Writers of Pro Football Prospectus 2008

Most Recent FO Features

McKinnonJer14.jpg

» Week 7 Quick Reads

Did Jerick McKinnon prove against Buffalo that he can be a feature back for Minnesota? Plus the best passers, runners, and receivers of Week 7.

27 Oct 2010

FEI: Everybody Punts

by Brian Fremeau

Auburn is the new No. 1 team in the country according to FEI, and Cam Newton's Heisman candidacy surged forward in a relentless ground attack on Saturday. As I discussed last week in an ESPN Insider piece, the game was supposed to feature a Top 10 offense in Auburn against a Top 10 defense in LSU. After surrendering 440 yards on the ground, however, LSU's defensive FEI ranking dropped from No. 3 to No. 21, and Newton's 28 carries and 7.8 yards per attempt deserved most of the credit.

Auburn more than doubled LSU's total yardage in the game but wasn't able to bury LSU on the scoreboard. The credit for that belongs to LSU's extraordinary field position advantage in the game -- .622 FPA, or a 62 percent share of the total starting field position value in the game. Based on starting field position alone, an average offense against an average defense would have expected to score 20.7 points on LSU's drives and only 12.6 points on Auburn's drives. Auburn's FPA was the fifth most severe disadvantage overcome by a winning team this season. And unlike most starkly imbalanced field position games, no defensive or special teams touchdowns factored into the outcome. LSU kept Auburn's prolific offense from producing a more prolific scoring margin with a superstar performance from its punting team.

One thing I've been meaning to do for a while here in the FEI column is break down the FPA numbers into smaller components. The starting field position for any given drive may came about due to a variety of factors. First, since football is played in alternating possessions and every yard line on the field has at least some level of offensive score expectation value, part of the starting field position value is attributed automatically and isn't earned by a team's offense, defense, or special teams. But the rest of the value comes from the combination of the offensive, defensive, and special teams performances of the preceding possession. In the coming weeks, we'll take a look at those components one by one. This week, let's isolate on the field-position exchange on a punt.

The NCAA maintains a ranking of FBS "Net Punting" leaders, one way to measure a punt team's effectiveness. By their methodology, the total number of return yards and touchback yards is subtracted from the total number of gross punting yards, and the result is divided by the number of punts. Another idea proposed by Bud Elliott of TomahawkNation.com last week suggested that punting ought to be measured by percentage of available field to be gained. For example, a punt from midfield that results in a touchback nets 60 percent of the available yardage (30/50 = .60); one that is downed at the 5-yard line nets 90 percent of the available yardage (45/50 = .90). It's a good idea for punts in which a touchback is in play, but it doesn't translate well for punts from deep inside a team's own side of the field.

I was interested in identifying the national average expectations for a punting team in terms of field-position value and then measuring each punt for each team against that value. I first collected the data from every non-garbage time punt in the last three years, 19,862 punts in all. From each drive-ending punt position, I then calculated the average starting field position for the opponent's next drive. For instance, the 403 punts from midfield in the last three years netted an average starting field position for the opponent at its own 18-yard line. The 413 punts from a team's own 20-yard line in the last three years netted an average starting field position for the opponent at its own 41-yard line.

Every yard line translates into an expected field-position value. An average offense against an average defense is expected to score 1.2 points per possession from its own 18-yard line and 2.2 points per possession from its own 41-yard line. In the same manner that offensive efficiency is calculated, Punting Efficiency (PE) is simply the actual value of post-punt field position divided by the expected value of post-punt field position, minus one (in order to represent a team's efficiency above or below average). A good punt team will have a PE below zero, pinning opponents with less starting field position value than expected. A good punt return team will have a PE above zero, creating more starting field position value than expected.

It is important to note that this metric should be attributed to the respective punt unit and punt return unit, not just the punter or punt returner involved. It does not take into account the hang time or roll of the punt, the coverage, etc. Rather, the resulting field position of the entire play is taken into account. Punts returned for a touchdown are credited with a field position value of 6.96 points. Punts that are fumbled are credited with the field position value of the recovered fumble. PE does not include any adjustment for the strength of opposition faced.

Against Auburn on Saturday, LSU punted six times, none of which were returned by Auburn. From LSU's punting position, Auburn should have expected an average starting field position on the subsequent six drives of approximately its own 20-yard line. Instead, two of the punts were downed at the 1-yard line, one at the 2-yard line, one at the 9-yard line, one at the 10-yard line and one trickled into the end zone for a touchback. Auburn's actual field position value following those six punts was 35 percent below its expected value (-.346 PE), the eighth-best single game punting effort of the college football season. The tables below list the Top 10 punting and punt return teams according to year-to-date PE.

Punt Efficiency Top-10 Punt Return Efficiency Top-10
Rank Team PE Punts Rank Team PE Punts
1 Oregon -.192 19 1 Utah .528 41
2 Middle Tennessee -.136 37 2 Maryland .340 27
3 Missouri -.135 28 3 Oregon .336 29
4 Michigan -.132 22 4 UNLV .335 22
5 Florida State -.129 22 5 Michigan State .270 37
6 Nebraska -.116 20 6 Northern Illinois .260 27
7 LSU -.109 30 7 Texas Tech .248 35
8 UTEP -.098 23 8 Boise State .243 28
9 Alabama -.097 18 9 Kent State .237 35
10 South Carolina -.095 19 10 Fresno State .233 33

The Oregon Ducks are the only team ranked in the Top 10 for both punting and punt returning according to PE. LSU (No. 7 in punting PE, No. 11 in punt returning PE) is the only other team ranked in the top 20 in both categories. The Utah Utes rank first in punt returning PE and dead last in punting PE. They also have only 16 non-garbage time punts this year, fourth fewest in the country.

Three and Out

Week 3: Three-and-outs, Available Yards, and Explosive Drives
Week 4: Reaching the Red Zone, Methodical Drives, and Late and Close Efficiency
Week 5: Converting 10+ Yard Drives Into Scores, Points Per Possession, and Scoring After Three-and-outs
Week 6: Yards Per TD Drive, Playing With 2- or 3-Score Lead, Third Downs Per First Down Series
Week 7: FEI Team Resumes for South Carolina, Arizona State, and Utah

In the tables below, the Game Efficiency, Offensive FEI, Defensive FEI, and "Game" FEI (GFEI) for each team in each game is provided. The ranking of those individual unit and game performances is provided. Note that there have been 395 FBS vs. FBS game played to date, meaning that there have been 790 individual game performances for each category.

The opponent FEI rating is also provided, as well as a general relevance factor for that particular result for that team. As stated in the FEI principles, my system rewards playing well against good teams, win or lose, and punishes losing to poor teams more harshly than it rewards defeating poor teams. In the formula, the relevance factor is partly a function of the relative ratings of the two teams. The least relevant results receive about one-eighth as much weight as the most relevant results. For simplicity, I've generalized the relevance data here into three equally distributed categories, High, Med, and Low.

No. 6 Oregon Ducks (6-0)
Date Wk Opponent Result Opp
FEI
GE GE
Rk
OFEI OFEI
Rk
DFEI DFEI
Rk
GFEI GFEI
Rk
Relevance
9/4 1 New Mexico W 72-0 118 .803 6 .760 173 .191 393 .197 207 Low
9/11 2 at Tennessee W 48-13 88 .333 105 .336 341 -.182 209 .183 218 Low
9/25 4 at Arizona State W 42-31 41 .081 288 .486 270 .051 322 .292 133 Med
10/2 5 Stanford W 52-31 14 .148 235 1.238 56 -.345 155 .500 27 High
10/9 6 at Washington State W 43-23 100 .260 159 .644 216 .271 435 .044 342 Low
10/21 8 UCLA W 60-13 80 .672 14 1.536 21 -.074 257 .451 57 Low

Oregon followed up a seemingly commanding No. 1 FEI rating last week with a thrashing of UCLA on Thursday night. So what's with the tumble down the FEI ratings? It actually has nothing to do with the victory over the Bruins, which was impressive according to drive efficiency data. Arizona State (previously ranked No. 13 but stuck on only one FBS victory) took a nosedive after flailing against California, and the Sun Devils are the second-best team Oregon has faced. A good win against USC should bring Oregon back in good graces with FEI, but this week's ratings indicate USC may have the upper hand at home.

No. 16 Wisconsin Badgers (6-1)
Date Wk Opponent Result Opp
FEI
GE GE
Rk
OFEI OFEI
Rk
DFEI DFEI
Rk
GFEI GFEI
Rk
Relevance
9/4 1 at UNLV W 41-21 113 .248 167 .175 415 .813 659 -.177 569 Low
9/11 2 San Jose State W 27-14 115 .272 145 -.729 711 .510 551 -.316 684 Low
9/18 3 Arizona State W 20-19 41 .016 381 .892 125 .509 550 .083 306 Med
10/2 5 at Michigan State L 24-34 10 -.130 541 1.257 51 -.158 218 .377 82 High
10/9 6 Minnesota W 41-23 84 .495 49 1.089 81 -.288 170 .251 161 Low
10/16 7 Ohio State W 31-18 17 .232 178 1.818 6 -.598 86 .563 14 High
10/23 8 at Iowa W 31-30 19 .016 382 1.798 7 .327 468 .451 56 High

Wisconsin has notched its two most impressive wins of the year in back-to-back weeks, topping Ohio State in Madison and Iowa on the road. The Badgers were ranked No. 32 in FEI only two weeks ago, and they were languishing in the mid-40s before that with a less-than-impressive schedule to start the year. It's possible Michigan State won't give up its seat at the top of the Big Ten standings, but Wisconsin appears poised to fill the void if they do.

No. 29 Georgia Bulldogs (4-4)
Date Wk Opponent Result Opp
FEI
GE GE
Rk
OFEI OFEI
Rk
DFEI DFEI
Rk
GFEI GFEI
Rk
Relevance
9/4 1 Louisiana Lafayette W 55-7 106 .532 38 .350 334 -.238 190 .104 282 Low
9/11 2 at South Carolina L 6-17 3 -.185 576 .130 434 -1.095 14 .382 81 Med
9/18 3 Arkansas L 24-31 23 -.083 506 .742 177 -.740 56 .119 271 High
9/25 4 at Mississippi State L 12-24 26 -.321 679 .463 280 .321 465 .013 380 High
10/2 5 at Colorado L 27-29 63 -.026 419 .699 195 .913 682 .040 344 High
10/9 6 Tennessee W 41-14 88 .429 72 .491 267 .648 607 .133 254 Low
10/16 7 Vanderbilt W 43-0 73 .723 12 1.623 17 -.408 137 .524 20 Med
10/23 8 at Kentucky W 44-31 31 .155 231 .141 427 -.155 219 .452 55 High

Being the highest-rated four-loss team doesn't exactly meet even minimal expectations at programs with Georgia's history, but the Bulldogs haven't completely embarrassed themselves this year. Every loss has been competitive, the last three weeks have been dominant, and Georgia ranks in the top 20 in season-long Game Efficiency, a remarkable stat for a .500 team. A win over Florida this weekend keeps an SEC East title within reach.

If you have a suggestion for an FEI Team Resume you'd like to see, drop me a line on Twitter or in the comment section here. I'm happy to answer data inquiries or provide team tables for bloggers that are interested.

FEI Week 8 Top 25

The principles of the Fremeau Efficiency Index (FEI) can be found here. 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, not play-by-play 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. 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 to date. SOS listed here does not include future games scheduled.

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 RW) represent the average expected team wins for games scheduled but not yet played.

Offensive FEI (OFEI) and Defensive FEI (DFEI) are the opponent-adjusted ratings of all non-garbage-time drives from scrimmage. Field Position Advantage (FPA) is the share of the value of total starting field position for the season earned by each team against its opponents. Field Goal Efficiency (FGE) is the point value per field goal attempt earned by the field goal unit.

Only games between FBS teams are considered in the FEI calculations. The FEI ratings published here are a function of the results of games played through October 23.

FEI ratings for all 120 FBS teams are listed in the stats page section of FootballOutsiders.com. Click here for current ratings; the pull-down menu in the stats section directs you to 2007 through 2009 ratings. There are also now separate pages for offensive and defensive FEI ratings for 2010.

Rk Team FBS
Rec
FEI LW
Rk
GE GE
Rk
SOS SOS
Rk
FBS
MW
FBS
RW
OFEI OFEI
Rk
DFEI DFEI
Rk
FPA FPA
Rk
FGE FGE
Rk
1 Auburn 8-0 .308 2 .159 17 .224 11 9.4 2.5 .726 1 -.517 6 .524 35 -.015 65
2 Missouri 6-0 .259 12 .218 12 .572 86 9.7 4.3 .500 7 -.597 2 .525 33 .511 21
3 South Carolina 4-2 .251 5 .146 21 .178 3 8.7 4.2 .643 4 -.348 20 .502 62 -.293 89
4 Boise State 6-0 .251 4 .452 1 .518 72 10.9 5.6 .340 18 -.441 10 .536 21 .431 27
5 Nebraska 5-1 .245 10 .265 8 .568 84 9.6 4.3 .345 17 -.520 5 .536 22 1.022 1
6 Oregon 6-0 .239 1 .325 5 .615 94 8.9 3.5 .404 14 -.327 22 .570 5 .629 12
7 TCU 7-0 .231 11 .347 3 .721 105 10.0 3.6 .307 21 -.516 7 .555 9 .620 13
8 LSU 6-1 .227 6 .087 33 .194 8 8.4 3.1 .198 36 -.332 21 .577 2 .674 10
9 Virginia Tech 6-1 .221 9 .274 7 .375 45 8.9 3.1 .481 8 -.251 30 .552 11 .436 26
10 Michigan State 7-0 .220 7 .152 18 .560 81 9.4 3.4 .297 23 -.571 4 .496 69 .603 14
11 USC 5-2 .218 17 .219 11 .445 53 10.1 4.3 .537 6 -.049 52 .567 6 -.949 114
12 Alabama 7-1 .217 16 .300 6 .330 31 8.1 1.5 .473 9 -.296 26 .579 1 .131 51
Rk Team FBS
Rec
FEI LW
Rk
GE GE
Rk
SOS SOS
Rk
FBS
MW
FBS
RW
OFEI OFEI
Rk
DFEI DFEI
Rk
FPA FPA
Rk
FGE FGE
Rk
13 Oklahoma 6-1 .216 8 .135 24 .343 35 9.7 4.2 .415 13 -.422 11 .517 43 -.016 66
14 Stanford 5-1 .215 3 .224 10 .373 43 8.3 3.6 .299 22 -.206 36 .547 12 .596 15
15 Arizona 5-1 .213 21 .185 14 .599 88 8.2 3.1 .241 29 -.413 14 .516 44 .153 48
16 Wisconsin 6-1 .212 15 .149 20 .276 21 9.0 3.7 .643 3 -.193 37 .504 59 .524 20
17 Ohio State 7-1 .204 19 .342 4 .365 40 9.8 3.3 .432 12 -.365 16 .536 20 .075 58
18 Miami 4-2 .201 24 .060 46 .288 24 8.2 4.0 .211 34 -.625 1 .504 58 .160 45
19 Iowa 4-2 .184 20 .189 13 .380 46 8.1 3.7 .249 27 -.325 23 .510 52 -.121 75
20 Oregon State 3-3 .182 14 -.026 71 .142 1 7.1 4.0 .437 10 .201 87 .540 17 -.447 97
21 North Carolina State 4-2 .181 18 .093 31 .507 70 8.1 3.6 .149 40 -.306 25 .530 29 -.060 71
22 Clemson 3-3 .148 25 .074 39 .202 9 6.9 3.2 .209 35 -.443 9 .510 51 -.281 87
23 Arkansas 4-2 .139 34 .033 56 .180 4 6.4 2.9 .684 2 -.230 33 .473 90 .765 7
24 Pittsburgh 3-3 .137 37 .084 34 .476 64 7.4 3.4 .220 32 -.126 41 .537 18 .081 56
25 Navy 4-2 .133 42 .087 32 .762 110 9.1 4.3 .255 26 -.320 24 .485 80 -.878 111

Posted by: Brian Fremeau on 27 Oct 2010

6 comments, Last at 27 Oct 2010, 10:25pm by TomTom

Comments

1
by andrew :: Wed, 10/27/2010 - 2:00pm

When the season's long and the choice is yours alone,
When you're sure you've had enough of sending for Chris Hanson
Don't let yourself go for't, 'cause everybody kneels and everybody punts sometimes

Sometimes every play is wrong. Now it's time to kick away
When your drive is stalled again, (hold on, hold on)
If you feel like going on fourth, (hold on)
When you think you can convert fourth and two from your own 'thirty, well hang on

'Cause everybody punts. Take comfort in your d'fense
Everybody punts. Don't throw the ball. Oh, no. Don't throw the ball
If you go for't on 4th and two, no, no, no, they'll say you're through

If you're on your own in this game, the days and nights are long,
When you think you've had too much of coaching to hang on

Well, everybody punts sometimes,
Everybody kneels. And everybody punts sometimes
And everybody punts sometimes. So, hold on, hold on
Hold on, hold on, hold on, hold on, hold on (to the ball)
Everybody punts. You are not alone.

2
by mm (not verified) :: Wed, 10/27/2010 - 2:42pm

5 of the 6 SEC West schools in top 26. So far South Carolina is the only team that's beaten one of those schools when they're not playing each other.

Surprising to see Florida State so low. Still, several ACC schools have wandered into the top 25.

You know what would be cool in the future with F+/- rankings? Graphs for each conference displaying each school's F+/- rankings. If you could put them all on the same page we could compare different conferences by depth, quality at the top, quality at the bottom, etc. by sight.

So which schools in your top 25 have played the most 'high relevance' games so far? My guess is Oklahoma, LSU, and Auburn are up there.

3
by Kal :: Wed, 10/27/2010 - 3:08pm

Interesting. So FEI and S&P are starting to agree on Oregon, if only in slight increments. USC should be a good make it or break it game; it's by far their toughest test on the road. Still, after seeing the efficiency they showed at UCLA (where they punted once in the game and scored on every non-garbage time possession) I'm surprised to see them go down.

4
by CDB (not verified) :: Wed, 10/27/2010 - 4:50pm

I think "by far" is going a little overboard. They play @ OSU, who is every bit as good as USC and will be incredibly pumped should Oregon show up undefeated.

5
by Kal :: Wed, 10/27/2010 - 5:03pm

OSU is not as good as USC is, I don't think. They don't have the offensive threats with Rodgers gone, and they don't have the defensive talent. Their coaching is better, Katz is going to be a stud in the future, and Quizz is amazing - but it's not as threatening a game as USC is, especially playing at USC.

Oregon State's also been really up and down this season; losing at Washington really hurts them in my mind, IMO.

6
by TomTom (not verified) :: Wed, 10/27/2010 - 10:25pm

One thing that interests me is does Oregon actually have a game where every drive was considered significant? ASU perhaps, but any others? A couple of things that are hidden in these stats is that Oregon special teams have been magnificent (look at that punting, lovely punting), and that the second halves have been better than the first (eg 7 points given in 4th quarter). The latter might be a reason for Oregon to be underrated.

By the way, love everything you do, keep it up. GO DUCKS!

PS. Cal and OSU scare me more than USC. Either Oregon blows out due to turnovers and special teams being strong (like they have been), or Kelly outcoaches Kiffin in the 4th quarter. There's a reason USC lost by 3 points each game, and FEI doesn't notice the bad coaching, because it's just a simple field goal drive.