Writers of Pro Football Prospectus 2008

24 Sep 2014

FEI Week 4: Mean Wins Update

by Brian Fremeau

The first four weeks are in the books and now is as good an opportunity as any to check in on the preseason FEI projections. As I wrote in the final installment of the FEI Primer four weeks ago, we like being right, but it can be a heck of a lot of fun to find out how much we were wrong.

On the one hand, not much has dramatically changed at the very top of the FEI ratings. Alabama, Oregon, Florida State, Stanford, and Oklahoma were projected to be top-five teams in FEI, and all five are still ranked among the top eight in the current rankings. Part of the reason there hasn't been much movement among that group is due to the fact that preseason projections are still a factor, albeit one that only accounts for roughly 40 to 45 percent of each team's Week 4 rating. But FEI isn't alone. The Crimson Tide, Seminoles, Sooners, and Ducks are all ranked in the top four of the Associated Press poll, Coaches poll, and the SBNation Composite Top 25 (which includes our F/+ ratings as one of four factors). Stanford clings to their perch in FEI because they dominated possessions in their lone loss in all ways but the conclusion of those drives.

The overall rankings don't tell the whole story, however. There have been somewhat significant changes in the win forecasts for certain teams. Stanford's projected FEI rating (.253) isn't dramatically different than its current FEI rating (.248), but their season win total projection has dropped by 1.4 games in four weeks. Instead of 8.4 FBS mean wins (9.4 total including its win over FCS UC Davis), Stanford is now projected to finish the year with only 7.0 FBS mean wins (8.0 total). Alabama's mean wins forecast has dropped from 9.6 FBS mean wins (10.6 total) to only 8.3 FBS mean wins (9.3 total) -- and that's despite the fact that the Crimson Tide are undefeated and rank fifth nationally in net points per drive. They haven't looked flawless, but they've looked very good. The problem for the Tide is that the rest of the SEC West has looked better than projected.

Arkansas has climbed from No. 66 to No. 38 in four weeks, increasing its mean wins projection by 2.2 games. Mississippi State has climbed from No. 36 to No. 18 (+1.5 mean wins projection). Ole Miss has climbed from No. 30 to No. 11 (+1.5 mean wins). Texas A&M has climbed from No. 18 to No. 4 (+1.8 mean wins). Auburn has climbed from No. 8 to No. 3 (+1.1 mean wins). Alabama (-1.3 mean wins) and LSU (-1.0 mean wins) are still top-10 teams in the FEI ratings, but they aren't the projected division leaders anymore.

The SEC West is going to be a wild ride, but it is still way too early to call how the standings in that loaded division are going to shake out. It's obviously too early to call anything definitively in college football; a top-10 team has lost as at least a seven-point favorite in each of the first four weekends. But I do feel pretty confident that we can call one of our biggest misses of the year -- we were wrong about East Carolina.

The Pirates were ranked No. 61 in the FEI preseason projections. Our F/+ projections (and those of pretty much everyone else) slated ECU fourth in the American Athletic Conference race, behind Central Florida, Cincinnati, and Houston. The Pirates were projected to win 4.8 FBS games against a schedule that included three Power Five, top-40, non-conference opponents in the first four weeks of the season. FEI gave ECU a 53 percent chance of going 0-3 in those games, and only a 7 percent chance of going 2-1 or better. We were wrong about East Carolina.

The Pirates went toe-to-toe with South Carolina on the road in week two, nearly matching the Gamecocks in available yards. East Carolina won on the road at Virginia Tech a week later, then obliterated North Carolina last weekend by a final score of 70-41. The Pirates have generated only modest value on turnovers (+1.6 points) and field position (+4.6 points). They've simply been very strong against pretty good competition.

The fact that preseason projection data is still included in the FEI ratings is actually still holding them back a bit (No. 37 overall), and yet they have increased their projected mean wins total from 4.8 wins at in the preseason to 8.8 wins in the first four weeks. The Pirates top all other teams in mean wins added in that span.

FEI Season Win Projection Top Ten Positive Changes
Wk 4
Rank
Team FBS
W-L
Preseason
Proj MW
Current
Proj MW
Change +/-
37 East Carolina 2-1 4.8 8.8 +4.0
76 Georgia Southern 1-2 3.4 7.3 +3.9
12 Utah 2-0 4.5 7.2 +2.8
42 Iowa State 1-1 3.2 5.9 +2.7
75 Louisiana Tech 2-1 4.4 6.8 +2.4
68 Memphis 1-1 4.0 6.3 +2.4
38 Arkansas 2-1 2.6 4.9 +2.2
27 BYU 4-0 7.8 9.8 +2.0
32 Indiana 1-1 4.6 6.6 +2.0
41 Penn State 4-0 6.7 8.7 +2.0
FEI Season Win Projection Bottom Ten Negative Changes
Wk 4
Rank
Team FBS
W-L
Preseason
Proj MW
Current
Proj MW
Change +/-
98 Fresno State 0-3 5.9 3.5 -2.4
54 Texas 1-2 6.8 4.4 -2.4
128 Troy 0-3 4.5 2.0 -2.5
65 Northwestern 0-2 6.0 3.5 -2.6
107 Rice 0-3 5.7 3.1 -2.6
51 Michigan 2-2 8.1 5.4 -2.6
113 Kent State 0-3 5.6 2.8 -2.8
62 North Carolina 1-1 6.5 3.4 -3.1
114 Vanderbilt 1-3 5.3 2.1 -3.2
121 Louisiana Lafayette 0-3 7.5 3.7 -3.7

Most teams haven't dramatically altered their fortunes for 2014. More than half of all teams (50.8 percent, to be precise) have had a total mean wins shift from preseason to date of less than one game, including Florida State (+0.4 mean wins), Oregon (+0.2 mean wins), and Oklahoma (+0.0 mean wins). The influence of preseason data on the ratings will continue to be reduced over the next two weeks and eliminated altogether beginning with the Week 7 FEI ratings.

FEI 2014 Week 4 Ratings

The Fremeau Efficiency Index (FEI) is a college football rating system based on opponent-adjusted drive efficiency. Nearly 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 function of the starting field position and outcome of non-garbage possessions. Opponent adjustments are calculated with special emphasis placed on quality performances against good teams, win or lose.

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. SOS listed here includes all regular season games scheduled. A multifaceted approach to measuring schedule strength is available here.

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 RMW) represent the average total games a team with the given FEI rating should expect to win against the remaining opponents on its schedule.

These ratings are partially influenced by preseason projections, a function of Program FEI ratings, previous-year FEI and garbage time data, previous-year turnover-neutral, special teams-neutral, and field position-neutral FEI, returning starters, recruiting success, and quarterback reliance. As the season progresses and actual 2014 data continues to be collected, the weight given to projection data will be reduced each week until Week 7, at which point it will be eliminated from the rankings entirely. Offensive and defensive FEI ratings will also debut in Week 7.

Rank Team FBS
W-L
FEI LW GE GE
Rk
SOS SOS
Rk
FBS
MW
FBS
RMW
1 Florida State 2-0 .314 2 .059 53 .171 50 9.8 8.2
2 Oregon 3-0 .298 1 .237 16 .100 33 9.2 6.5
3 Auburn 3-0 .267 12 .249 14 .031 2 7.9 5.3
4 Texas A&M 3-0 .259 11 .423 3 .042 6 8.0 5.4
5 USC 2-1 .256 6 .151 28 .060 14 9.1 6.9
6 Stanford 1-1 .248 4 .210 20 .030 1 7.5 6.0
7 Alabama 4-0 .241 8 .318 9 .079 24 8.1 4.3
8 Oklahoma 4-0 .236 10 .363 6 .226 59 10.0 6.3
9 LSU 2-1 .229 5 .111 34 .051 10 7.5 5.2
10 UCLA 3-0 .219 23 .065 48 .064 16 8.2 5.8
11 Mississippi 3-0 .215 7 .479 1 .073 21 7.6 4.8
12 Utah 2-0 .209 33 .315 10 .042 5 7.0 5.2
Rank Team FBS
W-L
FEI LW GE GE
Rk
SOS SOS
Rk
FBS
MW
FBS
RMW
13 Georgia 2-1 .206 9 .299 11 .135 43 8.1 6.0
14 Notre Dame 3-0 .204 13 .351 7 .050 9 8.4 5.6
15 Oklahoma State 1-1 .202 19 .075 44 .061 15 7.5 6.3
16 Missouri 2-1 .194 3 .174 25 .157 48 8.0 5.6
17 Baylor 2-0 .193 24 .444 2 .181 52 8.1 6.1
18 Mississippi State 4-0 .190 31 .255 13 .070 18 7.3 4.1
19 Ohio State 2-1 .189 17 .202 21 .254 66 9.3 6.8
20 Wisconsin 1-1 .189 20 .225 17 .268 67 8.7 7.4
21 Clemson 0-2 .184 22 -.115 92 .078 23 7.5 7.0
22 Kansas State 1-1 .182 21 -.013 70 .091 28 7.1 6.1
23 South Carolina 3-1 .182 15 .061 51 .097 31 7.3 4.7
24 Arizona State 2-0 .174 28 .176 24 .072 20 6.5 4.6
25 Michigan State 1-1 .171 14 .220 18 .107 35 7.6 6.4

Posted by: Brian Fremeau on 24 Sep 2014

2 comments, Last at 26 Sep 2014, 3:36pm by skunkfunk

Comments

1
by Lance :: Wed, 09/24/2014 - 2:57pm

I really wish Oklahoma State would stop scheduling the I-AA schools. Or FCS-- whichever. Makes things so confusing and it doesn't help anyone. Gundy and Co. should be able to get a real D-I school that's on the lower tier. Tulsa? Hi. What about New Mexico State? UTEP? Just stop with the out-of-division schools at least.

2
by skunkfunk :: Fri, 09/26/2014 - 3:36pm

Does Florida State count?

Gundy has said before that he wants the easiest non-conference schedule he can get.

http://footballscoop.com/news/mike-gundy-fsu/