clock menu more-arrow no yes mobile

Filed under:

The Toolbox: Offensive Line Stats

New, 4 comments

One of the many areas in which football stats differ from those of other sports is with the nature of the offensive line.  We can figure out ways to evaluate individuals at the skill positions, and potentially even at a few positions on defense, but without knowing the playcall and knowing the role of every lineman, how can one possibly hope to evaluate an offensive line as a whole?  Even full game-charting data would only get us a small portion of the way there.

So we do what we can.  There are two primary measures we use for evaluating offensive line play.  Obviously there are no individual evaluations here -- the only thing we have that includes a single lineman's name is when they commit a penalty -- but with the following two measures (which you've already seen used quite a bit in the Summer Vacation series), we can at least get a general idea for how well an offensive line performed as compared to other units on offense.

Adj. Line Yards

From the NFL definition of Adjusted Line yards at the Football Outsiders glossary:

Adjusted Line Yards (ALY): Statistic which attempts to, even to a small extent, separate the ability of a running back from the ability of the offensive line. Adjusted Line Yards begin as a measure of average rushing yards per play by running backs only, adjusted in the following way:

  • 0-4 yards: 100% strength
  • 5-10 yards: 50% strength
  • 11+ yards: not included
  • runs for a loss: 120% strength

How did Aaron and company come up with those numbers?  From an expanded explanation:

We have enough data amassed that we can try to separate the effect that the running back has on a particular play from the effect of the offensive line (and other offensive blockers) and the effect of the defense. A team might have two running backs in its stable: RB A, who averages 3.0 yards per carry, and RB B, who averages 3.5 yards per carry. Who is the better back? Imagine that RB A doesn't just average 3.0 yards per carry, but gets exactly 3 yards on every single carry, while RB B has a highly variable yardage output: sometimes 5 yards, sometimes -2 yards, sometimes 20 yards. The difference in variability between the runners can be exploited to not only determine the difference between the runners, but the effect the offensive line has on every running play.

We know that at some point in every long running play, the running back has gotten past all of his offensive line blocks. From here on, the rest of the play is dependent on the runner's own speed and elusiveness, combined with the speed and tackling ability of the defensive players. If Tiki Barber breaks through the line for 50 yards, avoiding tacklers all the way to the goal line, his offensive line has done a great job -- but they aren't responsible for most of that run. How much are they responsible for?

For each running back carry, we calculated the probability that the back involved would run for the specific yardage on that play, based on that back's average yardage per carry and the variability of their yardage on every play. We also calculated the probability that the offense would get the yardage based on the team's rushing average and variability without the back involved in the play, and the probability that the defense would give up the specific amount of yardage based on its average rushing yards allowed per carry and variability. For example, based on his rushing average and variability, the probability in 2004 that Tiki Barber would have a positive carry was 80% while the probability that Giants would have a positive carry without Barber running was only 73%.

Yardage ends up falling into roughly the following combinations: Losses, 0-4 yards, 5-10 yards, and 11+ yards. In general, the offensive line is 20% more responsible for lost yardage than it is for yardage gained up to four yards, but 50% less responsible for yardage gained from 5-10 yards, and not responsible for yardage past that. Thus, the creation of Adjusted Line Yards.

...

The system is far from perfect. We don't know when a guard is pulling and when a guard is blocking straight ahead. We know that some runners are just inherently better going up the middle, and some are better going side to side, and we can't measure how much that impacts these numbers. We have no way of knowing the blocking contribution made by fullbacks, tight ends, or wide receivers.

There is a decent possibility that the parameters for line yards could be a little different at the collegiate level -- just like the parameters for success rates are different; one day I'll get to that piece of the to-do list.  For now, we simply use the NFL definition.  Even if there are differences, they will almost certainly not be significant enough to render the current measure worthless.

The Adj. Line Yards figure you see both after the jump and in the Summer Vacation profiles basically combines a team's Line Yards per carry with an adjustment for the quality of opponent.  I am sharing both the raw and adjusted figures below.

Adj. Sack Rates

This one is a little easier to figure out.  Sack Rate is simply (Sacks) / (Sacks + Passes).  It tells you how frequently a passer was sacked per pass attempt.  The "adjusted" portion of Adj. Sack Rates has the same sentiment as the adjustment for line yards -- it looks at who the pass attempts came against and how many sacks would have been expected against said opponents.  Again, below you'll see both raw and adjusted sack figures.

For purposes of team profiles, I like to look a lot at standard downs and passing downs sack rates.  It is part of the equation for the "Need for Blitzes" portion of the Defensive Footprint.  But for the rates below, it simply looks at all pass attempts -- standard down or passing down -- that took place while a game was "close."

Data

For perspective, I am listing the below teams according to their Adj. Run-Pass Ratio.  As you'll see, there is a bit of a correlation between Run-Pass Ratio and Adj. Line Yards.  To find your team the fastest, I would recommend simply doing a Ctrl-F search.

Quick note: like all the other opponent-adjusted stats I work with, Adj. Line Yards and Adj. Sack Rate are calibrated so that 100.0 = perfectly average. Above 100 is good, below 100 is bad. You'll notice that the extremes on both ends of the Adj. Sack Rate scale are much further from average than on the Adj. Line Yards scale. Among other things, that is because teams allow between about 10 and 40 sacks in a given year, but even the least-frequent rushing team rushes hundreds of times. The scale is a bit more sensitive on the sack rates side of the equation, then.

Offense Adj. Run-Pass Rk Sack Rate Rk Adj. Sack Rate Rk LY/Carry Rk Adj. LY Rk
Army 83.7% 1 5.8% 58 85.4 78 3.24 20 106.8 46
Air Force 81.3% 2 2.5% 10 211.0 12 3.52 7 112.7 20
Georgia Tech 79.2% 3 8.0% 101 77.8 95 3.28 15 111.8 23
Navy 78.8% 4 6.7% 76 75.6 96 3.43 11 109.2 31
Nebraska 67.2% 5 9.9% 117 60.4 113 3.22 22 108.4 35
Auburn 65.9% 6 6.6% 74 106.9 51 3.56 3 137.1 1
Illinois 65.6% 7 7.5% 88 68.9 107 3.27 16 113.7 16
Mississippi State 65.1% 8 7.8% 97 82.5 89 3.06 43 108.6 34
Central Florida 64.9% 9 5.8% 59 96.8 61 3.07 40 103.6 58
Northern Illinois 63.5% 10 3.4% 20 160.1 23 3.24 19 109.6 29
Wisconsin 63.3% 11 5.3% 49 89.5 70 3.53 6 120.8 6
Eastern Michigan 62.3% 12 6.6% 71 92.6 67 2.91 72 102.0 61
Ohio 62.1% 13 6.1% 61 87.7 72 3.01 50 105.8 50
TCU 61.9% 14 2.9% 15 186.0 16 3.54 4 114.4 13
Oregon 61.7% 15 2.6% 12 246.4 6 3.18 26 104.9 53
Kansas State 60.6% 16 10.1% 118 57.2 116 3.08 37 101.7 62
LSU 60.1% 17 6.9% 80 95.8 62 2.95 61 112.7 21
Western Kentucky 60.0% 18 8.9% 109 69.5 105 3.05 44 98.8 68
Nevada 59.5% 19 2.7% 14 227.7 9 3.63 1 123.6 3
Virginia Tech 59.5% 20 9.0% 111 74.6 98 3.17 29 113.4 17
Offense Adj. Run-Pass Rk Sack Rt Rk Adj. Sack Rt Rk LY/Carry Rk Adj. LY Rk
Ole Miss 59.2% 21 4.2% 34 152.6 28 2.76 84 95.8 81
UCLA 58.9% 22 7.6% 91 84.3 81 3.14 32 106.8 45
Connecticut 58.6% 23 3.7% 25 169.2 19 2.98 56 111.5 24
South Florida 58.5% 24 7.2% 86 94.3 63 2.92 70 110.6 26
Utah State 58.5% 25 8.0% 100 75.4 97 2.57 103 89.4 97
North Texas 58.1% 26 7.9% 98 70.2 104 2.97 59 96.1 79
Wake Forest 57.8% 27 6.1% 63 84.3 82 2.65 94 91.9 92
Michigan 57.7% 28 2.2% 6 228.8 8 3.51 8 127.6 2
Ohio State 57.6% 29 5.3% 50 104.6 54 3.25 18 119.1 8
Ball State 57.3% 30 5.0% 45 111.2 48 2.67 93 91.0 95
West Virginia 56.5% 31 6.6% 72 97.0 60 2.79 82 97.1 75
Stanford 56.1% 32 1.6% 1 386.8 1 3.17 28 107.7 39
Boston College 56.1% 33 7.8% 94 83.3 87 2.64 96 97.2 74
Temple 55.7% 34 7.8% 96 73.7 100 3.05 46 108.1 36
Middle Tennessee 55.5% 35 4.4% 37 119.2 41 2.84 75 91.4 93
Louisville 55.3% 36 4.6% 42 134.1 34 2.92 67 106.3 47
South Carolina 55.2% 37 6.3% 67 100.7 55 2.95 63 110.0 28
Toledo 55.1% 38 5.2% 48 127.8 36 2.69 90 97.3 73
Florida International 55.1% 39 4.5% 39 125.6 37 2.97 58 101.7 63
Fresno State 54.9% 40 9.7% 114 61.4 112 2.65 95 93.0 89
Offense Adj. Run-Pass Rk Sack Rt Rk Adj. Sack Rt Rk LY/Carry Rk Adj. LY Rk
Wyoming 54.9% 41 9.3% 112 63.5 111 2.13 119 76.4 119
Purdue 54.4% 42 4.3% 36 110.0 50 2.74 88 91.1 94
Rice 54.4% 43 6.2% 65 86.1 76 2.82 81 87.6 101
Florida 54.3% 44 5.4% 51 120.1 40 2.77 83 104.9 52
California 53.8% 45 7.1% 83 91.3 69 3.16 30 107.2 42
Georgia 53.7% 46 7.1% 84 86.9 74 3.19 24 115.2 11
New Mexico State 53.5% 47 4.1% 31 129.1 35 2.60 99 81.1 114
Kansas 53.5% 48 9.6% 113 47.5 119 3.08 39 102.0 60
Northwestern 53.5% 49 9.7% 115 48.0 118 3.06 42 109.1 32
UNLV 53.3% 50 7.0% 81 83.4 85 2.50 109 88.9 98
Miami-FL 53.0% 51 3.6% 23 173.8 18 3.19 23 112.4 22
Clemson 52.7% 52 3.5% 21 188.5 14 2.59 100 96.7 77
Minnesota 52.7% 53 3.2% 17 157.4 26 2.89 74 104.3 56
New Mexico 52.5% 54 8.4% 102 55.5 117 2.53 106 83.5 110
Florida State 52.5% 55 6.8% 79 94.0 65 2.82 79 113.2 18
Southern Miss 52.1% 56 3.6% 24 158.4 24 3.10 36 98.1 70
Iowa 52.0% 57 5.5% 52 84.0 83 3.00 54 103.2 59
Michigan State 52.0% 58 5.7% 56 86.5 75 3.03 47 107.7 38
Rutgers 51.9% 59 14.6% 120 46.8 120 2.45 112 86.1 107
Iowa State 51.8% 60 7.8% 93 78.0 94 2.83 78 99.9 66
Offense Adj. Run-Pass Rk Sack Rt Rk Adj. Sack Rt Rk LY/Carry Rk Adj. LY Rk
Boise State 51.8% 61 1.9% 4 316.1 3 3.23 21 107.0 44
BYU 51.6% 62 4.0% 30 140.8 30 3.29 13 104.6 55
Colorado 51.5% 63 4.7% 43 116.4 43 2.74 87 89.9 96
Pittsburgh 51.4% 64 6.1% 64 106.2 52 3.19 25 122.7 5
Florida Atlantic 51.4% 65 8.6% 106 71.6 103 2.32 118 78.0 118
UL-Monroe 51.1% 66 7.9% 99 74.0 99 2.57 102 83.4 111
Washington 50.9% 67 5.5% 53 115.5 44 3.11 35 105.8 49
Penn State 50.7% 68 2.4% 8 193.9 13 2.83 77 109.4 30
Syracuse 50.5% 69 8.5% 105 69.2 106 3.15 31 114.3 14
Vanderbilt 50.4% 70 8.9% 108 65.9 110 2.68 92 95.1 82
Tulsa 50.1% 71 5.8% 57 83.8 84 3.51 9 107.4 41
Buffalo 50.1% 72 5.6% 54 92.5 68 2.69 91 87.0 104
Utah 50.0% 73 2.5% 11 237.7 7 2.94 65 97.1 76
Alabama 49.8% 74 8.7% 107 78.7 93 3.35 12 118.8 9
Akron 49.7% 75 7.8% 95 73.0 102 2.52 107 93.0 88
Maryland 49.7% 76 5.6% 55 120.5 39 2.61 97 93.2 87
Louisiana Tech 49.6% 77 4.2% 33 136.1 33 2.98 57 104.3 57
Memphis 49.4% 78 8.4% 102 67.4 109 2.58 101 87.3 102
USC 48.5% 79 3.9% 27 160.8 22 3.54 5 123.1 4
San Diego State 48.4% 80 1.8% 2 276.1 4 2.94 64 88.9 99
Offense Adj. Run-Pass Rk Sack Rt Rk Adj. Sack Rt Rk LY/Carry Rk Adj. LY Rk
Baylor 48.4% 81 4.5% 38 111.8 47 3.01 49 107.9 37
UTEP 48.3% 82 4.6% 41 113.6 46 3.14 33 98.2 69
Oregon State 47.8% 83 8.5% 104 87.6 73 3.00 55 110.2 27
Texas 47.6% 84 4.3% 35 116.8 42 2.95 62 95.8 80
Kent State 47.3% 85 4.6% 40 114.7 45 2.44 113 83.7 109
Texas A&M 47.3% 86 6.8% 78 93.7 66 3.07 41 104.8 54
North Carolina 47.2% 87 7.8% 92 82.7 88 2.74 86 94.5 84
Kentucky 47.1% 88 4.0% 28 161.2 21 2.92 71 107.2 43
Washington State 47.0% 89 11.2% 119 60.0 114 2.48 111 82.7 112
Arkansas State 46.4% 90 5.1% 46 111.2 49 3.01 51 97.7 72
Virginia 46.4% 91 3.7% 26 137.4 32 3.29 14 111.1 25
Tennessee 46.2% 92 9.0% 110 68.7 108 2.56 104 92.6 91
UAB 46.1% 93 2.7% 13 187.4 15 2.92 68 100.9 64
Tulane 45.8% 94 7.2% 87 73.4 101 2.82 80 94.2 86
Oklahoma 45.8% 95 3.5% 22 158.3 25 2.60 98 92.6 90
Arkansas 45.5% 96 6.1% 62 100.2 57 3.13 34 119.5 7
Missouri 45.3% 97 3.3% 19 156.3 27 3.02 48 106.0 48
UL-Lafayette 45.0% 98 7.1% 85 85.8 77 2.38 116 80.4 117
Oklahoma State 44.8% 99 1.8% 3 322.1 2 3.44 10 116.3 10
Colorado State 44.6% 100 9.8% 116 57.8 115 2.75 85 96.3 78
Offense Adj. Run-Pass Rk Sack Rt Rk Adj. Sack Rt Rk LY/Carry Rk Adj. LY Rk
Houston 44.3% 101 2.4% 9 220.7 10 3.26 17 105.0 51
Troy 44.2% 102 4.0% 28 138.4 31 2.93 66 94.6 83
Marshall 44.2% 103 3.0% 16 180.9 17 2.42 115 80.6 116
Arizona State 44.0% 104 6.6% 72 97.7 59 3.01 53 99.6 67
Miami-OH 43.7% 105 6.6% 70 85.2 79 2.35 117 81.9 113
Cincinnati 43.2% 106 7.0% 82 100.4 56 3.18 27 114.8 12
Notre Dame 42.7% 107 4.1% 32 148.6 29 2.90 73 107.5 40
NC State 42.5% 108 6.7% 75 88.4 71 3.08 38 113.7 15
Duke 41.5% 109 5.2% 47 106.1 53 2.49 110 86.0 108
San Jose State 41.4% 110 6.4% 68 84.8 80 2.44 114 86.6 106
Arizona 41.3% 111 6.3% 66 99.6 58 2.97 60 98.0 71
Western Michigan 40.8% 112 5.8% 59 94.3 64 2.92 68 94.3 85
Central Michigan 40.6% 113 6.7% 76 80.4 91 2.50 108 80.8 115
SMU 40.6% 114 6.5% 69 83.4 86 3.61 2 113.1 19
Bowling Green 40.4% 115 7.5% 89 81.0 90 1.81 120 67.8 120
Indiana 39.6% 116 2.1% 5 220.2 11 2.53 105 87.1 103
Texas Tech 39.6% 117 3.2% 18 165.5 20 3.01 52 100.2 65
Idaho 35.5% 118 7.5% 90 79.4 92 2.83 76 88.1 100
East Carolina 35.2% 119 2.3% 7 261.8 5 3.05 45 109.1 33
Hawaii 28.6% 120 4.8% 44 122.7 38 2.73 89 86.7 105

Cam Newton was incredible in 2010, but he had quite possibly the best run blocking in the country doing him some favors as well (it is always a possibility, of course, that he was doing them some favors along the way too, and that rubbed off on their line numbers). That's how Newton's Adj. POE was not the highest among QBs.

How does Hawaii's Alex Green end up atop the Adj. POE rankings for 2010? Well, running when everybody expected a pass certainly helped ... but what perhaps made an even larger difference is the fact that his line did him very few favors. What Adj. POE sees is that most of the yards he generated, he generated on his own.

If you look closely enough, you can see Hawaii wasn't alone in the "don't run much, don't block the run well" crowd. While there is no correlation whatsoever between run-pass ratios and sack rates (which somewhat makes sense, since being pass-heavy could mean quick hitch after quick hitch and low sack rates, or it could mean old school, downfield passing and high sack rates), there is a reasonably strong one between run-pass ratios and line yards. A 0.33 correlation, to be exact. Not huge, but solid.

Click to enlarge.

So sack rates depend on scheme and intent (and a passer's own ability to avoid the rush). Line yards depend, at least partially, on practice and repetition.

I cannot emphasize enough that these measures are incredibly imperfect, even more flawed than other advanced stats. This is a given. There's only so much we can glean from a line of play-by-play. Hell, for that matter, there's only so much we could do if we had perfect, complete charting data.

But this is a start. You can learn more about a line with these stats than without them, which means they're worth something, even if we shouldn't overreact to the numbers' conclusions.