over Florida State in this week’s FEI ratings. But they certainly weren’t the only results that mattered.

We can figure out which games mattered most by taking another close look at Game Factors, the single-game opponent adjusted efficiency data that produces each team’s overall rating. Opponent-adjusted overall game efficiency (GFEI), offensive efficiency (OFEI), and defensive efficiency (DFEI) are provided for every team in every game, and the weight each game receives in the formula is provided as well.

Let’s begin with Alabama. The Crimson Tide now have four wins over FEI top-25 opponents (No. 14 Virginia Tech, No. 15 Texas A&M, No. 22 LSU, No. 25 Mississippi). All four games rank in the 97th percentile of all overall team performances played to date this season. Based on the weight given to those games in the FEI formula, those four results account for 64.5 percent of Alabama’s FEI rating.

FEI Game Factors - Alabama Crimson Tide
Team Wk Opponent Final NG Final GFEI Rk OFEI Rk DFEI Rk GW
Alabama 1 Virginia Tech W 35-10 W 35-10 .660 5 1.600 61 -.542 146 .170
Alabama 3 Texas A&M W 49-42 W 49-42 .542 26 2.020 17 .121 406 .166
Alabama 4 Colorado State W 31-6 W 24-6 .137 346 -.004 693 -.405 186 .093
Alabama 5 Mississippi W 25-0 W 25-0 .576 14 1.035 193 -1.436 4 .151
Alabama 6 Georgia State W 45-3 W 38-3 .287 181 .638 355 .063 378 .042
Alabama 7 Kentucky W 48-7 W 34-7 .321 156 .090 650 -.470 167 .072
Alabama 8 Arkansas W 52-0 W 35-0 .460 56 1.055 185 -.550 143 .068
Alabama 9 Tennessee W 45-10 W 42-3 .525 34 1.255 129 -.413 184 .080
Alabama 11 LSU W 38-17 W 38-17 .644 7 1.790 32 -1.271 7 .158

A blowout win over Tennessee is also highly ranked according to GFEI (96th percentile), and that cluster of five victories accounts for 72.5 percent of Alabama’s FEI rating. This cluster of victories is what distinguishes Alabama from the rest of the team resumes compiled by other contenders. The chart below plots the GFEI distribution for the Crimson Tide over each of the last five weeks. You can see how the cluster of Alabama’s best games to date has been built over time.

This chart helps illustrate the changes in opponent adjustments for each game in each week as well. While the cluster of top victories has been relatively consistent over the last month, some of the individual GFEI results have had a big impact. Which win has been Alabama’s best according to GFEI? The opening weekend victory over Virginia Tech is the answer to that question this week, but that particular result has varied considerably over the last month.

The Hokies were ranked No. 21 in the Week 7 FEI ratings, 5-1 against FBS opponents at the time. A win over Pittsburgh bumped them up in Week 8 and a close loss to Duke in Week 9 didn’t significantly change their profile. But a loss to Boston College in Week 10 dropped Virginia Tech back to No. 24, and as illustrated above, that substantially changed the FEI perspective on Alabama’s win over the Hokies. One week later, the Hokies trounced Miami on the road and the GFEI factor in that game leapt forward once again.

That Virginia Tech win over Miami had a significant impact on the Seminoles as well. A week ago, we discussed how Florida State had recorded three of the top-10 opponent-adjusted single game performances of the season to date. The win over Miami was rated then as the 7th best of the college football season. Once Miami lost to Virginia Tech, however, that win dropped to only the 29th best of the season, and Florida State’s overall profile suffered a bit.

FEI Game Factors - Florida State Seminoles
Team Wk Opponent Final NG Final GFEI Rk OFEI Rk DFEI Rk GW
Florida State 1 Pittsburgh W 41-13 W 41-13 0.721 3 2.033 15 0.055 369 0.152
Florida State 3 Nevada W 62-7 W 45-7 0.244 233 0.430 474 -0.647 111 0.082
Florida State 5 Boston College W 48-34 W 48-34 0.210 256 0.878 252 0.195 449 0.125
Florida State 6 Maryland W 63-0 W 42-0 0.337 139 2.021 16 0.015 350 0.086
Florida State 8 Clemson W 51-14 W 41-7 0.815 1 1.716 42 -1.022 26 0.176
Florida State 9 North Carolina State W 49-17 W 42-10 0.117 368 1.094 171 0.174 431 0.099
Florida State 10 Miami W 41-14 W 35-14 0.539 29 1.835 25 -1.116 15 0.182
Florida State 11 Wake Forest W 59-3 W 52-0 0.508 40 0.575 385 -0.878 49 0.098

The Seminoles have a wider distribution of game results than Alabama, and they still have two of the top-3 single-game overall performances (against Clemson and Pittsburgh). As mentioned last week, the end-of-year FEI perspective on the Seminoles may come down to how strong Clemson, Miami, and Pittsburgh finish the year. The Hurricanes didn’t help that cause last weekend.

The new No. 2 team in the FEI ratings is Stanford, fresh off their win over Oregon. That win was impressive (and would have been much more impressive in the eyes of FEI had the Cardinal not allowed a flurry of points from the Ducks late in the game). But shouldn’t Stanford be held back by their loss to Utah? The Utes, of course, are still looked upon favorably in the FEI ratings (hold that thought), though that particular result has been declining in the eyes of FEI over the last month.

FEI Game Factors - Stanford Cardinal
Team Wk Opponent Final NG Final GFEI Rk OFEI Rk DFEI Rk GW
Stanford 2 San Jose State W 34-13 W 34-13 0.158 322 0.244 584 -0.273 231 0.060
Stanford 3 Army W 34-20 W 34-20 0.018 485 -0.224 801 -0.569 137 0.052
Stanford 4 Arizona State W 42-28 W 39-14 0.794 2 1.746 38 -1.612 2 0.151
Stanford 5 Washington State W 55-17 W 41-3 0.475 54 0.543 406 -0.486 161 0.068
Stanford 6 Washington W 31-28 W 31-28 0.268 203 0.683 342 -0.305 222 0.118
Stanford 7 Utah L 21-27 L 21-27 0.417 78 0.845 270 -0.272 232 0.167
Stanford 8 UCLA W 24-10 W 24-10 0.445 64 1.222 136 -0.905 46 0.123
Stanford 9 Oregon State W 20-12 W 20-12 0.419 74 0.283 558 -0.970 32 0.110
Stanford 11 Oregon W 26-20 W 26-20 0.540 27 1.736 39 -0.982 30 0.150

In the same span, Stanford’s best win has looked better and better. Arizona State has climbed all the way to No. 4 overall in the FEI ratings this week, and Stanford’s dominant win over the Sun Devils back on September 21st has been elevated to the No. 2 overall single-game efficiency mark of the year.

I feel compelled to discuss Utah one more time (though I certainly cannot promise this will be the last). Utah lost to Arizona State 20-19 over the weekend. All eight of the Utes’ opponents to date are ranked in the FEI top 40, more than any other team. The Utes are one of only eight teams with at least three wins against FEI top-40 opponents. They are one of only six teams with at least two wins over FEI top-20 opponents.

Victories Over FEI Top-40 and Top-20 Opponents
Team Record FEI
Rank
Top-40
Wins
Top-20
Wins
Stanford 8-1 2 5 3
Arizona State 6-2 4 4 3
Alabama 9-0 1 4 2
USC 7-3 13 4 1
BYU 6-3 11 4 0
Utah 3-5 12 3 2
Notre Dame 7-3 28 3 2
Virginia Tech 6-3 14 3 0

I mentioned on Twitter over the weekend that I have been waiting for a team to finally put Utah in its place in the FEI ratings by blowing them out, and I thought the Sun Devils were a good candidate to do so. Instead, Utah grabbed a lead and nearly hung on before Arizona State prevailed in the end. Once again, turnovers doomed the Utes. In four of their five losses, the value lost on turnovers by Utah exceeded their scoring deficit in the game. They would be 7-1 (and ranked seventh) in turnover-neutral FEI.

Oregon is next for Utah. We’ll see if they can play yet another close game against yet another strong opponent. It certainly has the potential to be one of the most influential games in determining next week’s FEI ratings.

FEI Week 11 Top 25

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 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.

Other definitions:

  • SOS Pvs: Strength of schedule based on the likelihood of an elite team going undefeated against the given team's schedule to date.
  • SOS Fut: Strength of schedule based on the likelihood of an elite team going undefeated against the given team's remaining schedule.
  • FBS MW: Mean Wins, the average number of games a team with the given FEI rating would be expected to win against its entire schedule.
  • FBS RMW: Remaining Mean Wins, the average number of games a team with the given FEI rating would be expected to win against its remaining schedule.
  • OFEI: Offensive FEI, the opponent-adjusted efficiency of the given team's offense.
  • DFEI: Defensive FEI, the opponent-adjusted efficiency of the given team's defense.
  • STE: Special Teams Efficiency, the composite efficiency of the given team's special teams units - field goals, punt returns, kickoff returns, punts, and kickoffs.
  • FPA: Field Position Advantage, the share of the value of total starting field position earned by each team against its opponents.

These FEI ratings are a function of results of games played through November 9th. The ratings for all FBS teams can be found here. Program FEI (five-year weighted) ratings and other supplemental drive-based data can be found here.

Rk Team FBS
Rec
FEI LW GE GE
Rk
SOS
Pvs
Rk SOS
Fut
Rk FBS
MW
FBS
RMW
OFEI Rk DFEI Rk STE Rk FPA Rk
1 Alabama 9-0 .325 2 .400 3 .294 51 .584 34 9.9 1.7 .480 13 -.716 4 6.199 1 .607 1
2 Stanford 8-1 .315 3 .183 14 .134 13 .524 27 10.2 2.6 .408 17 -.810 2 3.856 2 .571 4
3 Florida State 8-0 .298 1 .462 1 .548 95 .831 67 10.3 2.8 .527 11 -.620 7 -.882 91 .569 6
4 Arizona State 6-2 .267 8 .123 29 .073 5 .534 30 8.2 2.4 .638 6 -.593 8 -.244 71 .529 31
5 Baylor 7-0 .267 6 .456 2 .621 106 .532 29 9.9 3.4 .661 3 -.430 17 -1.376 98 .532 28
6 Oregon 7-1 .262 4 .367 4 .176 22 .510 25 8.9 2.3 .560 10 -.455 13 1.481 32 .559 9
7 Missouri 8-1 .248 5 .217 8 .327 56 .565 33 9.1 1.4 .274 30 -.577 9 -.502 83 .543 17
8 Ohio State 8-0 .238 10 .334 6 .569 101 .782 61 9.9 2.6 .650 4 -.292 27 3.669 3 .563 8
9 Wisconsin 6-2 .237 19 .209 9 .158 18 .853 69 8.8 2.7 .480 14 -.553 11 -.458 82 .512 49
10 South Carolina 7-2 .224 12 .108 33 .188 24 .862 73 8.7 1.7 .583 9 -.447 15 -2.873 117 .464 103
11 BYU 6-3 .215 7 .084 43 .193 28 .725 50 8.2 1.6 .110 49 -.673 6 -.131 65 .518 44
12 Utah 3-5 .214 11 -.028 71 .060 4 .481 22 6.8 2.2 .358 19 -.390 22 1.810 25 .490 77
Rk Team FBS
Rec
FEI LW GE GE
Rk
SOS
Pvs
Rk SOS
Fut
Rk FBS
MW
FBS
RMW
OFEI Rk DFEI Rk STE Rk FPA Rk
13 USC 7-3 .210 13 .118 30 .161 19 .399 14 9.5 2.0 .269 32 -.732 3 .931 37 .545 14
14 Virginia Tech 6-3 .205 24 .038 53 .120 12 .970 120 8.2 1.9 .009 64 -.930 1 -1.031 93 .495 72
15 Texas A&M 7-2 .194 16 .183 15 .233 34 .356 9 7.8 1.7 .737 1 .138 85 2.673 9 .566 7
16 Georgia 5-3 .192 14 .020 59 .251 40 .448 17 7.4 1.9 .621 7 -.127 48 .242 59 .458 110
17 Auburn 8-1 .191 17 .145 23 .333 59 .372 11 8.1 .9 .301 26 -.427 18 3.405 6 .517 45
18 Central Florida 7-1 .190 18 .238 7 .437 74 .941 95 10.3 3.8 .463 15 -.233 35 2.611 10 .537 24
19 Oklahoma State 7-1 .185 15 .169 18 .668 110 .480 21 8.6 1.8 .231 38 -.492 12 .369 56 .549 11
20 UCLA 7-2 .174 29 .141 24 .059 3 .338 7 7.1 1.3 .230 39 -.417 19 2.842 8 .523 40
21 Michigan State 7-1 .174 23 .189 12 .521 89 .765 59 8.9 2.3 .011 63 -.711 5 .407 52 .573 3
22 LSU 6-3 .173 9 .127 27 .088 9 .783 62 7.3 1.5 .679 2 .110 80 2.181 16 .526 35
23 Georgia Tech 5-3 .160 26 .068 47 .238 35 .624 41 6.2 1.0 .281 28 -.270 31 1.851 23 .496 71
24 Kansas State 5-3 .157 36 .174 16 .287 50 .913 85 8.3 2.6 .272 31 -.240 34 3.185 7 .534 26
25 Mississippi 5-3 .156 20 .049 51 .082 6 .622 40 6.5 2.0 .260 35 -.415 21 -.231 70 .477 88

Posted by: Brian Fremeau on 13 Nov 2013

4 comments, Last at 15 Nov 2013, 5:56am by Brian Fremeau

Comments

1
by Will :: Wed, 11/13/2013 - 8:08pm

Arizona State & BYU combine to give Wisconsin a big boost (and I assume Ohio State a somewhat smaller one).

Will

2
by Brian Fremeau :: Thu, 11/14/2013 - 9:44am

Yep. Wisconsin's loss to Arizona State is the 13th best performance of the season per Game Factors. Their win over BYU ranks 15th: http://www.bcftoys.com/2013-game-factors/

Five losses rank among the top 50 in GFEI:
4. Washington loss to Stanford
10. Oregon loss to Stanford
13. Wisconsin loss to Arizona State
37. UCLA loss to Stanford
50. Texas A&M loss to Alabama

3
by mm(old) (not verified) :: Thu, 11/14/2013 - 9:07pm

Is the RMW being calculated correctly?:

LSU has 2 games left: A&M & Arkansas, & 1.5 RMW. Even if you give it 100% chance of winning against Arkansas, that gives it a 50% chance of beating A&M.

A&M has 2 games left: LSU & Mizzou, & 1.7 RMW. Even if you give it a 100% chance of winning against Mizzou (doubtful, since Mizzou is ranked higher), that would give it a 70% chance of beating LSU.

4
by Brian Fremeau :: Fri, 11/15/2013 - 5:56am

Good catch. There is an error there and it should read Texas A&M with only 0.8 RMW.