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.

19 Oct 2017

FEI Week 7 Ratings

by Brian Fremeau

Through the first seven weeks of the season, I reduce the weight of preseason projection data in the weekly FEI ratings until it is eliminated completely from the formula. This week represents the first FEI ratings of the year in which no preseason projection data is included at all. Every team is measured only by what it has or has not yet accomplished to date.

I look forward to this point in the season every year for the sometimes unexpected surprises I find when the spreadsheets process the drive data. Sometimes I see it coming before this week, as is the case with Alabama. I wrote two weeks ago about how the SEC has been underwhelming to start the year, and that has led to the Crimson Tide having rolled to a comfortable 7-0 start to the year without having faced a strong slate. Alabama has the nation's second-best raw game efficiency data behind Central Florida, a measure of how well a team maximizes its own possessions and limits those of its opponents. But their non-conference win over Florida State doesn't measure as impressively now as it did then, and Alabama hasn't faced an FEI top-20 opponent yet. They'll move up if they're just as dominant the rest of the year, but they've slipped to No. 6 overall in this week's FEI ratings.

Opponent adjustments are significant in the FEI ratings formula, boosting teams that have impressive wins coupled with solid games against good teams, win or lose. Iowa State is a great example, having already faced three opponents ranked in the FEI top 30, with losses to Iowa and Texas and an impressive road win against Oklahoma. FEI ranks Iowa State No. 15 overall this week. Compare the Cyclones with one of the teams against which they have a head-to-head loss, No. 20 Iowa (4-2 against FBS opponents), and it might be a head-scratcher. Iowa State has a slightly better overall game efficiency rating, but the key difference is the win over Oklahoma in comparison to Iowa's loss to Penn State -- each team's only game to date against an opponent ranked in the top 10.

This week is also the first time in the year when I publish all of the supplemental ratings for offenses, defenses, and special teams. The methodology for calculating the data hasn't changed, but I've changed the way I'm presenting opponent-adjusted offensive (OFEI) and defensive (DFEI) efficiency.

The nation's No. 1 offense belongs to Oklahoma State this week, with a 4.35 OFEI rating. That rating is intended to represent a per-possession scoring value, adjusted for both starting field position and opponent defenses faced, and representing the Cowboys' offense's specific contribution to scoring. Oklahoma State is averaging a nearly identical total in points per non-garbage drive (4.34), but that number is representative of all scoring, field goals or touchdowns, at the conclusion of a given drive. Their offense has been better than that, responsible for setting up what should have been an average of 4.68 points per drive -- the Cowboys' raw offensive efficiency -- when starting field position is removed from the equation and Oklahoma State's field goal success (or lack thereof -- they're 9-of-13 on the year) is also removed. That's adjusted back down to 4.35 points in the OFEI rating when the strength of the defenses they've faced is considered.

The nation's No. 1 defense belongs to Georgia. The Bulldogs are giving up only 0.76 points per drive. When starting field position and opponent field goal success is removed, their defensive efficiency brings that number down to 0.91 points per drive. And when we consider the offenses Georgia has faced, with particularly strong efforts against Notre Dame's No. 5-ranked offense and Mississippi State's No. 24-ranked offense, their DFEI rating brings that adjusted per-possession scoring value down to 0.64 points per drive.

The numbers presented in this new way also help contextualize how much impact special teams plays into per-possession scoring rates. The best offenses generate more than four points per drive, and the best defenses allow less than one point per drive. But the best special teams units cumulatively generate less than 0.30 points per possession -- Kentucky leads the way this week with a 0.23 STE rating -- and the worst give up about that much value per possession to their opponents. That adds up, and particularly excellent or poor special teams play in a given game can have a big impact, but it is a fractional impact over the course of a year in comparison to offensive and defensive efficiency data.

Traditionally, the FEI ratings remain relatively stable from the midpoint in the year through the end of the regular season, but there have been dramatic movements from teams from time to time. Often, that can be tied to injuries that may, for instance, derail a hot start. But it can also be tied to dramatic changes in schedule strength played at the start of the year versus the finish.

There are many ways I present strength of schedule data on my own site, even though I isolate only one of those methods in the weekly ratings posted here. The average number of losses expected by an elite team against a given schedule is my preferred primary schedule strength rating. For overall schedule strength, Maryland has the nation's toughest regular season slate according to this method. For schedule strength to date, Arkansas has played the nation's toughest slate. But what about schedule strength remaining, and better yet, which teams will play a schedule most dramatically different than the one they've played thus far?

The table below lists 20 teams with big changes in schedule strength difficulty -- the ten teams whose future schedules are most difficult, compared to the teams they have played so far, and ten whose schedules are about to get a lot easier.

More Difficult Remaining Schedule Than Schedule To Date
Team SOS SOS
Rk
Previous
SOS
PSOS
Rk
Remain
SOS
RSOS
Rk
delta
Kansas 2.24 2 0.57 64 1.67 1 1.10
Georgia Tech 2.05 10 0.55 68 1.50 2 0.95
Texas Tech 2.18 4 0.69 44 1.49 3 0.80
Navy 1.06 62 0.14 119 0.92 28 0.78
Michigan 1.83 19 0.53 70 1.30 5 0.76
Penn State 1.43 42 0.38 88 1.05 20 0.67
Rutgers 1.89 17 0.61 58 1.28 8 0.66
North Carolina State 1.54 34 0.44 83 1.10 17 0.66
Illinois 1.46 41 0.41 84 1.05 21 0.64
Ohio State 1.63 29 0.51 73 1.12 14 0.61
Less Difficult Remaining Schedule Than Schedule To Date
Team SOS SOS
Rk
Previous
SOS
PSOS
Rk
Remain
SOS
RSOS
Rk
delta
UNLV 0.77 81 0.71 40 0.06 124 -0.65
Vanderbilt 1.23 54 0.95 16 0.29 75 -0.66
Cincinnati 0.95 71 0.81 27 0.14 96 -0.66
Indiana 1.95 14 1.31 2 0.64 48 -0.67
Appalachian State 0.83 78 0.76 37 0.08 117 -0.68
Kent State 1.03 63 0.86 23 0.17 90 -0.69
Boston College 1.87 18 1.29 3 0.58 56 -0.71
Akron 1.03 64 0.92 17 0.11 104 -0.81
Arkansas 1.71 27 1.36 1 0.35 68 -1.01
Missouri 1.24 53 1.14 6 0.10 108 -1.04

The Big Ten East has many strong teams, and most of the games to be played between those contenders are yet to occur. Ohio State, Penn State, and Michigan all have a round-robin forthcoming, and there are other pitfalls for each as well. On the flipside, Missouri and Arkansas, both winless in SEC conference play thus far, can each feel better that the back half of their respective schedules is more manageable than the front half has been.

FEI 2017 Week 7 Ratings

The Fremeau Efficiency Index (FEI) is a college football rating system based on opponent-adjusted drive efficiency. Approximately 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 measure of net success on non-garbage possessions, and FEI opponent adjustments are calculated with special emphasis placed on quality performances against good teams, win or lose. SOS ratings represent the average number of losses an elite team (two standard deviations better than average) would have against the team's entire schedule.

Offensive FEI (OFEI) is value generated per drive adjusted for starting field position and opponent defenses faced. Defensive FEI (DFEI) is value generated per opponent drive adjusted for starting field position and opponent offenses faced. Special Teams Efficiency (STE) is the average value generated per possession by a team's non-offensive and non-defensive units.

Ratings for all 130 teams can be found here.

Rk Team Rec FEI GE Rk SOS Rk OFEI Rk DFEI Rk STE Rk
1 TCU 5-0 .312 .217 13 1.79 20 3.07 15 1.15 9 .20 3
2 Georgia 6-0 .285 .327 6 1.32 49 3.33 8 .64 1 .04 41
3 Clemson 6-1 .261 .249 9 1.30 51 2.92 20 1.08 8 -.02 77
4 Penn State 6-0 .253 .331 5 1.43 42 2.90 22 .91 4 .18 5
5 Ohio State 6-1 .247 .392 3 1.63 29 4.02 3 .86 3 .05 33
6 Alabama 7-0 .244 .398 2 .87 75 3.27 9 .71 2 .10 12
7 Oklahoma 5-1 .243 .198 14 2.14 7 4.08 2 2.12 59 -.13 122
8 Miami 4-0 .239 .133 19 .96 69 2.99 17 1.59 27 -.09 115
9 Notre Dame 5-1 .237 .226 12 2.15 6 3.47 5 1.29 12 -.01 73
10 Washington 5-1 .206 .344 4 .88 74 3.13 12 .98 7 .04 40
11 Oklahoma State 5-1 .199 .320 7 1.89 16 4.35 1 1.66 31 -.13 124
12 USC 6-1 .196 .113 27 1.57 33 2.95 18 1.92 46 .00 62
13 Central Florida 5-0 .185 .408 1 .39 109 3.68 4 1.47 20 .10 13
14 Michigan 5-1 .185 .119 24 1.83 19 1.82 98 .96 6 .09 17
15 Iowa State 3-2 .179 .123 21 2.07 9 2.66 33 1.96 48 .14 7
Rk Team Rec FEI GE Rk SOS Rk OFEI Rk DFEI Rk STE Rk
16 Stanford 5-2 .179 .162 15 1.76 23 3.17 11 2.29 69 .10 16
17 Auburn 4-2 .166 .236 11 1.76 24 2.53 41 .94 5 .07 24
18 Wisconsin 6-0 .166 .242 10 .74 82 2.77 26 1.18 10 .03 46
19 North Carolina State 5-1 .161 .116 26 1.54 34 3.13 13 2.09 58 -.10 119
20 Iowa 4-2 .156 .105 29 1.76 25 2.68 31 1.85 41 .08 21
21 Michigan State 5-1 .145 .059 43 1.97 13 2.02 81 1.25 11 -.05 101
22 San Diego State 5-1 .143 .060 41 .57 97 2.07 75 1.51 21 .03 45
23 Florida State 2-3 .132 -.031 80 1.98 12 2.20 63 1.32 13 .00 63
24 Texas Tech 3-2 .126 .059 42 2.18 4 3.03 16 2.27 66 -.03 84
25 Arizona State 3-3 .123 -.001 70 1.58 32 2.74 28 2.03 56 -.02 76
26 West Virginia 3-2 .121 .140 17 2.12 8 3.23 10 2.38 73 -.02 74
27 Virginia Tech 4-1 .121 .146 16 1.52 36 2.18 66 1.39 16 .09 19
28 Texas 3-3 .120 .047 49 2.46 1 1.95 87 1.42 17 -.04 88
29 Virginia 4-1 .120 .119 23 1.06 61 2.69 30 1.59 28 -.08 109
30 Washington State 5-1 .116 .120 22 1.40 45 2.25 57 1.44 19 -.05 96

Posted by: Brian Fremeau on 19 Oct 2017

0 comments