Tuesday, November 04, 2014

Decoding Linehan - Week 9 - Arizona

http://cowboysblog.dallasnews.com/2014/11/decoding-linehan-missing-opportunities-through-the-air.html/

When analyzing the offensive performance of the Cowboys in Week 9 against Arizona, we are struck with how the QB position is still the basis for all offensive success. Regardless of how many offensive linemen are taken in the 1st round of the draft and regardless of the quality of the skill position players, we can look all around the league and see perennial doormats who are attacked relentlessly by every defense on their schedule because they do not have a QB who demands the full fear and respect of the defenses they face.
This is a very real issue faced by teams who do not own a top-level QB year after year. In fact, the Cowboys dealt with this between Troy Aikman and Tony Romo for years. You simply don't know your advantages as an offense until you must work through a few games without having any fear in your opponent's hearts.
It is difficult to operate as an offense to run smoothly and efficiently when the defense sets up in an aggressive posture to "dare" the QB to beat him, and then he demonstrates that he is unable to do so.
These are the realities of starting a backup QB. Today, it is Brandon Weeden, but last year it was Kyle Orton, and before that we have seen Brad Johnson, Jon Kitna, and even Stephen McGee have all taken a turn at the helm. It all comes back to the fact that starting QBs are highly compensated because there is a far smaller supply of competent QBs than there is a demand. And when they get hurt, you will have a significant drop off in performance because your backup is going to make a small fraction of what you pay Romo - and for very good reason.
In a league of 32 teams where each team needs at least 2, you can see that there are at minimum 64 job openings for Quarterbacks in the NFL. Now, name how many that you think there are that would make a top defense like Arizona feel the slightest twinge of fear and respect? 10? 12? Surely, no more than 15. Weeden is not on that list.
Let's demonstrate what we are talking about a bit before we look at the numbers. This might help with the discussion of "why didn't Linehan call better plays?" or "where were the easy underneath routes to free up easy targets?" It seems that most logical ideas were tried by the coaching staff and against that defense on that day, nothing was working.
Since it is Tuesday morning, we are left with merely the TV copy to demonstrate these points, but I think it will work on this occasion (although it would be ideal to have the All-22s).
Here are two consecutive 3rd Down-and-long situations in the 3rd Quarter (there are many to choose from, but these 2 I think are the most clear cases of Weeden predetermining throws). The Cowboys run the exact same play and the Cardinals seem to be in a very similar underneath coverage of man. And, on both occasions, Weeden misses Cole Beasley who is wide open underneath and seemingly able to easily convert these crucial 3rd Downs.

Above, 3rd and 7.  Cowboys have trips to the left, Bryant to the right.  They need to get to the 30 yard line and we asked, "where are the man-beating rub routes underneath?"  Behold, right here on the screen.  Witten is impeding Beasley's man with his out-breaking route.  Beasley is wide open as he crosses at the 27.  I am confident that if you are aware he is there, Dan Bailey could have converted this 3rd Down if forced to play emergency QB.
Instead, Weeden is looking right the entire time.  He is locked on Bryant and the Cardinals have Peterson on him - knowing that if Bryant is going to run his most likely route in this scenario - the deep in - he will have safeties coming down hill to cause issues.  Meanwhile, there is Beasley with his man about 10 yards behind him and wondering why the Cowboys didn't take an easy 3rd Down conversion and instead tried to force the ball to where the window was going to be tight and Bryant was going to feel an impact.
So, I am sure they go to the sideline and look at the pictures and discuss what they had there.  Now, it is 3rd and 9 on the very next drive.  The Cowboys will call the exact same play with the exact same protection and everything.  The Cardinals will try to blitz, so look left this time.

Here we go again, and again Witten does a nice job of getting in the way of Beasley's guy, although clearly that guy is expecting it because he just saw this play. Nevertheless, I think Cole is quite capable of taking a pass at the 14 yard line and getting to the sticks with his wheels. Of course, Weeden doesn't see him again and instead tries to reach Romo on the sideline even though Witten is drawing a corner in man-coverage. You will see below that this is basically Weeden's only significant throw to the left side of the field all day and he misses where you can't miss.
So, in both cases of a crucial 3rd Quarter in a game you were still in, it isn't merely the throws that were off. It is also the ability to see the whole field and make a throw where the percentages are in your favor. The easiest throws for a backup QB have to be the one's right in front of you against man coverage. Those are 7-on-7 throws of the most basic level. But, Weeden didn't see them and Beasley was pretty much open on both occasions.
When you have that level of play from your QB, you quickly deduce 2 things. 1) There is no level of play calling that can compensate for that performance. and 2) it is vital to keep Tony Romo upright at all costs or the season will quickly nosedive.
In case you are a masochist, here is a brief QB game-log for those 15 starts during Romo's career where he has been unable to answer the bell due to fitness. He missed 3 in 2008 with the thumb, 10 in 2010 with the collarbone, last year's finale with the back, and now Sunday. 15 starts and here are the performances:
As you can see, Kitna in 2010 actually settled in and wasn't bad for a few weeks here and there.  Otherwise, it has been pretty rough.  Which, by the way, you would expect and maybe hope for when your starting QB is in.  It hurts to look at it from this perspective, but if you simply apply the "do we miss him when he is gone" test to your QB, then it is easier to know if you are paying the right guy.
So, the average performance from your back-up QB during this stretch is 19 for 31, 218 yards, 1 TD and 1 INT for a rating of 78.7.  An average Tony Romo start during that same stretch is 22 for 34, 270 yards, 2 TDs, 0.9 INTs, and a passer rating of 96.4.
But, what stats have never done for us is show with great accuracy the effects of how one part of the offense feeds another.  For instance, if you have a good QB, how much more unlikely is it that Arizona stacks the box and forces you to stop running?  This is the long-running fight I have with people that don't value running the ball.  Good running makes passing so much easier.  And, good passing  also makes running easier.  It is a symphony of football where you are trying to make the defense unsure of what you are doing and therefore they can not deploy the adequate number of troops to stop everything you want to do.  If they sense you can only do one thing, then that will not function as well - even though it is not the fault of that specific portion of your attack.
That is why, this week, people will complain about the running game and the offensive line and say, what happened to them?  Well, in my opinion (and I will try to show this later in the week), the running game suffered because the passing game was gone.  Arizona deployed troops to gang up on the offensive line and there was no way to chase them off that idea unless your QB could make them pay.  So, complaining about the running game not performing is somewhat misguided because they are running against 8 defenders rather than 6 or 7 as they have when they were impressing you so much earlier in the season.
The Cowboys did find some running room - 21 carries under center yielded 75 yards - but, clearly, when you are at 3.6 a carry, you cannot sustain drives by just running the ball into strong fronts. You have to have some ways to back off the big and physical Cardinals.
So, that led people to wonder about screens and other ways to help along a QB - which they did hit on early in the game for 40 yards - but the Cardinals also know what you want to do to free up space and they are sitting on it. Arizona played a very smart defensive game and if they were going to leave anything exposed, it was going to take some precision deeper throws from Weeden to make them pay. And they won that bet.
Offensive Participation:  A fully healthy offense has started to go the other direction as on this occasion, they were playing without 3 regulars.  Romo is an issue all of its own, but then you add 2 reserves on the offensive line and it starts to become problematic.  Of course, that is the NFL game - being able to deal with attrition - but without Doug Free and Ron Leary does present issues.  I thought Parnell continues to play pretty well, but the drop off to Bernadeau was seen in those inside runs.  It would have been interesting to see how Leary could have handled Calias Campbell and Dan Williams and whether that 4th and 1 in particular is any different.  All snap numbers courtesy of PFF and they include all snaps including plays that were not official because of penalties.
STATS FOR WEEK 9 AGAINST ARIZONA
I think this is where we get a good look at how everything is connected.  1st down failures set up 2nd and real long.  That can easily lead to 3rd and real long.  On Sunday, both happened.  And 3rd and 6 is just not the same as 3rd and 8 on average.  You could feel the noose tightening all day long and they seemed behind the sticks and off schedule all day - largely because the Cardinals decided they were not going to "lose" 1st down to DeMarco Murray.  They would deploy all the troops necessary to make sure that did not happen.
=====
PASSING CHART This season, we're attempting to track both passing and drive progression. John Daigle has designed a fantastic chart.  Each color, for instance, represents the possession number listed in the key. The numbers are separated by the half. If you were to start from the bottom and work your way up, you would be tracking that possession from beginning to end. The dotted-lines are incompletions. Large gaps between throws are mostly YAC or carries.
Week 9 Summary
Wow.  Think the Cowboys were right-handed on Sunday?  Odd choice, given that the left corner is where Patrick Peterson lives.  But, this throw chart tells you that the issues are pretty clear when you have these tools to study.  Forcing the ball to 88 is not the way to win.
DRIVE STARTERS - The 1st play of each drive can often reveal the intent of a coach to establish his game plan. How committed is he to the run or pass when the team comes off the sideline? We track it each week here.
This week, the Cowboys did attempt to foil the Arizona plan on 1st down to pass them out of that, but they did not land too many punches.  50% run to start drives is well below their routine, but they saw what was happening.  They just didn't connect on those passes.
2013 Total: 176 Drives - 84 Run/92 Pass - 47% Run
2012 Total: 173 Drives - 76 Run/97 Pass - 44% Run
2011 Total: 181 Drives - 79 Run/102 Pass - 44% Run
* This statistic doesn't count the 1-play kneel down drives.
SHOTGUN SNAPS
Shotgun snaps are fine on 3rd Down and in the 2 minute drill. But, we track this stat from week to week to make sure the Cowboys aren't getting too lazy in using it. They are not efficient enough to run it as their base, and with a 15%/85% run/pass split across the league, there is no way the defense respects your running game. When shotgun totals are high, the Cowboys are generally behind, scared of their offensive line, or frustrated.
2013 Total: 566/945 - 59.8% Shotgun
2012 Total: 565/1038 - 54% Shotgun
2011 Total: 445/1012 - 43.9% Shotgun
TOTALS BY PERSONNEL GROUPS (Before you study the data below, I would recommend that if the numbers for the groupings are unfamiliar, that you spend some time reading a more expanded definition of the Personnel Groupings here.)
 * - Knee Plays are not counted in play calls.
Look at 12 personnel - both in shotgun and under center.  When the Cowboys went to 2 TEs, it looked like the Cardinals were prepared to deal with it quite well.  13 snaps for a grand total of 20 yards!  That is easily the least productive game I can recall from that grouping.
PLAY-ACTION PERFORMANCE
Wk 1: 1/5, 9 Yds, 3 INT, 1 FD
Wk 2: 4/5, 39 Yds, 1 Sack, 2 FD
Wk 3: 3/3, 88 Yds, 1 TD, 2 FD
Wk 4: 6/8, 76 Yds, 1 TD, 4 FD
Wk 5: 2/4, 38 Yds, 1 Sack, 2 FD
Wk 6: 1/4, 47 Yds, 1 Sack, 1 FD
Wk 7: 3/5, 55 Yds, 1 Sack, 2 TD, 1 FD
Wk 8: 5/6, 92 Yds, 1 TD, 2 FD
Wk 9: 1/1, 1 Yd
2014 Total: 26/41, 62.5 Cmp%, 445 Yds, 5 TD, 3 INT, 15 FD, 4 Sack
Play-action was not much of a factor
BLITZING Romo/Weeden Pass Rushers Against Dallas - 36 Passes vs Arizona

Wk 1: SF Blitzed Dallas 1/40 - Blitzed 2.5%
Wk 2: TEN Blitzed Dallas 12/33 - Blitzed 36%
Wk 3: STL Blitzed Dallas 11/23 - Blitzed 47%
Wk 4: NO Blitzed Dallas 11/32 - Blitzed 34%
Wk 5: HOU Blitzed Dallas 11/42 - Blitzed 26%
Wk 6: SEA Blitzed Dallas 5/33 - Blitzed 15%
Wk 7: NYG Blitzed Dallas 5/25 - Blitzed 20%
Wk 8: WAS Blitzed Dallas 21/40 – Blitzed 52%
Wk 9: AZ Blitzed Dallas 13/36 - Blitzed 36%
2014 Total: Opponents Blitzed Dallas 90/304 - Blitzed 29%
2013 Total: Opponents Blitzed Dallas 210/616 - Blitzed 34%



Thanks to John Daigle for his work on the charts and graphs.
SUMMARY AND LOOK AHEAD:  
I have now spent the better part of two days saying harsh things about the performance of Brandon Weeden. That is a difficult position to take on a guy who is making 600k (or roughly 500k less per year than the Cowboys pay their long snapper, LP Ladouceur).
Weeden can play better and it argues loud and clear that if Tony Romo is lost for the season, this thing is in really bad shape.
But, here are some realities. The running game is really good and have a powerful RB in DeMarco Murray who was battling his tail off on Sunday and actually made some nice runs with very little support on Sunday. They also have invested a lot in their supporting cast. That, coupled with my concern about Romo being fit for the stretch run is why if I was king, I would have left Romo in Dallas to get right for the final 6 in a stretch that begins on Nov 23 at Giants Stadium.
However, I know that is not a popular stance and I know that the Cowboys - Jerry Jones, Jason Garrett, Tony Romo, and the Cowboys medical staff - seem to all feel that he can play against Jacksonville on Sunday in London. Therefore, the onus now goes squarely on the pass protection and the offensive scheme to limit at all costs the number of hits Romo takes in the next few weeks. They must understand that a punt is ok in most situations to preserve their QB1 health. They know what happens if he is missing and therefore, treating a 3rd down in the 2nd Quarter has to be done with caution and common sense rather than asking him to take a big hit. If the game is on the line, you can adjust accordingly, but being reckless with his back is silly if you don't calculate the risk/reward ratios constantly.
Of course, in the NFL, the opponents know this and use it against you. They will challenge with blitzes and sit on his escape throws and ideas. This chess game could lead to a less productive offensive performance, but the numbers are for fantasy players, not the Cowboys. They must play with a very conservative posture early on and not play into the urge to go right back to being Superman.
I anticipate Romo will be playing hurt on Sunday, and I think the Cowboys need to plan on altering their ideas to accommodate him - not risk him again. Because i think you can clearly see above that if you don't have #9, then you don't have a realistic chance to compete.
If we have learned nothing else from that poor Sunday offensive performance, at least the Romo appreciation level received an upgrade.

No comments: