Writers of Pro Football Prospectus 2008

Most Recent FO Features

KhanSha1.jpg

» Futures: My Expansion Franchise

You've just been awarded an NFL expansion team and must build your personnel department. How would you do it? Matt Waldman takes on the exercise.

05 Dec 2013

VN: Hello, Old Friends

by Bill Connelly

In my lengthy SEC Championship preview Thursday morning at SB Nation, I referenced an old friend of mine, Adjusted Points.

Remember Adj. Score? It's a measure I lean on pretty heavily in my season preview series; it takes a team's performance in a given game and tells you how that team would have done against a perfectly average opponent, with a perfectly average number of breaks, in a given game. It's good for trend-spotting, but it's also good for gauging the effect of injuries.

I tend to revisit this measure once a year or so. Here's my original 2010 piece, here's 2011, and here's 2012. I use it extensively in my offseason previews at SBN. It's an interesting measure, but even more than with other measures, it's not incredibly useful until late in the season (or in the offseason). It's hard to spot trends while they're happening. But now that we're into December, it's time to find out what it can tell us.

I used Adjusted Points in the piece above to talk about a couple of key points for Auburn-Missouri -- namely, that James Franklin's injury hurt Missouri at least as much as key opponent injuries (Georgia's Todd Gurley, Florida's Jeff Driskel, etc.) helped the Tigers, and that Auburn's offense has been devastating over the last two months (though the defense has sprung some leaks of late). But what else can we glean from the Adj. Score data now that there's enough data to adjust?

Below is a hefty data dump, complete with the following measures:

  • Adjusted Off. Points per Game, Adjusted Def. Points Points per Game, Adjusted Scoring Margin (offensive PPG minus defensive PPG), and Adjusted Record (how many games would a team have won against this "perfectly average" opponent?). These are all based on the same Adjusted Points concept.
  • Weighted Scoring Margin (Wtd. Margin). This takes the same Adjusted Scoring Margin concept but gives three percent more weight to each proceeding game, meaning recent performance carries a bit more weight.
  • Momentum (Mo). For these purposes, this is simply a team's Weighted Scoring Margin subtracted from its Adjusted Scoring Margin. Baylor, for instance, is 3.8 points worse when you more heavily weight recent games. Temple, meanwhile, is 3.1 points better. This gives us a hint regarding who has improved or regressed, or who has been best by injuries (or who has gotten healthier). This is not necessarily due to "momentum," per say, but it was the best name I could think of. Simple "Change" or "Shift" could have also worked, but that's not as much fun.
  • Covariance (Covar). This is another measure I address at some point each season (and more heavily in the offseason). Here's what I wrote about it last year. This simply looks at who had their best performances (according to Adjusted Scores) against good teams and who had their best against bad teams. The higher the number, the further a team is on the "best against worst" scale. (Teams are ranked in that order. No. 1 doesn't mean "best" in any way; it only means "best against worst.") As we see below, there are assets and liabilities with being at either pole on the Covariance list.
Team Record Adj. W/L F/+ Rk Adj. Off. Pts Rk Adj. Def. Pts Rk Adj. Margin Rk Wtd. Margin Rk Mo Rk Covar Rk
Florida State 12-0 12-0 1 46.6 1 16.9 3 29.7 1 29.7 1 0.0 68 -119.0 115
Baylor 10-1 9-2 8 45.0 2 21.4 14 23.7 2 19.9 3 -3.8 125 375.6 2
Alabama 11-1 12-0 2 38.6 11 17.7 5 21.0 3 21.1 2 0.1 66 173.8 14
Oregon 10-2 12-0 6 42.8 4 23.4 24 19.5 4 18.2 5 -1.3 112 -59.6 100
Louisville 10-1 11-0 17 37.9 12 18.8 6 19.1 5 16.9 7 -2.2 117 164.5 17
Ohio State 12-0 11-1 5 42.5 5 23.7 26 18.8 6 19.2 4 0.3 54 22.2 70
Wisconsin 9-3 11-1 9 36.7 16 19.7 7 16.9 7 15.7 8 -1.2 110 91.0 36
Georgia 8-4 11-1 21 39.4 8 23.7 27 15.7 8 14.9 9 -0.8 96 30.0 67
Clemson 10-2 10-2 16 36.1 17 21.0 12 15.1 9 17.0 6 1.9 11 -49.3 97
LSU 9-3 11-1 18 39.3 9 24.3 32 15.0 10 14.3 11 -0.6 86 168.2 15
Texas A&M 8-4 10-2 23 43.6 3 28.7 76 14.9 11 13.5 13 -1.4 116 64.6 53
Auburn 11-1 12-0 7 39.5 7 25.7 48 13.8 12 14.8 10 1.1 26 174.4 13
Stanford 10-2 12-0 3 34.2 25 20.6 11 13.6 13 13.7 12 0.2 64 -40.1 95
Arizona State 10-2 11-1 4 35.4 20 22.4 19 13.1 14 12.4 15 -0.7 88 -7.8 84
Central Florida 10-1 10-1 22 38.9 10 26.8 56 12.1 15 11.4 18 -0.7 89 -44.3 96
Michigan State 11-1 10-2 12 25.8 89 13.8 1 12.0 16 12.3 16 0.3 55 -18.1 86
Oklahoma State 10-1 10-1 10 33.2 32 21.5 16 11.7 17 12.6 14 0.9 30 -257.0 124
Washington 8-4 11-1 19 32.9 37 21.7 17 11.1 18 10.5 22 -0.6 84 160.3 18
Missouri 11-1 11-1 13 34.3 24 23.3 23 11.0 19 10.7 20 -0.3 74 62.3 55
Marshall 9-3 12-0 42 36.0 18 25.0 41 11.0 20 12.1 17 1.1 24 70.6 47
Kansas State 7-5 10-2 25 37.1 14 26.8 57 10.3 21 10.5 21 0.2 58 -67.5 104
Miami 9-3 9-3 27 40.7 6 30.9 91 9.8 22 7.4 36 -2.4 119 45.0 62
Iowa 8-4 10-2 28 30.0 55 20.3 8 9.7 23 9.9 23 0.2 61 -20.6 87
BYU 8-4 10-2 26 31.0 50 21.3 13 9.7 24 9.1 26 -0.6 83 33.1 66
South Carolina 10-2 12-0 15 35.3 21 25.7 47 9.6 25 8.7 27 -0.9 99 13.7 75
Virginia Tech 8-4 12-0 20 25.3 97 16.0 2 9.2 26 8.5 28 -0.7 90 -55.4 99
USC 9-4 10-3 11 30.7 52 21.8 18 8.9 27 7.4 35 -1.4 115 98.1 33
Georgia Tech 7-5 9-3 30 37.3 13 28.4 73 8.9 28 9.2 25 0.4 50 293.7 3
Bowling Green 9-3 8-4 48 33.0 34 24.8 37 8.3 29 11.0 19 2.8 3 199.6 11
Oklahoma 9-2 9-2 29 32.9 38 24.8 39 8.1 30 7.5 33 -0.5 82 112.5 28
Utah State 8-4 11-1 32 25.6 95 17.6 4 8.1 31 7.9 32 -0.1 71 218.2 7
Arizona 7-5 10-2 31 32.0 43 24.1 29 7.9 32 6.7 41 -1.2 107 0.6 81
UCLA 9-3 10-2 14 32.6 39 24.8 38 7.8 33 8.0 31 0.1 65 67.8 48
Cincinnati 9-2 8-3 63 33.1 33 25.3 44 7.8 34 9.4 24 1.6 14 8.0 77
Mississippi State 6-6 11-1 43 30.1 54 22.5 20 7.6 35 7.1 40 -0.5 78 158.0 19
North Carolina 6-6 11-1 40 32.4 40 25.1 43 7.3 36 8.4 29 1.2 22 153.4 21
East Carolina 9-3 10-2 37 33.6 29 26.5 52 7.1 37 7.5 34 0.4 48 -122.0 116
Notre Dame 8-4 10-2 24 33.6 28 26.6 54 7.0 38 7.1 39 0.1 67 24.2 68
Boise State 8-4 10-2 39 31.3 47 24.5 33 6.8 39 7.1 38 0.4 52 44.8 63
Fresno State 10-1 8-3 52 33.7 26 27.1 61 6.6 40 8.0 30 1.4 17 15.6 73
Penn State 7-5 8-4 61 29.4 60 23.2 22 6.2 41 6.4 44 0.3 57 24.2 69
Oregon State 6-6 8-4 49 34.6 23 29.5 83 5.1 42 5.7 46 0.5 41 19.0 72
Nebraska 8-4 9-3 44 31.8 44 26.7 55 5.1 43 6.4 45 1.3 20 -27.9 92
Northern Illinois 12-0 9-3 53 35.9 19 30.9 90 5.0 44 7.3 37 2.3 7 -127.4 117
Utah 5-7 7-5 33 28.6 65 23.7 25 5.0 45 2.5 62 -2.4 120 121.4 26
TCU 4-8 9-3 46 25.4 96 20.6 10 4.8 46 6.5 43 1.7 13 -151.9 118
Texas Tech 7-5 9-3 60 32.9 36 28.3 72 4.6 47 3.7 54 -1.0 103 112.0 31
North Texas 8-4 8-4 50 28.5 66 24.1 30 4.4 48 6.5 42 2.1 8 113.0 27
Ole Miss 7-5 10-2 36 29.8 58 25.4 46 4.4 49 5.0 48 0.7 33 -71.3 106
Duke 10-2 9-3 34 31.2 49 27.0 60 4.1 50 4.3 51 0.2 62 81.1 41
Michigan 7-5 7-5 35 29.0 62 25.4 45 3.6 51 2.9 56 -0.8 94 67.6 49
Pittsburgh 6-6 7-5 59 28.4 69 24.9 40 3.5 52 2.3 63 -1.2 108 63.5 54
UTSA 7-5 10-2 67 32.0 42 28.6 75 3.4 53 5.0 49 1.5 15 82.0 40
Ball State 10-2 8-4 51 34.6 22 31.2 96 3.4 54 3.9 53 0.5 44 -90.4 109
Florida Atlantic 6-6 7-5 71 24.7 98 21.5 15 3.2 55 5.6 47 2.4 5 200.8 9
Toledo 7-5 9-3 62 33.6 27 30.5 87 3.1 56 4.2 52 1.1 25 -29.2 93
Houston 8-4 9-3 41 30.8 51 27.9 70 2.9 57 3.5 55 0.6 36 -110.4 113
Texas 8-3 6-5 38 29.1 61 26.6 53 2.5 58 2.7 60 0.2 59 -63.2 103
Western Kentucky 8-4 5-7 82 31.6 46 29.1 81 2.5 59 4.5 50 2.0 9 -27.4 91
Boston College 7-5 6-6 55 31.2 48 28.9 78 2.3 60 2.8 58 0.5 45 97.1 34
UL-Lafayette 8-3 6-5 84 33.4 31 31.1 93 2.3 61 2.8 57 0.5 43 132.4 25
Maryland 7-5 8-4 64 28.4 68 26.2 51 2.2 62 1.3 66 -0.9 102 104.4 32
Florida 4-8 7-5 45 22.5 105 20.4 9 2.2 63 -0.6 73 -2.8 123 -5.4 83
Indiana 5-7 6-6 57 36.9 15 35.1 110 1.8 64 -1.8 78 -3.5 124 53.6 59
Washington State 6-6 7-5 58 28.4 67 26.8 58 1.6 65 1.3 65 -0.3 75 189.3 12
Rice 9-3 6-6 66 26.2 85 24.7 36 1.5 66 2.7 59 1.2 21 12.5 76
Buffalo 8-4 6-6 73 26.3 84 25.1 42 1.1 67 2.2 64 1.0 27 97.0 35
Tennessee 5-7 7-5 76 28.2 73 27.1 62 1.1 68 0.0 70 -1.1 105 156.7 20
Vanderbilt 8-4 6-6 54 27.2 75 26.2 50 1.0 69 0.1 69 -1.0 104 112.3 30
Northwestern 5-7 5-7 56 29.9 56 28.9 77 1.0 70 0.3 67 -0.7 91 -96.7 111
Colorado State 7-6 7-6 68 28.4 70 27.7 66 0.7 71 2.6 61 1.9 10 88.8 38
Memphis 3-8 5-6 69 24.3 99 23.9 28 0.4 72 -0.9 76 -1.4 113 -229.9 122
Syracuse 6-6 5-7 77 25.7 92 25.9 49 -0.2 73 -0.8 74 -0.7 87 469.6 1
San Jose State 6-6 6-6 78 33.0 35 33.3 106 -0.3 74 -0.8 75 -0.5 79 -114.6 114
Navy 7-4 5-6 65 31.7 45 32.5 102 -0.8 75 0.2 68 1.0 29 60.2 56
San Diego State 7-5 5-7 93 23.6 103 24.5 34 -1.0 76 -0.4 72 0.6 34 -173.3 120
Virginia 2-10 3-9 75 21.3 111 22.9 21 -1.6 77 -2.1 79 -0.4 77 254.4 5
Illinois 4-8 6-6 70 33.5 30 35.1 111 -1.6 78 -2.4 83 -0.8 95 13.9 74
Rutgers 5-6 4-7 94 25.9 88 27.7 68 -1.8 79 -4.4 93 -2.5 121 79.5 43
Arkansas 3-9 3-9 87 28.3 72 30.5 88 -2.2 80 -3.0 86 -0.8 93 64.7 52
South Alabama 5-6 5-6 79 28.9 64 31.2 94 -2.3 81 -2.4 81 -0.1 70 57.2 58
UNLV 7-5 4-8 95 26.7 78 29.0 80 -2.3 82 0.0 71 2.3 6 38.9 65
SMU 5-6 3-8 89 29.4 59 31.9 99 -2.4 83 -2.3 80 0.2 60 -0.2 82
N.C. State 3-9 5-7 88 26.9 77 29.5 82 -2.6 84 -4.0 90 -1.4 114 20.2 71
Ohio 7-5 7-5 104 26.3 83 28.9 79 -2.6 85 -5.2 99 -2.6 122 112.4 29
Minnesota 8-4 4-8 47 25.8 90 28.5 74 -2.7 86 -2.4 82 0.3 56 67.3 50
Tulsa 3-9 4-8 91 24.3 100 27.4 64 -3.1 87 -3.8 89 -0.7 92 49.9 61
West Virginia 4-8 4-8 81 26.4 82 30.1 86 -3.7 88 -4.9 94 -1.3 111 -16.8 85
Troy 6-6 3-9 106 29.9 57 33.6 107 -3.8 89 -3.2 87 0.5 40 203.3 8
Colorado 4-8 3-9 96 25.7 94 29.5 84 -3.8 90 -5.0 97 -1.2 109 284.8 4
Middle Tennessee 8-4 4-8 74 28.3 71 32.4 101 -4.1 91 -2.7 85 1.5 16 -90.0 108
Arkansas State 7-5 6-6 92 26.5 80 30.6 89 -4.1 92 -5.0 95 -0.8 98 138.4 24
Hawaii 1-11 4-8 85 23.5 104 27.7 67 -4.1 93 -3.3 88 0.9 31 71.0 46
Wake Forest 4-8 5-7 80 20.0 117 24.2 31 -4.2 94 -4.2 92 0.0 69 65.8 51
Temple 2-10 6-6 97 30.4 53 34.7 109 -4.3 95 -1.2 77 3.1 1 -254.9 123
Tulane 7-5 3-9 72 19.9 118 24.6 35 -4.6 96 -4.0 91 0.6 35 73.5 45
Kentucky 2-10 3-9 98 26.5 79 31.5 98 -4.9 97 -7.3 104 -2.4 118 200.1 10
Wyoming 5-7 3-9 102 27.0 76 32.2 100 -5.1 98 -6.0 101 -0.9 100 -36.1 94
Kent State 4-8 2-10 103 26.0 87 31.2 95 -5.2 99 -2.6 84 2.5 4 88.7 39
Akron 5-7 4-8 107 22.0 106 27.3 63 -5.4 100 -5.0 96 0.4 51 -164.1 119
Iowa State 3-9 3-9 83 25.8 91 31.4 97 -5.6 101 -5.1 98 0.5 42 6.0 79
Connecticut 2-9 3-8 100 21.1 112 26.9 59 -5.8 102 -5.2 100 0.6 37 -26.3 89
Texas State 6-6 2-10 108 20.9 113 27.8 69 -6.8 103 -7.4 105 -0.5 80 41.9 64
UL-Monroe 6-6 3-9 111 20.9 114 28.2 71 -7.3 104 -6.3 102 1.0 28 244.2 6
New Mexico 3-9 2-10 109 32.2 41 40.9 121 -8.7 105 -7.9 108 0.8 32 145.6 23
South Florida 2-9 3-8 90 18.5 123 27.4 65 -9.0 106 -7.8 107 1.2 23 -84.1 107
Nevada 4-8 1-11 86 26.4 81 35.5 113 -9.1 107 -7.7 106 1.3 18 80.2 42
Central Michigan 6-6 1-11 110 23.8 101 32.9 104 -9.1 108 -7.3 103 1.8 12 6.2 78
Purdue 1-11 2-10 113 21.9 107 31.1 92 -9.2 109 -10.1 109 -0.9 101 76.9 44
California 1-11 2-10 105 25.7 93 35.4 112 -9.7 110 -10.1 110 -0.4 76 -307.8 125
Kansas 3-9 1-11 101 18.8 121 29.6 85 -10.8 111 -11.4 113 -0.5 81 -61.5 101
Louisiana Tech 5-7 1-11 112 21.9 109 32.8 103 -11.0 112 -11.2 111 -0.2 73 -49.4 98
Air Force 2-10 1-11 114 26.2 86 37.5 119 -11.4 113 -12.0 114 -0.6 85 90.0 37
Army 2-9 1-10 99 27.7 74 39.2 120 -11.5 114 -11.3 112 0.2 63 2.0 80
Southern Miss 1-11 1-11 119 19.7 119 33.3 105 -13.6 115 -13.0 117 0.6 38 50.9 60
UAB 2-10 0-12 115 28.9 63 42.6 124 -13.7 116 -13.9 118 -0.2 72 -62.6 102
Western Michigan 1-11 1-11 117 19.6 120 33.8 108 -14.2 117 -12.8 116 1.3 19 -210.7 121
Georgia State 0-12 2-10 121 20.7 115 35.7 115 -15.0 118 -12.2 115 2.8 2 -25.0 88
Idaho 1-11 2-10 116 20.5 116 35.8 116 -15.2 119 -14.9 119 0.4 53 -69.4 105
Massachusetts 1-11 0-12 120 18.5 122 36.3 117 -17.8 120 -17.4 120 0.4 47 -96.7 110
Eastern Michigan 2-10 0-12 125 21.9 108 41.6 122 -19.7 121 -20.8 121 -1.2 106 58.6 57
UTEP 2-10 0-12 118 21.5 110 41.6 123 -20.1 122 -20.9 122 -0.8 97 164.8 16
New Mexico State 2-10 0-12 122 23.8 102 45.5 125 -21.7 123 -21.3 123 0.4 49 -26.5 90
Miami (Ohio) 0-12 0-12 123 13.6 124 37.2 118 -23.6 124 -23.1 124 0.6 39 -104.0 112
Florida International 1-11 0-12 124 11.8 125 35.7 114 -23.9 125 -23.5 125 0.4 46 151.8 22

Okay, so what does this data tell us?

Baylor has faded. Key injuries have slowed the Bears down on both sides of the ball. They can still score points, but it is far more difficult for hem to outrun and outscheme opponents without receiver Tevin Reese and, until recently, running backs Lache Seastrunk and Glasco Martin. Meanwhile, the loss of linebacker Bryce Hager has potentially been equally damaging. Regardless, the Bears are the "coldest" team in the country, having lost 3.8 points per game from their earlier pace.

Nine teams have played well enough to beat an average team every week of the season: Alabama, Auburn, Florida State, Louisville, Marshall (!), Oregon, South Carolina, Stanford, and perhaps most surprisingly (or most questionably), Virginia Tech. All but one of these teams has lost a real game, however.

The 10 BCS conference teams with the most momentum: Temple (+3.1), Clemson (+1.9), TCU (+1.7), Cincinnati (+1.6), Nebraska (+1.3), North Carolina (+1.2), South Florida (+1.2), Auburn (+1.1), Oklahoma State (+0.9), Ole Miss (+0.7).

The 10 BCS conference teams with the least momentum: Baylor (-3.8), Indiana (-3.5), Florida (-2.8), Rutgers (-2.5), Utah (-2.4), Miami (-2.4), Kentucky (-2.4), Louisville (-2.2), Texas A&M (-1.4), USC (-1.4). Either because of injuries or other attrition-related factors, this list looks about right, at least until you get to Louisville and USC.

Your top 11 "Best against Best" teams: California (record: 1-11), Oklahoma State (10-1), Temple (2-10), Memphis (3-8), Western Michigan (1-11), San Diego State (7-5), Akron (5-7), TCU (4-8), Northern Illinois (12-0), East Carolina (9-3), Florida State (12-0). Combined record of these 11 teams: 66-64.

Your top 10 "Best against Worst" teams: Syracuse (6-6), Baylor (10-1), Georgia Tech (7-5), Colorado (4-8), Virginia (2-10), UL-Monroe (6-6), Utah State (8-4), Troy (6-6), Florida Atlantic (6-6), Kentucky (2-10), Bowling Green (9-3). Combined record of these 11 teams: 66-65.

We tend to give more respect to the teams that bring it most in big games, but this can cut in one of two ways. Either it can push a great season over the top, as it has for Northern Illinois and Florida State, or it could mean that you come up short against really good teams, then lay veritable eggs against bad teams, and basically lose all of your games (California, Temple, etc.). Or, in Oklahoma State's case, it means you're capable of whipping Oregon but losing to West Virginia.

A thought exercise

What happens if we take teams' momentum and apply it to this week's conference championship games? Here are this week's F/+ picks; here's what happens if we apply momentum factors to the Old Method picks at that link:

  • SEC Championship (Auburn vs. Missouri). Old pick: Auburn by 6.8. Momentum-adjusted pick: Auburn by 8.2.
  • ACC Championship (Duke vs. Florida State). Old pick: FSU by 27.9. Momentum-adjusted pick: FSU by 27.5.
  • Big Ten Championship (Ohio State vs. Michigan State). Old pick: Ohio State by 1.3. New pick: Ohio State by 1.3.
  • Pac-12 Championship (Stanford at Arizona State). Old pick: ASU by 0.1. New pick: Stanford by 0.8.
  • Conference USA Championship (Marshall at Rice). Old pick: Marshall by 3.5. New pick: Marshall by 3.4.
  • MAC Championship (Northern Illinois vs. Bowling Green). Old pick: NIU by 2.4. New pick: NIU by 1.9.

No major changes there, but we'll see which set of picks ends up closer.

This week at SB Nation

Really happy with this week's work at SBN. I'm happy with it most weeks, but … this week in particular.

Last Friday
My heart is full

Monday
From the Egg Bowl to Auburn, college football just topped itself

Tuesday
The Numerical, Week 14: It's really, really hard to win three straight titles
Missouri 28, Texas A&M 21: Beyond the box score

Wednesday
Win probabilities: Projecting college football's championship weekend
Week 15 F/+ picks
Week 14 advanced box scores

Thursday
2013 SEC Championship breakdown: 10 things to know about Auburn-Missouri
2013 SEC Championship game: Perusing the Auburn 2-deep

Posted by: Bill Connelly on 05 Dec 2013

6 comments, Last at 06 Dec 2013, 12:19pm by Aaron Brooks Good Twin

Comments

1
by Aaron Brooks Go... :: Thu, 12/05/2013 - 5:43pm

So... we've learned that Auburn has basically become Ohio State, and that Virginia Tech is even Michigan Statier than Michigan State?

2
by Aaron Brooks Go... :: Thu, 12/05/2013 - 6:09pm

Also, it's important to point out, that the last team to win 3 straight NCs (Army, 1944-1946) not only was playing professional players, they were literally drafting their competition's best players.

The NCAA death-penalty'd SMU for far less than that.

So... if Missouri nipped Auburn, Duke crushed FSU, Michigan State won like they took the "State" out of Ohio, and Oklahoma State won like they put the "State" back into Texas...

Is there any chance we could have an Alabama-Baylor NC game involving two non-conference champs?

4
by justanothersteve :: Thu, 12/05/2013 - 9:20pm

Baylor is ranked ninth in the BCS. They'd have to jump seven teams. I'd say somewhere between slim and none.

6
by Aaron Brooks Go... :: Fri, 12/06/2013 - 12:19pm

I just thought it would be important to match an Alabama team that can't win their own division, let alone conference, against an equally-qualified opponent.

3
by Kal :: Thu, 12/05/2013 - 6:22pm

This week has been really excellent. You're definitely a best against best author. :)

One of the biggest things I've been looking at is the variance angle. Covariance gets it somewhat, but only somewhat; what I'd like to figure out is whether or not a team could (if they play as good as they could) beat another team or has shown any indication that they could do that. Some teams are average but highly variant; others are average and consistent. Oregon had been hugely consistent from week to week until this year, for instance. Alabama seemed to jump all over the place.

FEI gives this up via Game Factors, which have been hugely valuable. This is somewhat done on the S+P breakdowns for the game. Have you thought about doing histograms of performance, similar to what DVOA has done on a per-team basis during the playoffs?

5
by analyze-football (not verified) :: Fri, 12/06/2013 - 5:28am

You're great. Even amazing!