Writers of Pro Football Prospectus 2008

Most Recent FO Features

WentzCar16-1.jpg

» Week 11 DVOA Ratings

DVOA has finally climbed on board the Wentz Wagon! The Eagles move into the No. 1 spot, but they aren't the only strong, well-balanced team in the NFL this year. New Orleans, Pittsburgh, and the Los Angeles Rams make this one of the best seasons ever for multiple teams over 30% in DVOA, and Minnesota isn't far behind.

08 Nov 2017

FEI Week 10 Ratings

by Brian Fremeau

As debate over the race for the College Football Playoff ramps up in the month of November, it's inevitable that fans and pundits will latch on to select numbers to make a case for or against a given team. Sometimes those numbers can be misleading, or seem to tell one story when in reality they tell another.

On Tuesday night, for instance, I posted the total point differential teams have accumulated against opponents ranked in the latest CFP committee top 25 rankings. Notre Dame stands head and shoulders above every other team by this measure, having racked up a +75-point margin in four games against currently ranked playoff teams (a one-point loss to Georgia, a 35-point win over USC, a 20-point win over Michigan State, and a 21-point win over North Carolina State). The next best cumulative margin (+29) belongs to both Georgia (in two games) and Clemson (in three games). Alabama is further down the list at +14, a single 24-10 victory over LSU.

Cumulative scoring margin will naturally favor a team that has more opportunities, whereas average scoring margin will tell a different one. Notre Dame has a +18.8 points per game margin against the CFP top 25; Georgia is at +14.5; Alabama is at +14.0; and Clemson is at +9.7. That's still a selective list, of course, since all games against the subset of teams are given equal footing whether the opponents were in the top five or barely ranked in the top 25. And games against opponents that might be good but are just outside the CFP top-25 subset are ignored altogether. Both the cumulative and per-game data sets tell a story, but neither one does a great job of telling the entire, more complex story of the season to date.

One of the drivers that led me to run the numbers on cumulative scoring margin in the first place was an exploration of dominance. The playoff selection committee has used "game control" language in describing how teams have impressed or not impressed in the past, and that language has stuck with me. Our colleagues at ESPN maintain a metric they call Game Control, defined as the chance an average top-25 team would control games from start to end the way the given team did, given the schedule. They don't further define what control of games from start to end necessarily means, but it likely has something to do with the in-game win probabilities a given team has over the course of its games.

Another way to evaluate how a team maintains control in a given game is to simply calculate how often it takes a lead, maintains a lead, and grows a lead in a given game. Victory, even by a narrow margin, is the primary goal for every team, but teams are more likely to be consistently victorious if they rarely trail, and frequently play with a multiple-score lead.

Alabama leads the nation in average scoring margin, outscoring their opponents by 26.0 points per game in non-garbage time. The Crimson Tide have had the lead on their opponent on 86.7 percent of their non-garbage possessions, also best in the nation. They have only trailed on 1.1 percent of their non-garbage possessions. They have never trailed by more than a single score this year. They have led by at least three scores (17-point margin or better) on 35.0 percent of non-garbage possessions, and by four scores (25-point margin or better) on 14.4 percent of non-garbage possessions. They've epitomized consistent dominance unlike any other team this season.

Contrast the Crimson Tide with another undefeated team, the Miami Hurricanes. Miami's average margin of victory this year is 10.9 points per game in non-garbage time. They have trailed their opponent on 22.8 percent of their non-garbage possessions this season. They have led by 17 or more points on only 2.8 percent of non-garbage possessions, and haven't led by 25 or more points at all this year in non-garbage time (or at all, for that matter, against FBS competition).

Both teams are undefeated, of course, and there's certainly no reason Miami will be held back from competing for a national championship if the Canes keep winning, regardless of how dominant or non-dominant those victories will be down the stretch. But nearly every team that has reached the College Football Playoff in its first three iterations managed to be more dominant than the Hurricanes have been to date. All but one of those teams led their opponents by at least 17 points on at least 10 percent of their non-garbage possessions. The only exception among the playoff teams since 2014 was the Florida State Seminoles in the first year of the playoff. More so than the Hurricanes this year, Florida State lived on the edge in 2014, managing to go undefeated while leading their opponents on only 40.3 percent of possessions. But even that team had a higher percentage than this year's Miami Hurricanes of possessions played with a three-score lead or better -- Florida State played 5.0 percent of non-garbage drives in 2014 with a lead of 17 points or more, a percentage nearly twice as high as Miami has recorded to date.

With major matchups this weekend between top teams -- including Miami playing host to Notre Dame, Auburn hosting Georgia, and Oklahoma hosting TCU – there may be limited opportunities for the nation's best to take a commanding lead and maintain it against a formidable opponent. But those that can and do may be well-positioned for postseason success down the stretch.

Click here for more data on the distribution of scoring margins by possession for all 130 teams.

FEI 2017 Week 10 Ratings

The Fremeau Efficiency Index (FEI) is a college football rating system based on opponent-adjusted drive efficiency. Approximately 20,000 possessions are contested anually 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 Georgia 8-0 .300 .346 3 1.51 29 3.59 4 .52 1 .04 41
2 Notre Dame 8-1 .297 .264 6 1.89 9 3.67 3 1.15 7 .02 47
3 Clemson 8-1 .279 .229 10 1.21 46 2.80 25 .99 5 .00 59
4 Oklahoma 8-1 .257 .182 15 1.79 14 4.17 1 2.35 71 -.08 108
5 Penn State 7-2 .248 .244 9 1.37 37 3.17 10 1.39 14 .15 5
6 TCU 7-1 .248 .219 12 1.76 16 2.70 33 1.38 13 .19 1
7 Central Florida 7-0 .223 .302 4 .43 108 3.28 7 1.34 11 .07 22
8 Ohio State 7-2 .220 .251 8 1.60 26 3.55 5 1.48 21 .01 56
9 Alabama 9-0 .217 .373 1 .89 74 3.22 8 .73 2 .10 12
10 Auburn 6-2 .216 .252 7 1.76 15 2.85 22 .94 4 .05 35
11 Washington 7-1 .213 .362 2 .55 101 3.19 9 .91 3 .10 14
12 Iowa State 5-3 .208 .112 25 1.82 13 2.42 45 1.48 20 .07 25
13 Miami 7-0 .204 .122 21 .95 68 2.87 20 1.47 19 -.08 110
14 Iowa 6-3 .202 .108 26 1.87 10 2.65 34 1.47 18 .05 36
15 Oklahoma State 7-2 .199 .203 13 1.62 23 3.67 2 1.70 26 -.13 127
Rk Team Rec FEI GE Rk SOS Rk OFEI Rk DFEI Rk STE Rk
16 Wisconsin 9-0 .195 .267 5 .65 91 2.78 28 1.13 6 .04 40
17 USC 8-2 .153 .085 33 1.42 34 3.11 12 2.03 48 -.02 78
18 Michigan State 7-2 .146 .043 48 1.94 7 2.19 66 1.45 17 -.05 97
19 Mississippi State 6-2 .141 .191 14 1.57 27 2.85 23 1.35 12 .06 30
20 Virginia Tech 6-2 .126 .168 16 1.35 39 2.22 59 1.23 10 .12 7
21 Michigan 7-2 .121 .123 20 1.61 25 2.31 52 1.40 15 .04 42
22 North Carolina State 5-3 .113 .044 47 1.70 19 2.90 18 2.10 50 -.08 113
23 Washington State 7-2 .111 .090 30 1.07 59 2.01 84 1.18 8 -.07 106
24 Stanford 6-3 .109 .121 23 1.51 30 2.72 31 2.23 65 .11 9
25 Texas 4-5 .107 .037 51 2.18 1 1.68 110 1.19 9 -.01 68
26 Boise State 7-2 .101 .136 18 .40 113 2.56 39 1.98 44 .10 13
27 Boston College 5-4 .100 .009 60 1.83 12 2.27 57 1.62 25 .06 31
28 Toledo 7-1 .097 .156 17 .64 94 2.96 16 2.21 63 .11 10
29 Arizona 5-3 .095 .077 36 .75 79 3.34 6 2.65 95 -.05 95
30 Houston 6-3 .094 .084 34 .59 97 2.06 77 1.56 24 .01 51

Posted by: Brian Fremeau on 08 Nov 2017

5 comments, Last at 15 Nov 2017, 9:08pm by Brian Fremeau

Comments

1
by Happy Fun Paul :: Thu, 11/09/2017 - 5:36pm

I'm confused. Perhaps I don't understand, but I thought that FEI is a combination of GE (general efficiency) and SOS (strength of schedule, where "ratings represent the average number of losses an elite team (two standard deviations better than average) would have against the team's entire schedule").

From last week, I sort-of understand why Alabama is currently ranked so low (#9)-- because their SOS is so low. Nevertheless, Alabama does have a higher SOS and higher GE than #7 Central Florida. So... why is 'Bama ranked behind UCF?

2
by Brian Fremeau :: Thu, 11/09/2017 - 6:29pm

The numbers aren't strictly a combination of GE and SOS, but those are solid indicators of a team's FEI rating. Alabama has a better overall SOS (including regular season games not yet played against Mississippi State and Auburn), but does not have a stronger SOS to date than UCF. UCF SOS to date is ranked 107th, Alabama's SOS to date is ranked 119th. SOS data fleshed out here: http://www.bcftoys.com/2017-sos/

3
by Happy Fun Paul :: Fri, 11/10/2017 - 9:47pm

Ah! Thanks for the clarification.

4
by Happy Fun Paul :: Fri, 11/10/2017 - 9:57pm

Oh, for what it's worth, I'd prefer having PEL (expected elite team losses to date) replace SOS (against overall schedule*) in the top 30 table. I think PEL would be more useful, and that was what I thought was being shown in the first place.

* I assume SOS is
(1) against a team's overall regular season schedule, not including conference championships and beyond, since opponents aren't yet known, and
(2) that SOS (and PEL, for that matter) get readjusted every week depending on how a team's opponents fare in their subsequent games.
Correct?

5
by Brian Fremeau :: Wed, 11/15/2017 - 9:08pm

Thanks for the suggestion. I've revised it for this week and going forward, and will use PEL (schedule played to date) for the SOS column in the weekly FEI ratings.