Writers of Pro Football Prospectus 2008

Most Recent FO Features


» 2018 Free Agency Cost-Benefit Analysis

Is Kirk Cousins the best free-agent quarterback in recent memory? Should Trumaine Johnson or Malcolm Butler have gotten the larger contract? And what makes a free-agent contract good or bad, anyway?

15 Nov 2007

Fremeau Efficiency Index: Week 11

by Brian Fremeau

The Fremeau Efficiency Index (FEI) 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 (Division 1-A) teams are considered. The Week 11 Ratings represent the results of games played through Saturday, November 10, 2007.

Rank Team W-L FEI Last Wk GE GE Rank
1 LSU (9-1) 0.289 1 0.261 7
2 Oregon (8-1) 0.276 2 0.247 9
3 West Virginia (8-1) 0.237 3 0.330 1
4 Ohio State (9-1) 0.230 4 0.306 4
5 Florida (6-3) 0.227 7 0.151 16
6 Cincinnati (7-2) 0.215 16 0.184 13
7 Clemson (7-2) 0.208 13 0.249 8
8 Arizona State (9-1) 0.206 9 0.202 11
9 Oklahoma (9-1) 0.205 6 0.319 3
10 South Florida (6-3) 0.201 5 0.137 20
11 Georgia (7-2) 0.194 18 0.092 29
12 USC (8-2) 0.190 21 0.201 12
Rank Team W-L FEI Last Wk GE GE Rank
13 Missouri (8-1) 0.181 14 0.217 10
14 Virginia Tech (7-2) 0.180 19 0.092 28
15 Illinois (7-3) 0.172 29 0.118 22
16 California (6-4) 0.165 20 0.061 39
17 Boston College (7-2) 0.160 10 0.139 19
18 Connecticut (7-2) 0.156 11 0.106 26
19 Tennessee (7-3) 0.153 30 0.081 31
20 Alabama (5-4) 0.153 15 0.055 41
21 BYU (6-2) 0.150 24 0.160 15
22 Florida State (6-4) 0.143 12 0.063 38
23 Auburn (6-4) 0.143 8 0.077 34
24 Virginia (9-2) 0.141 27 0.088 30
25 Kansas (9-0) 0.140 23 0.321 2

Offensive Efficiency (OE) represents the raw per-possession scoring rate for each team, and Defensive Efficiency (DE) represents that of its opponents. Adjusted Offensive Efficiency (AOE) and Adjusted Defensive Efficiency modifies OE and DE for field position and opponent. Field Position Advantage (FPA) represents the difference between a team's average offensive field position and that of its opponents.

FEI Rank Team W-L OE OE Rank AOE AOE Rank DE DE Rank ADE ADE Rank FPA
1 LSU (9-1) 0.419 11 0.433 8 0.164 6 0.164 4 3.7
2 Oregon (8-1) 0.478 4 0.516 2 0.240 32 0.232 27 4.9
3 West Virginia (8-1) 0.448 7 0.429 12 0.191 14 0.194 15 5.6
4 Ohio State (9-1) 0.388 16 0.348 32 0.099 1 0.114 1 9.8
5 Florida (6-3) 0.508 1 0.550 1 0.360 94 0.319 82 3.9
6 Cincinnati (7-2) 0.342 31 0.373 25 0.170 7 0.186 12 6.0
7 Clemson (7-2) 0.399 14 0.405 13 0.155 4 0.185 11 5.3
8 Arizona State (9-1) 0.340 32 0.369 26 0.184 12 0.183 9 1.9
9 Oklahoma (9-1) 0.469 5 0.445 7 0.180 11 0.178 6 1.4
10 South Florida (6-3) 0.314 47 0.335 38 0.192 15 0.181 8 5.8
11 Georgia (7-2) 0.381 21 0.432 10 0.268 44 0.238 33 1.1
12 USC (8-2) 0.339 33 0.315 44 0.155 3 0.152 3 4.1
FEI Rank Team W-L OE OE Rank AOE AOE Rank DE DE Rank ADE ADE Rank FPA
13 Missouri (8-1) 0.440 9 0.487 3 0.260 38 0.232 28 -1.0
14 Virginia Tech (7-2) 0.246 82 0.261 74 0.141 2 0.148 2 4.9
15 Illinois (7-3) 0.328 41 0.401 15 0.227 25 0.216 19 -2.2
16 California (6-4) 0.328 40 0.381 21 0.280 47 0.267 48 -0.2
17 Boston College (7-2) 0.318 46 0.389 18 0.174 10 0.232 26 -0.3
18 Connecticut (7-2) 0.263 71 0.279 63 0.187 13 0.194 14 1.5
19 Tennessee (7-3) 0.385 18 0.394 17 0.291 56 0.264 47 4.0
20 Alabama (5-4) 0.285 61 0.326 40 0.260 37 0.226 22 1.8
21 BYU (6-2) 0.335 34 0.357 30 0.205 20 0.222 20 -1.8
22 Florida State (6-4) 0.233 94 0.266 70 0.205 21 0.222 21 4.7
23 Auburn (6-4) 0.284 63 0.316 43 0.221 24 0.184 10 -0.3
24 Virginia (9-2) 0.270 67 0.298 53 0.210 22 0.252 39 1.4
25 Kansas (9-0) 0.444 8 0.361 28 0.161 5 0.170 5 4.8

Click here for rankings of all 119 teams. Click here for Offensive/Defensive Efficiency rankings for all 119 teams.

Oregon was idle, LSU may as well have been, West Virginia escaped a scare, and Ohio State fell for the first time. But at the end of another wild weekend (more on this later), the FEI top four teams managed to hold their positions. And for the first time in a long time, the FEI top two match the polls, the BCS, the popular and electoral vote, Siskel & Ebert, four out of five dentists and pretty much any straw poll taken outside of the Plains states.

In the most important game of the weekend, Illinois managed to defeat Ohio State at its own game, protecting the football and sustaining drives, none more significant than the eight-minute game-ending possession. Ohio State only managed to run 11 total plays over the final 18:46 of the game, and threw interceptions on two of their three total second half possessions. The Buckeyes' loss has paved the way for a possible, if not yet probable, national championship showdown between the best team from each of the best conferences, LSU and Oregon.

The big three from the Big 12 -- Oklahoma, Missouri and Kansas -- may have something to say about the BCS before the end of the season, but they shouldn't get that chance. The only significant out-of-conference win by any team in the Big 12 remains Missouri's Week 1 victory over Illinois. The second tier of the conference consists of a Texas team racking up narrow escapes every week and a sometimes-explosive Texas Tech team that has fizzled against quality teams. And the third and fourth tiers are a mess, populated by a bunch of near-.500 teams (or worse) that have suffered a number of embarrassing and inexplicable losses. The good news for Kansas is that they may get the chance to play their way up and into the championship picture. FEI doesn't expect them to, though.

The Big East, meanwhile, has flown under the radar once again, mostly due to the failure of Louisville and Rutgers to live up to contender expectations. The Mountaineers will play Cincinnati and Connecticut in the next two weeks, games that don't have the marquee value of last season's Big East homestretch, but contests that will ultimately determine the conference championship. And if anyone bothers to pay attention, this week's WVU vs. Cincinnati game has the most potential, pitting the Mountaineers' potent offense against a Cincinnati defense that has held five opponents to 14 points or less and all but two opponents under their season scoring average.


Florida (6-3; BCS No. 12; AP No. 14; FEI No. 5)
The Gators are the first SEC team to close up conference play, and probably mercifully so. Three disappointing conference losses have squashed their chances at defending not only last season's national title, but the conference crown as well. Florida dropped games in the final moments to the top two teams from the SEC West (LSU and Auburn), and, more damagingly, stumbled against rival Georgia in Jacksonville. In spite of all that, the Gators offense has been outstanding, averaging just over a touchdown scored for every two possessions on the season. They are second only to Oregon in Adjusted Offensive Efficiency in long-field drives, and are the only FBS team to rate in the top 10 AOE from every area of the field. Florida Atlantic and Florida State remain on the regular season schedule, and both should be appetizers before they feast on an overmatched bowl opponent.


Virginia (9-2; BCS No. 14; AP No. 16; FEI No. 24)
Before Saturday, the Cavaliers had been living on the edge in nearly every game this season. Even at 8-2, they had actually held a lead on their opponents in less than half of their possessions on the year, and held a multiple-score lead on only 21 percent of their possessions. The floodgates finally opened for Virginia in a 48-0 thrashing of reeling Miami, spurred by six first-half offensive drives begun on the Hurricane side of the field. Despite that effort, the Cavaliers have still over-performed their Pythagorean Winning Percentage to date by 1.5 games, the fifth-largest margin among FBS teams. UVA will face an improving Virginia Tech team in their final regular season test next week for a spot in the ACC championship game.

Wild Weekend

Navy defeated North Texas 74-62 on Saturday, setting a new record for total scoring in a single non-overtime major college game. The game featured 60 first downs and 19 touchdowns. A single stretch in the second quarter accounted for eight straight touchdown drives by the teams and 475 yards of offense on just 30 plays, all in less than 12 minutes of game time. There were two three-and-out punts in the game versus nine touchdown drives of three plays or fewer, only two of which started in opponent territory.

That absurd offensive showcase was just part of a remarkably high-scoring weekend in college football. In the 53 games played between FBS opponents, none of which went into overtime, 11 teams scored at least 50 points and 28 teams scored at least 40. The total points scored over the weekend totaled 3,233, the most in a single week since at least the beginning of the 2003 season (the FEI era). Week 11 was the third week this year that topped the 3,000 mark in total points scored. Five of the top nine scoring weekends since 2003 have been recorded this year alone.

Field goal percentage is up more than 2 percent nationally, and the change in the kickoff rule has placed teams only about a yard closer to the end zone at the start of drives. The clock rule changes from last season are contributing more plays and possessions this year, but even accounting for the additional plays, Week 11 was remarkable. Games over the weekend featured a scoring rate of 2.3 points per competitive possession, the second highest rate in the FEI era. (2005's Week 14 tops that list, featuring USC scoring 66 points and Texas scoring 70, upstaging one another in a final showcase before their championship bowl battle.) Whether teams shoot for similar style points down the stretch this season remains to be seen. Another race for the BCS may provide the incentive, but with the increase in offensive production this season, teams may need to run up their score just to stay in the lead.

Posted by: Brian Fremeau on 15 Nov 2007

39 comments, Last at 17 Nov 2007, 9:15pm by Sid


by Sundown (not verified) :: Thu, 11/15/2007 - 3:39am

Can we just officially make the "i" in FEI stand for irrelevant?

The BCS just called and they think this is jacked up.

After seeing Alabama on there at 5-4 I started skimming to see where Notre Dame wound up. I suppose it'll take a narrow loss to Duke to get them into the top 25.

by Temo (not verified) :: Thu, 11/15/2007 - 3:49am

I dunno, the list is starting to look better now than it has in the past.
Besides, I think an Oregon-LSU championship would be fun.

by Fourth (not verified) :: Thu, 11/15/2007 - 4:37am

Seriously Sundown, nearly the whole point of this, like DVOA, is to provide a less subjective look beyond win-loss record when evaluating teams. Why don't you hop on over to the DVOA rankings thread and gripe about Philly's top 12 spot...

by lionsbob (not verified) :: Thu, 11/15/2007 - 4:44am

Alabama's 5-4 (6-4, but not counting the West Carolina win) is on there because..well they had Mississippi State beat except for a 100 yard interception return for a TD. They had LSU beat except for two turnovers that started LSU drives inside the 15.

The index is not an eyeball test, it works with stats alone. Considering Notre Dame is 1-9 and can't score on most high schools I doubt they have much in common with Alabama.

by Gary (not verified) :: Thu, 11/15/2007 - 4:52am

(Alabama) is clearly ranked (too high) because (they are 5-4). (BCS) is way better than this.

by Shawn (not verified) :: Thu, 11/15/2007 - 5:36am

College football teams don't play nearly enough out of conference games for comparisons of conference to conference strength of schedule.

DVOA works because AFC and NFC teams play against each other throughout the year and teams from different divisons continue to play each other. That simply isn't the case in college football. DVOA has a Weighted DVOA which emphasizes the importance of late season results. This can't happen with FEI because late season results essentially don't matter if your team doesn't reside in a conference that accumulated a good strength of schedule based on a handful of games the first couple of weeks of the season.

More importantly, college football is not based on the same system as any other sport. It is essentially a single elimination tournament from the very beginning of the season. That means teams from BCS conferences that lose are essentially eliminated until all other teams from BCS conferences have lost.

If the FEI system were meant to predict future results, it would be interesting. If it were meant as a helpful seeding tool for a national tournament, it would be interesting.

Because it values drive success rates over win/loss - and adds in a very questionable strength of schedule element - it is virtually worthless. That college football doesn't have a playoff is a travesty. But because it doesn't, you can't run a system that pretends that it does.

That Kansas would be ranked 25th after manhandling teams all season is an indication that there is still a lot of work to do here. And for posters who claim this is more objective because it's based on a formula, well, the other computer rankings are based on formulas also and Kansas is ranked 2nd overall on those.

by Fourth (not verified) :: Thu, 11/15/2007 - 5:57am

09/01 C Michigan W 52-7
09/08 SE Louisiana W 62-0
09/15 Toledo W 45-13
09/22 Fla Int'l W 55-3
10/06 at #24 KSU W 30-24
10/13 Baylor W 58-10
10/20 at Colorado W 19-14
10/27 at Texas A&M W 19-11
11/03 Nebraska W 76-39
11/10 at Oklahoma St W 43-28
11/17 Iowa St
11/24 #5 Missouri

And of course Kansas gets the Big 12 title game if they beat Mizzou. I'm not going to pick apart every game, the schedule speaks for itself. The best game there is probably a 5 point win at Colorado, or maybe the 8 point win at A&M.

Kansas probably isn't the 25th best team in the nation right now. They probably aren't a top 5 team either, but luckily we will get to find out. If they "manhandle" good teams like Mizzou and Oklahoma, instead of beating up on creampuffs and skating by mediocre teams (ok, they blew out one mediocre team--Nebraska), they will earn a chance at the title.

For what it's worth, Kansas would also probably receive a pretty large boost in FEI should they play well and win out. Seeing as right now they are 2nd before schedule adjustments, and that schedule is about to get tough for the first time all year, some more manhandling should put them in the top 10 here too.

By the way, I wonder if there is an FOMBC for college...

by Becephalus (not verified) :: Thu, 11/15/2007 - 7:21am

This is just one method for rating the strength of the teams. I personally like it better than most methods, but who cares about that.

The important thing to be aware of is that no rating system can be all things to all people. Why get so worked up over this? The system doesn't "need" work based on what it does with one team, and the components are not "questionable", they just are what they are, if you prefer different components go look at a different system, or if one doesn't exist create it yourself.

by RickD (not verified) :: Thu, 11/15/2007 - 9:32am

I have to say that I like this. I've been thinking that the Big 12 is way overrated in the national polls. It's refreshing to see Kansas at 25.

by Brooklyn Buckeye (not verified) :: Thu, 11/15/2007 - 12:32pm

The FEI does seem to do a better job at passing the oh-so-important "eye test" as the season goes on. Cincy and Clemson still seem overrated here (maybe somewhere between 15-20?), and Kansas seems underrated (top 10?).

Anyway, it's good to see two Big Ten teams in the top 25...I think Illinois was unfairly neglected all year following a dreadful performance against Iowa. It's too bad they had to get their respect by beating up on my Buckeyes!

by Dennis (not verified) :: Thu, 11/15/2007 - 12:40pm

I like the concept of the FEI, but it seems like something is really off when a team like Kansas can keep winning and keep dropping every week. Yes, they've played a weak schedule, but their schedule isn't any weaker than it was 2 weeks ago and they've fallen 11 spots.

#6 is right that there just aren't enough games between the conferences to do a meaningful analysis of relative strength.

I had an idea on the I-AA school issue. Why not say a I-AA team is equivalent to the #100 I-A team (or some other arbitrary rank) and use that to factor in the games against I-AA teams. That way teams that lost to I-AA teams get penalized and teams that beat I-AA teams don't get a whole lot of credit.

by pete (not verified) :: Thu, 11/15/2007 - 12:51pm

any system which says kansas is the 25th best team in the country needs to be looked at further.

Meanwhile, LSU has played 4 down-to-the wire games and is no. 1.

When it comes to impressive wins, I can think of only one for LSU. Theyre big wins over FLA and AUB came at home, and each came in the closing seconds. LSU destroyed Va Tech, albeit at home. Meanwhile, Oregon went into Ann Arbor and absolutely embarrassed Michigan.

That said, i love the effort. Its just that the lack of parity in college football makes FEI difficult to use as a predictive model.

by Trevor (not verified) :: Thu, 11/15/2007 - 1:08pm

for shits and giggles...the potential BCS matchups if this was the ranking. No bowl is picking a 6-3 USF in the BCS game, so i swapped them with GA. Pretty good games...

LSU v. Oregon - NC
West Va. v. Georgia - Sugar
Ohio State v. Arizona State - Rose
Clemson v. Florida - Orange
Oklahoma v. Cincy - Fiesta

by Trevor (not verified) :: Thu, 11/15/2007 - 1:11pm

is kansas really better than Hawaii (both are undefeated)? I know they're in the Big12, but the conference foes they've faced could easily be good WAC teams. I haven't seen them so I don't know, but they're conference affliation shouldn't help them so far since they haven't played UT, Okla, or Mizzou yet.

by bartleby (not verified) :: Thu, 11/15/2007 - 1:15pm

Gene Siskel is dead.

by AHBM (not verified) :: Thu, 11/15/2007 - 1:25pm

Expanding on #15, you meant to say "Ebert & Roeper", Brian.

by Richard (not verified) :: Thu, 11/15/2007 - 2:06pm


Seriously, can we just make the first comment in this thread every week some trite, recycled reference to this index's relevancy.

Damn. At least be original. Or, even better ...

Read the methodology and develop a critique. Or, track the FEI rankings with reference to how predictive they are of future (or even past) performance. Something! ANYTHING!


by DangerGnat (not verified) :: Thu, 11/15/2007 - 2:16pm

I've been following this ranking every week, and each week I become more baffled. How does Oklahoma rank 7 in AOE and 6 in ADE, yet only 9th overall? They are behind WVU, Cincy, Clemson, and ASU, who all have lower ranks in both AOE and ADE. Perhaps I am not understanding the entire ranking process, but this just doesn't seem right. Kansas also seems very low overall when you look at their AOE/ADE rankings.

by footballprofessor (not verified) :: Thu, 11/15/2007 - 2:17pm

Two words..."analysis", "paralysis".

Don't worry Brian, I got the Siskel and Ebert joke. Don't worry about the numbskulls that are so into analyzing the boogers out of everything that they have to call you out on it.

To those of you who want Kansas a little higher than 25th - go find a rating system that doesn't take into account strength of schedule. Mine doesn't, so you could start there where Kansas is ranked #1. Brian's system is great, it analyzes the game from several interesting angles, and we should appreciate it for what it is - a DVOA like attempt to analyze the college game. Not perfect, but in my opinion better than the BCS.

by strannix (not verified) :: Thu, 11/15/2007 - 2:20pm

I'm wondering if there's any serious analytical argument to be made that Tim Tebow isn't the best player in America. Anyone? Or is this already generally accepted?

by Yinka Double Dare (not verified) :: Thu, 11/15/2007 - 2:35pm

20: Who knows -- no one else has the entire offense run through him on almost every play, because most other quarterbacks have running backs that they hand it to at least 20 times a game, instead of also being the primary running back. Not even McFadden has to carry his team's entire load, he at least has Felix Jones to help out a pretty fair amount. Tebow's only 'running backs' that Meyer trusts are the wide receivers.

No one can take that kind of punishment though over the long haul, not even Tebow. He's just fortunate that the injury that finally happened was on his off-shoulder and so he was still able to play. They better hope next year that Moody doesn't end up in the Kestahn Moore Doghouse, because I don't think Florida can win a title with Tebow having to be the primary running back. 25 carries a game plus all those passing attempts will take their toll, he'll get hurt again, and I suspect that next time he will not be so fortunate and it will be something that keeps him out of a few games.

Tebow is basically the perfect QB for that offense though. I don't know if he's the best player, but he has to be in the discussion. I am definitely sick of all the Tebow ball-washing by commentators though. It's to the point that you can't watch a Florida game anymore unless you mute it -- they're giving him the Favre treatment already, and he's only a sophomore.

by strannix (not verified) :: Thu, 11/15/2007 - 2:59pm

I guess what I mean isn't so much "best", which is overly broad, as "the most valuable", which is more of an analytical question. Because so much of the offense runs through him, I'd have to think that his college-football-DVOA-equivalent would be absolutely through the roof given how productive the offense has been.

by DavidH (not verified) :: Thu, 11/15/2007 - 3:35pm


Very good question. Only explanation I can come up with is that it comes down the field position difference, but that doesn't make sense to me. It seems that the poor field position will already be factored in by virtue of them scoring less points or giving up more points.

For another perspective, here is the top 25 if you rank all teams by [AOE-ADE]:

1 Oregon 0.28364
2 LSU 0.26872
3 Oklahoma 0.26732
4 Missouri 0.25467
5 Ohio State 0.23455
6 West Virginia 0.2342
7 Florida 0.23153
8 Clemson 0.21973
9 Georgia 0.19429
10 Kansas 0.19082
11 Cincinnati 0.18696
12 Arizona State 0.18567
13 Illinois 0.18485
14 USC 0.1626
15 Boston College 0.15773
16 Arkansas 0.15313
17 South Florida 0.15309
18 Texas 0.14721
19 Texas Tech 0.14156
20 Louisville 0.13795
21 BYU 0.13454
22 Oklahoma State 0.13318
23 Auburn 0.13242
24 Tennessee 0.13068
25 Michigan 0.1263

Somehow the Big 12 comes out looking like a powerhouse. Another way could be to do a [AOE/ADE], which produces the same order as a Pythagorean formula:

1 Ohio State 3.06524610372458
2 LSU 2.6375380865326
3 Oklahoma 2.50348706411699
4 Oregon 2.22142795624839
5 West Virginia 2.20423693953106
6 Clemson 2.18785814682668
7 Kansas 2.12412371134021
8 Missouri 2.09634508588402
9 USC 2.06622950819672
10 Arizona State 2.01459016393443
11 Cincinnati 2.00397379443669
12 Illinois 1.85424465086187
13 South Florida 1.84388953199934
14 Georgia 1.81696240854428
15 Virginia Tech 1.76528266161753
16 Florida 1.72650538140513
17 Auburn 1.72108473099543
18 Michigan 1.6964817469946
19 Boston College 1.68130966264956
20 Texas 1.62832387212429
21 Arkansas 1.62583782900114
22 BYU 1.6057358966278
23 Penn State 1.54077133707709
24 Tennessee 1.49588282169013
25 Texas Tech 1.45221058011756

Doesn't emphasize the Big 12 quite as much.

Either way, Oklahoma and Kansas are way underrated by FEI in comparison.

by DavidH (not verified) :: Thu, 11/15/2007 - 3:36pm

Sorry 'bout the excessive post-decimal length

by Dennis (not verified) :: Thu, 11/15/2007 - 4:03pm

#12 Its just that the lack of parity in college football makes FEI difficult to use as a predictive model.

I totally disagree about the amount of parity - I think there is more parity in college football this year than there ever has been. That's why there aren't any clear standout teams.

by Dennis (not verified) :: Thu, 11/15/2007 - 4:06pm

#19 - I'm not saying Kansas should necessarily be ranked higher. I'm saying it doesn't make sense that they have dropped so much in the last two weeks when their strength of schedule hasn't changed.

by DangerGnat (not verified) :: Thu, 11/15/2007 - 5:09pm

Thanks for the response. I think both of your rankings make more sense, from a "numbers" view as well as my subjective view of how good the teams are. I guess I will have to dig into the long-winded explanation of the formula if I really want to find the answer to my question. AOE/ADE are supposed to be adjusted for opponent strength and field position, so I'm missing how Okla. can be ranked lower overall than both their AOE and ADE.

by Fourth (not verified) :: Thu, 11/15/2007 - 5:59pm

Re: 13
I believe it's against BCS rules to have 3 teams from the same conference all get bids. At least, that's how it was before they added the 5th game, I suppose that may have changed and I just didn't hear about it.

Re: 27, etc:
I think this is the explanation you are looking for how the Adjusted data changes, though it's not going to be very satisfying:

"The Fremeau Efficiency Index (FEI) weights the value of Adjusted Game Efficiency data by first evaluating the following criteria:

1. Who did you beat and how did you win those games?

2. Who did you lose to and how did you lose those games?

As the quality of the opponent decreases, the value of the first question receives less weight than the second.

FEI rewards teams for playing well against good teams, win or lose, and treats losing to poor teams more harshly than it rewards victories over poor teams."

So it sounds like first each game is broken down into drives then compared against how the opponent fared in other games. But then there is another sort of strength of win/weakness of loss component that goes into it (which turns AOE-ADE into FEI), and since Kansas's "strength of win" value hasn't changed much the last 2 weeks (as in it's still pathetic) they are dropping in FEI as other teams move past them. Is this fair? I'm not sure, but it might be a place to start in reviewing the numbers at season's end when we know a little bit more about Kansas. If Kansas proves the system wrong, this part probably needs to be tweaked.

by citizen jason (not verified) :: Thu, 11/15/2007 - 6:17pm

28: Good point about Kansas. Really, I don't know why people seem to be so up in arms about it, since there's no really good way to rank Kansas, I don't think. (Especially if you are using a stats-based method.) As was pointed out above, yes they are undefeated, but: they haven't beaten (or even played) a good team (in fact, Florida International has lost 21 straight ...); and they play in a conference where (this year) the only good win is against a 3-loss Big Ten team.

I like to think of it this way: how many teams below them would be undefeated playing that same schedule? My gut feeling is quite a few ...

by Kal (not verified) :: Thu, 11/15/2007 - 6:20pm

#26: it's not that Kansas has dropped so much (though they have; basically, the teams that they've beat have actually gone out and sucked even more), it's that the teams at the top have played other teams at the top and gained more traction and looked better.

The thing is, it'll all be dealt with one way or another at the end of the season. If Kansas is better than #25, it'll show up against Missou (and possibly Oklahoma). If they're about where FEI says they are, they'll lose.

I still think that FEI puts too much emphasis on strength of schedule for an overall ranking and too little emphasis on recent game performance for a predictive quality.

by The Boilermaster (not verified) :: Thu, 11/15/2007 - 7:26pm

I, for one, maintain that the FEI puts thrice as much emphasis on strength of schedule as it should.

by Todd M. (not verified) :: Thu, 11/15/2007 - 7:58pm

Cincy is the real deal. West Virginia is going down and the Bearcats are BCS bound.

by Pete (not verified) :: Fri, 11/16/2007 - 12:16pm

I wonder what Oregon's loss would look like , especially comparing with Dixon and without him. I wish him the best and hope he makes a healthy recovery. From what I know the interception and 1 of his other 2 incompletes (5 of 8 passing) were drops or at least considered to be the receivers' error.

What does this mean for the Heisman? Do the voters consider poor team performance without him as recognition of superior play? Or is this just a situation where his stats may not be as large as for Tebow or others? I am a Gator fan, but I think Tebow could improve his delivery (for such a strong guy his passes sometimes do not get there as quickly as hoped). The commentators praised his long passes recently, but I think his receiving corps is among the best and adjust a little to the ball in the air. Of course, there have been several dropped passes in many of the games, too, but that is part of the game.

How does Tebow project at the next level, or is it too soon to tell? He is about 6-3 and 235 pounds and benches over 400 lbs. Could he be capable at the next level at QB? Or does he project as an H-Back or Fullback or special teams?

by SoonerHQ (not verified) :: Fri, 11/16/2007 - 1:16pm

Brian wrote: "The only significant out-of-conference win by any team in the Big 12 remains Missouri’s Week 1 victory over Illinois."

OK, other than LSU's victory over Va Tech, what quality OOC wins do the SEC teams have?

by heath (not verified) :: Fri, 11/16/2007 - 2:12pm

How do your little rankings look now retards? Oregon gets beat by Arizona. Ha. And what big wins has Ohio St had??? NONE. How can you even consider having Alabama in the top 25? This poll is a joke. Maybe you should pick up a different hobby, like maybe ranking the top basket weavers in the nations or something.

by Tampa Bay Mike (not verified) :: Fri, 11/16/2007 - 3:51pm

Wow, Heath's post is so dumb that I can't tell whether or not it's intentionally so. Upon further review however, the supposed invalidation of any statistical model by Arizona beating Oregon (under the circumstances) is so foolish that I think he must be joking.

by heath (not verified) :: Fri, 11/16/2007 - 6:43pm

You are right, I expected Arizona to at least cover the spread versus, Oregon, but not beat them. And any poll can't take into account a star qb getting hurt in the first quarter, (with a Leaf as the backup). However, I think this poll itself is outrageous and the comments regarding the Big XII ridiculous. Alabama,Florida St, & South Florida have no business being in the top 25 teams. So when Michigan beats Ohio St, is Ohio St going to remain at #4? Personally, I think that Kansas may get upset this weekend, and is a threat to get upset every weekend. BUT, so is USC, LSU, and many others. To discredit a conference when they have 3 teams in the top 5 is insane. When bowl season comes around, we will see who is the best conference and who deserves their rankings.

by Joe, channeling Jim Rome (not verified) :: Fri, 11/16/2007 - 7:44pm

Memo to all the would-be statisticians out there: The A in DVOA doesn't stand for "Aaron." Putting your unknown name on your little models is fun and all, but it doesn't add any credibility. If your idea catches on, you'll become known as Aaron has. If not, you'll be just as anonymous with your name on the system as not...which would be completely.

But there is a waiver for anyone whose mom agrees to waive rent for the basement if you can prove you finally did something with your life. In cases like that, feel free to slap your name on your system as many times as you can. The Fremeau Efficiency Index: Presented by Fremeau, for Fremeau, and featuring Fremeau (FEIPBFFFAFF) has a nice ring to it and may include the most Fs as any acronym in history, which might get you another 6-pack of generic soda the next time mom goes shopping.

Good Night Now

by Sid (not verified) :: Sat, 11/17/2007 - 9:15pm

ok, Duke just got blasted by Notre Dame. Now do you understand why I argued last week that they didn't deserve their ranking? They're one of the worst teams in the country.