Wednesday, August 16, 2017

Cowboys OL Sacks Allowed Project

Dallas Cowboys quarterback Dak Prescott (4) receives a hand from guard Ronald Leary (65) after being sacked by Green Bay Packers outside linebacker Nick Perry during the fourth quarter of their NFC Divisional playoff game at AT&T Stadium in Arlington, Texas, Sunday, January 15, 2017. (Tom Fox/The Dallas Morning News)
Tom Fox/Staff Photographer
Dallas Cowboys quarterback Dak Prescott (4) receives a hand from guard Ronald Leary (65) after being sacked by Green Bay Packers outside linebacker Nick Perry during the fourth quarter of their NFC Divisional playoff game at AT&T Stadium in Arlington, Texas, Sunday, January 15, 2017. (Tom Fox/The Dallas Morning News)
It is getting harder and harder to get to the QB in the NFL.  Sacks are at an all-time low.  Offenses are designed to get the ball out quickly, which keeps QBs from getting beat up and keeps offenses moving forward.  

In 2016, the NFL had its lowest sack rate (5.8% of all pass attempts ended in sacks), since the league started tracking those statistics.  We don't know if it will continue to drop as we continue to evolve, but long gone are the days where the sack rate was up near 10%.
In 2016, the league averaged a sack for every 17.4 pass plays.  To show you how that has changed in just a few seasons, let's look back:
In 2013, the league averaged a sack for every 15 pass plays.
In 1997, the league averaged a sack for every 13.6 pass plays.
And, in 1984, the NFL averaged a sack for every 11.9 pass plays.  
We pass more than ever, get the ball out quicker than ever, and QBs are playing longer than ever before.  Tom Brady just turned 40 and won a Super Bowl last season.  Peyton Manning won a Super Bowl at age 39, the year before.  And QBs are taking fewer sacks.  I assume there is a relationship between those two things.  
So, as the sack rate falls, we value sacks more and more.  The only way to slow these teams down is to punish them physically.  But, it is so hard to get to the QB, because protection is better and the scheme's objective is often to win the fight by making it short.  A QB gets the ball out quicker than ever to routes that are designed to be open faster and to do their damage after the catch.  Coaches have been thinking this way for decades, but the evolution has brought us to a point where playoff teams average surrendering fewer than 2 sacks a game, despite passing the ball more than ever before.  
Let's look at the last decade.  In purple, notice the last 4 years where the sack rate has fallen each season to its current all-time low of 5.8%.
The blue line is the Dallas Cowboys offensive sack rate.  As you can see, for the last decade, the Cowboys sack rate on sacks allowed has continuously been lower than the league sack rate.  This is a credit to the offensive design of Jason Garrett and then Scott Linehan, as well as the mental and physical tools of Tony Romo and, now, Dak Prescott.  
The Cowboys allow very few sacks as a group and last season was no exception when the Cowboys conceded a total of 28 sacks.
In 2016, the Cowboys allowed 28 sacks on 511 pass attempts for a sack rate of 5.5%.  This means, stated another way, that the offense conceded a sack for every 18.3 pass attempts.  
They allowed the 7th-fewest sacks in the league, but since they also passed the ball relatively few times, that means their sack rate was actually only 14th in the league.  Oakland led both categories by allowing only 18 sacks which was 2.9% of their attempts or one sack for every 34.1 attempts.  Wow.
Now, looking back at just the raw sack totals from the last decade, here is how 28 allowed stacks up with the last 10 years:
As you can see on both charts, the 2011 season was the low-water mark for the Cowboys offensive line - oddly enough the year they thought starting Bill Nagy and Phil Costa on the offensive line was a great plan - and since then, they prioritized their OL with both draft picks and pay-checks and have basically grown into the model OL in the business.  
But, since then, the sack totals have never been a concern.  They have always been at or below the league averages and have protected quite well.  Both Romo and Prescott have shown times where they have cost themselves sacks by trying to hang on to the ball.  There are clearly QBs in this league that will not hold the ball in the chaos until the last second, but the Cowboys have QBs with good eye-levels down field and will take a hit for the cause in the right situation. 
Over the last week, I wanted to go back to those 28 (30 if you include 2 in the playoffs) sacks and evaluate each one to try to find the bust or the culprit that led to a breakdown in protection.  While it is true that they only allowed 30 sacks all season, it is also true that through the Cleveland game on November 6th, the total was 11.  
So, said another way, the Cowboys allowed 11 sacks in the first 8 games (1 sack for every 23.7 attempts or 4.2%) and 19 sacks in the last 9 (1 sack for every 15.3 attempts or 6.6%).  
That isn't great.  At all.  Down the stretch, with what we think is the best OL in the business, on a stretch run that might make it all the way to the Super Bowl, the pass protection started breaking down to a point where they allowed a sack rate of 6.6%.  Over the course of the entire year, that sack rate puts you at 25th in the league.
In other words, if you are ranked 25th in pass protection when it counts the most, you might not have the best OL in the league.  
Now, before we get too carried away, we should look for explanations.  That is why I went back and wanted to see what we could see from each sack.  I wanted to know if we over-rate this group or if there are some explanations.  So, I looked at all 30 sacks.
Now, the problem with this, of course, is that I am not looking at all 511 pass attempts.  We shouldn't evaluate JUST the sacks to get a full picture of the protection.  But, I also didn't have 3 weeks to do this project, so bear with me.  
Before we get to my observations, let's look at the whole list.  I invite you to examine them with your own eyes, if you would like.  This, as much of film watching can be, is a subjective exercise.  Assigning blame for sacks requires a little bit of opinion and that is where I might see this different than Scott Linehan.  But, since he didn't offer to do this for public appraisal, you are stuck with me.  Here is what I found:
Here are some observations from my study:
1) - Doug Free had a fine career and showed toughness and the ability to play hurt for a long, long time.  He should be remembered as a strong player who really contributed at one of the most demanding positions in football for a decade.  That said, he was leaking oil in 2016 and was fighting Father Time.  His retirement is not hard to understand after doing this study.  In several games - At Washington, At Minnesota, and At New York, in particular, he had a really rough time handling things on an island.  And, the Cowboys certainly were not giving him much help in these instances, because they wanted their TE and RB out to give shorter targets to the QB.  He had a nice run, but he was culpable on between 8-10 of these sacks.  La'el Collins will have a hard time adjusting to this spot, but Free's bar from 2016 is far more attainable than Tyron Smith.
2) - Tyron Smith was not immortal in 2016.  There were seasons where you didn't find a single sack or holding penalty to hang on Tyron.  In 2016, he was playing hurt, too.  And it showed at times as his name appears on this list several times.  I still would take him over any linemen in football.  I think he is insanely good and is still just 26 years old.  He was awesome, yet less awesome than normal.  We can only hope his back does not become an issue.  
3) - Ron Leary and Travis Frederick do not appear on this list at all.  Inside line positions are much easier to handle in pass protection and if a guard or center gave up more than 3 sacks in a season it would be a lot (for a tackle that is not a lot), but both seemed to have nothing to do with any of the 30 sacks.  
4) - No RB seemed to blame for any sack that I could see.  We debate pass protection skills of the RBs quite a bit, but it seems that in 2016, no RB pulled a "Chris Gronkowski" moment (ask your friends).  
5) - Zack Martin was very good in 2016.  But, he was run over by big Stephon Tuitt in Pittsburgh (Notre Dame on Notre Dame crime).  He also was the man closest to the scene for a sack against the 49ers and another against Tampa Bay where the LB would delay, wait for Zack to go help someone, and then sneak through his now-vacated spot to get to Dak on a delayed blitz.  He is a heck of a player, but he did show up a few times.
6) - As you would expect, the majority of these were in shotgun and 13 were on 3rd downs.  There were several play-action plays where the defense was sitting on the fake, and there were a number of occasions where Dak Prescott caused his own sack by taking too long to abort the broken down play.  
Overall, I would suggest that this is a very good offensive line, but they should expect that the Giants and Vikings game plans of rushing 4, dropping 7, and trying to make Dak have to work to find openings will continue to be a look the Cowboys see.  The Giants, in particular, had a wonderful plan to confuse and restrict the Cowboys passing attack.  We must ask how well the receivers can get open against tight man coverage, as nothing causes more sacks than fantastic coverage by a secondary.
They did well, but we must hope that in 2017, the offense can protect better down the stretch.  15 sacks after December 1st last year will be something that give Cowboys' opponents plenty of hope to get to Prescott when it counts.

No comments: