Were the NFL’s record passing yards mostly due to poor tackling?

Much was made about the NFL’s “record” passing season in 2011.  Indeed, there were a number of QBs who threw for over 5,000 a pretty remarkable feat.

However, when I compared the 2011 and 2010 performances using the advanced statistics I use on this blog the relevant overall numbers put up by QBs who took at least 300 snaps in each of the two seasons were nearly identical, which suggests the record number of yards were most likely: produced by the upgrade in passers, or produced by “Yards after the Catch” which would indicate poor tackling.

Before I show the chart, here is a primer on the two main “advanced” QB statistics I use:

Forward Yards per Snap: This statistic is inspired by Brian Burke’s “Air Yards” stat.  “Forward Yards per Snap” is an attempt to calculate the number of forward yards each Quarterback produced for his team with his arm or feet for every snap he took.  The statistic discounts the number of yards produced by his receivers, and also discounts the negative yards produced by each turnover (minus 30 yards per) and each sack (minus the yards lost).  Here is the formula:

FwdYds/Snap= Rushing Yards + Passing Yards – Receiver Yards after Catch – 30xturnovers – Sack Yards Lost / Snaps taken by the Quarterback

Quarterback Efficiency Rating (“QBER”): This statistic is inspired by David Berri’s QB Score and my own observation that sacks and incomplete passes tend to be drive killers (along with penalties) and that the QBs who produce more yards per “dead play” tend to be most successful (or so it seemed to me after watching endless hours of football).  So, QBER is, simply, the number of forward yards each QB produces (including receiver yards and QB rushing yards) for every “dead” play he produces.  Once again, QB yards are discounted for turnovers (30 yards) and yards lost on sacks.   QBER Formula:

QBER= Passing Yards + QB Rushing Yards – 30xQBturnovers – Yardslostonsacks / Incompletions + Sacks

The Chart for QBs who played in 2011 and 2010

So, with that introduction, here is the chart of QBs who took at least 300 snaps in both 2011 (those are the first set of numbers, and they are in order of best Fyds/Snap average) and 2010.

QBs SNAPS FRW YDS FYDS/SNP QBER SNAPS FRW YDS FYDS/SNP QBER
Rodgers 988 2308 2.33 23.89 880 1849 2.09 19.21
Vick 787 1834 2.33 16.71 723 1823 2.52 18.58
Brees 1117 2420 2.16 23.26 1063 1603 1.51 15.79
E Manning 1028 1974 1.92 16.08 1035 1410 1.36 14.14
Brady 1082 2047 1.89 19.63 973 1758 1.81 18.68
Ryan 1022 1871 1.83 14.85 1088 1876 1.72 13.96
Romo 925 1656 1.79 17.04 383 489 1.27 19.33
Palmer 635 1003 1.58 14.68 1038 1408 1.35 12.51
Schaub 648 1020 1.57 16.77 1029 1586 1.54 15.44
Rivers 1048 1552 1.48 15.89 1036 1653 1.59 18.13
Cutler 645 957 1.48 12.35 867 998 1.15 11.18
Roethlisberger 954 1378 1.44 14.44 806 1527 1.89 16.11
A Smith 993 1368 1.38 13.82 646 584 0.91 11.84
McCoy 850 1192 1.39 10.35 430 590 1.37 11.91
Young 207 275 1.33 11.44 317 694 2.19 15.39
McNabb 308 409 1.32 11.69 785 1121 1.43 11.86
Hasselbeck 876 1104 1.26 13.85 1153 908 0.78 10.75
Freeman 918 1115 1.21 12.19 903 1746 1.93 15.87
Orton 580 681 1.17 11.95 891 1479 1.66 13.04
Flacco 1036 1193 1.15 11.32 1016 1321 1.29 13.42
Fitzpatrick 992 1141 1.15 13.23 811 1178 1.45 11.87
Kolb 587 529 1.11 10.73 342 278 0.81 9.59
Cassell 574 563 0.98 10.39 1010 1222 1.21 13.17
Bradford 658 594 0.89 7.68 1053 888 0.84 10.45
Sanchez 1030 856 0.83 9.32 1045 1424 1.36 10.96
TOTALS 19458 30184 1.4788 14.142 20278 29989 1.4812 14.1272

Numbers nearly identical for both seasons

The source for the calculations was Yahoo Sports.

As you can see, for quarterbacks who took the qualifying number of snaps in each season, FYds/Snap and QBER remained nearly identical overall.  In each season, qualifying QBs produced about 1.48 forward yards per snap, and a QBER of about 14.13 or 14.14, somewhere in that range.

Those are remarkably consistent results.  However, within those results, there is some fluctuation.  For instance, only 29% of the number of FYds/Snap gained by each quarterback in 2011 is explained by his 2010 average (corr=0.54).  When we move to QBER, the explanatory value improves somewhat (corr=.72).  Thus, a little over half of the QBs QBER average is explained by his QBER in 2010.

Some interesting observations within the numbers:

1. Donovan McNabb was deemed “washed up” when he actually outperformed Matt Hasselbeck in Fyds/Snap.  Hasselbeck, however, had the better QBER.

2. The only two QBs who produced over 2.00 forward yards per snap in each season were Aaron Rodgers of the Packers and Michael Vick of the Eagles.

3. The Jets Mark Sanchez’s Fyds/Snap went way down, but his QBER didn’t decline that badly.  This suggests he really missed WR Braylon Edwards, and was forced to dump the ball off to backs a lot more.

4. Sam Bradford flat out sucks.  He has done nothing… nothing… to justify being considered the QB of the future in St Louis.  If the Rams pass on the Quarterback from Baylor, Robert Griffin III, they are fools.  But apparently new Rams coach Jeff Fisher “loves” Bradford.  Fisher is a bad quarterback evaluator.

5. The only QBs who were above the qualifying average in both seasons and in both QBER and FwdYds/Snap were: Aaron Rodgers, Michael Vick, Drew Brees, Tom Brady, Matt Schaub, Phillip Rivers, and Ben Roethlisberger.  The last two — especially Rivers — surprised me.  But, those, I would submit, are your “elite” Quarterbacks in the NFL.  You notice Eli is not on the list.  He almost made it, but his Fyds/Snap was just under average in 2010, otherwise he would have made it.

Advertisements

Tags: , , , , , , , , ,

2 Responses to “Were the NFL’s record passing yards mostly due to poor tackling?”

  1. Ra's Head Says:

    It seems that how a quarterback throws the ball may affect YAC in some way. If they throw behind the reciever, he may not be able to turn upfield for example. So I am not sure that the QB should get no credit for YAC, but I hadn’t thought much about that before. Certainly I agree that they not get full credit for it.

    Also, you may cover this somewhere, but where does the coefficient for turnovers (30) come from? I would think the value of a turnover would be hard to acertain since it depends on so many things.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s


%d bloggers like this: