Writers of Pro Football Prospectus 2008

Most Recent FO Features

SandersEmm10.jpg

» Scramble for the Ball: Quarter Pole Projections

Mike and Tom weigh the chances of this year's class of receivers, running backs and tight ends who are on pace to break the magical 1,000-yard mark for the first time.

01 Oct 2008

FEI Week 5 Ratings

by Brian Fremeau

The Fremeau Efficiency Index principles and methodology can be found here. Like DVOA, FEI rewards playing well against good teams, win or lose, and punishes losing to poor teams more harshly than it rewards defeating poor teams. Unlike DVOA, it 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, a measurement of the success rate of a team scoring and preventing opponent scoring throughout the non-garbage-time possessions of a game. Like DVOA, it represents a team's efficiency value over average.

Only games between FBS teams are considered. Since limited data is available at the beginning of the season, the ratings to date are a function of both actual games played and projected outcomes based on the 2008 Projected FEI Ratings. The weight given to projected outcomes is reduced each week until mid-October, at which point the projections will be eliminated entirely.

Rank Team Rec FEI vs. Top 10 vs. Top 40
1 Oklahoma 3-0 0.268 0-0 0-0
2 Penn State 4-0 0.234 0-0 0-0
3 Florida 3-1 0.232 0-0 0-1
4 Virginia Tech 3-1 0.228 0-0 3-1
5 USC 2-1 0.220 0-0 1-0
6 Texas 4-0 0.211 0-0 0-0
7 Alabama 5-0 0.209 0-0 1-0
8 Wake Forest 3-1 0.202 0-0 2-0
9 Missouri 3-0 0.191 0-0 0-0
10 Boise State 2-0 0.186 0-0 1-0
11 BYU 3-0 0.183 0-0 0-0
12 Mississippi 2-2 0.178 1-1 1-2
Rank Team Rec FEI vs. Top 10 vs. Top 40
13 Wisconsin 3-1 0.176 0-0 0-0
14 Vanderbilt 4-0 0.169 0-0 1-0
15 Utah 4-0 0.169 0-0 0-0
16 LSU 3-0 0.164 0-0 1-0
17 Georgia Tech 2-1 0.166 0-1 1-1
18 Georgia 3-1 0.164 0-1 0-1
19 Texas Tech 2-0 0.163 0-0 0-0
20 Oklahoma State 3-0 0.157 0-0 0-0
21 Florida State 1-1 0.156 0-1 0-1
22 North Carolina 2-1 0.153 0-1 0-1
23 Northwestern 4-0 0.153 0-0 2-0
24 Kansas 2-1 0.145 0-0 0-1
25 South Florida 4-0 0.145 0-0 1-0

The complete Week 5 FEI Ratings for all 120 FBS teams can be found here.

I have to confess that while USC was getting dominated on the line of scrimmage in Corvallis from the opening kickoff last Thursday night, our household was glued to the season premiere of The Office. (Spoiler Alert! Ryan is back, Dwight and Angela are ratcheting up their scandalous affair, and Jim and Pam got engaged.) I'm confident that Michael Scott was watching the game like many of you, if only for the sole purpose of snickering adolescently at any mention of the two teams' mascots. Our TV finally switched over at halftime in time to hear the real shocker of the night. 21-zip Oregon State over USC?

And as it turned out, Thursday night was nothing compared to Saturday. A collective "Huh?!" echoed in stadiums and in living rooms across the land as score updates rolled in. Michigan did what?! Florida lost how?! Georgia is down by how much?! A week ago, four SEC superpowers had positioned themselves firmly in the FEI top 10, Southern Cal was supposed to cruise to the BCS championship game, and teams that lost to weaker opponents were never supposed to move up in the rankings. At the count of three, everybody, let's review the Week 5 FEI Ratings: One, two, three ... Huh?!

It's important to keep a few key things in perspective. First, the 2008 Projected FEI Ratings are still in play, though their weight is diminished -- next week will be the last set of ratings that include projected data. Florida and USC are the primary beneficiaries of that projection influence at this point, softening the blow of their weekend losses and keeping them in the mix at the top. The more modest preseason expectations of Alabama are likewise holding them back just a little bit. Also, there still aren't a ton of games on the books, so FEI is significantly influenced by how competitive losing teams have been in their losses. No. 18 Georgia is the highest-ranked team to lose a game this year by more than a single score. Ohio State, meanwhile, hasn't yet recovered from its beatdown at the hands of USC.

The biggest factor in play in this week's ratings has everything to do with the question of who has played whom. Eighteen total games have been played between teams currently ranked in the FEI top 40. No. 4 Virginia Tech has played in four of those games, winning three, including two victories in a row on the road. Their fellow ACC member Wake Forest is one of only two other teams with multiple victories against the FEI top 40, including its win over No. 12 MIssissippi. The Rebels were bumped all the way up after their thrilling win over the Gators to add to competitive losses to Wake and undefeated Vanderbilt. And so on and so forth. There's obviously some circular logic at work in any opponent-adjusted system, and the effects of the relatively strong Virginia Tech and Wake Forest chains is probably artificially inflating a whole group of team profiles. But as the table demonstrates, most of the other teams simply haven't played anyone yet. None of the six Big 12 teams positioned in this week's FEI top 25 has defeated an FEI top 40 opponent yet this year; only Kansas has even played one.

So do we really know anything yet? Vanderbilt, Northwestern, and Duke are a combined 10-1 in FBS games so far. The Mountain West Conference has a better record against BCS conference opponents than any other league. Home SEC teams are 3-8 in conference games so far this season. Knowledge is power, but some of the results so far seem more like rumor than fact. We need more head-to-head results of teams at the top, but it might take a few more weeks to get them.

Pinning the Opponent Deep

With about 11 minutes to play in the Oregon State game, the Beavers lined up to punt from about midfield, clinging to a seven-point lead. The kick wasn't pretty, but it rolled deep into USC territory and was downed at the 6-yard line. Three-and-out, punt for USC. After blocking an OSU field goal attempt on the next series, the Trojans dug themselves a hole with a penalty that moved the ball back to their own 14-yard line. Three-and-out, punt. The next Oregon State possession burned all of USC's timeouts, and with 3:30 to play, another fortunate OSU punt rolled to a stop at the USC 2-yard line. Three plays later, a Beavers interception was returned to the 2-yard line to set up the game-icing score, and the rest is Corvallis history.

USC's average starting field position in the game was its own 24-yard line. Oregon State started on average from its own 30-yard line, excluding the final goal line field position; including it, they started on their own 36 on average. USC's 21 points were scored on its three shortest fields of the night. All four of Oregon State's touchdowns were scored on its shortest fields of the night as well. Good field position in general and a decisive advantage in field position margin was crucial in this victory for Oregon State. The running game was critical in moving the chains to create that field position advantage, but pinning USC deep with the game on the line with the those punts was possibly even more important.

Figure 1: Offensive Efficiency by Team Field Position
Figure 2: Opponent's Next Possession Efficiency by Team Field Position

On average, teams score on fewer than 20 percent of their drives begun inside their own 15-yard line, as illustrated in Figure 1. The difference in score expectation between a drive beginning from a team's own 5-yard line and its own 20-yard line is nearly ten percent. And that's just the effect on the offense. Figure 2 represents the score expectations of the opponent's next possession after a team's own drive, by field position start. Starting a drive deep in one's own territory is the only area of the field that significantly impacts the average success of the opponent's next possession. A punt downed inside the opponent's 5-yard line is as valuable offensively as it is defensively, if not more so.

If those punts had trickled into the end zone, might USC have been in position to turn the game from an upset into a victory? And what if either punt had been fielded and returned to an even more palatable position, like USC's final scoring drive that began in Oregon State territory after a too-little, too-late big kickoff return? Obviously, the game was ultimately won in the trenches, and OSU deserves all the credit for manhandling the line of scrimmage to give Jacquizz Rodgers the room he needed to work his magic. But it's important to mention where on the field that line of scrimmage was located throughout the night.

Posted by: Brian Fremeau on 01 Oct 2008

7 comments, Last at 02 Oct 2008, 1:14pm by MSC

Comments

1
by Parker W. (not verified) :: Thu, 10/02/2008 - 12:33am

Roll Tide

3
by pawnking (not verified) :: Thu, 10/02/2008 - 8:18am

Indeed, roll Tide. Bama was a little lucky to be up so much at halftime, but not lucky to win the game. This is a really good team, and I can't help but think the FEI is underrating them a bit.

2
by Justin (not verified) :: Thu, 10/02/2008 - 6:31am

Being a Hokie, I love that Virginia Tech is #4. Having watched the games, however, I have trouble believing that ranking.

4
by Rocco :: Thu, 10/02/2008 - 8:40am

Being a Hokie fan, I feel the same way. Right now, this team just isn't that good. I don't trust the defense at all- they let Nebraska back into the game far too easily. The passing game, while less of a disaster last week, is still a mess. By the end of the year they may end up putting it all together and being a top-15 team, but they aren't there yet.

5
by Anonymous (not verified) :: Thu, 10/02/2008 - 11:11am

Does the figure 1 graph match better on a linear or logistic regression? It's hard to tell by eyeballing.

6
by Spoilt Victoria... :: Thu, 10/02/2008 - 12:54pm

I would like to express how surreal this season has been for a Vandy fan, but I don't think I'm capable of doing so. If we somehow beat Auburn this weekend, I will know for certain that I am actually living in the Truman Show. The fact that there is a statistical measure, however short of data, that suggests that we are actually underrated just makes it even stranger.

College GameDay is going to be filmed here this weekend. What planet is this, again?

7
by MSC (not verified) :: Thu, 10/02/2008 - 1:14pm

I wish the FEI chart had a column for last week's rating because you are talking about teams moving up or down the rankings, but unless you go back to last week's column you can't see the effect on the ranking. Also it would be interesting to have a column with the BCS ranking in it to see how FEI compares to that.