Thursday, December 01, 2011

Cowboys Mailbag 12/1/12

Today, we will dip into some emails after I unveil a new feature to my blogs here that I admit is awful experimental.

It is a "passing chart" based on where Tony Romo is going with his passes. I am not so interested in where the play ended, but rather where the pass was intended and completed. Blue passes are completions, red are incompletions, and purple are the two interceptions from Thanksgiving Day.

Ideally, this will tell us things on a week to week basis such as: Where Tony Romo has tendencies to throw the ball (right handed QBs tend to throw right far more often, but Romo may not be as bad as many). How aggressive the game plan was in a given game (often times we rate aggressive game plans on a raw number of pass plays. However, this is a bad idea since not all pass plays are created equal. In the Jason Garrett offense, he substitutes short passes for run plays when he has no faith in a run between the tackles as we saw in New England and Minnesota the last time the Cowboys were in both places). It will also show us where they were attacking and what a defense was conceding.

The yellow line is the line of scrimmage for each play. We attempt to mark the release point of Romo and the catch point of the receiver. Hypothetically, this will show us what throws Romo can make versus Kitna. It can also show us how they are using Dez Bryant if we make a chart for Bryant catches for the season (as we will do when the season ends).

Anyway, it is experimental, and Tim Krajewski deserves the credit for the actual building of my concept. We will continue to tweak as we go, but below, please find the first public appearance of the Romo passing chart. Feel free to offer feedback so we can get it right. And try to ignore the yard markers, as they are only a reference point rather than specific information on where the play was located. The yellow line is the line of scrimmage.

Romo versus Miami, Week 11. Blue=completions. Red=incompletions. Purple=interceptions.



OK, now, on to our email:

I really enjoy your articles on Cowboy’s Personnel. Great reading. Here’s my question.

At times the Cowboys have different personnel in their groupings, i.e. sometimes it’s Murray and sometimes Jones or Choice. Same with tight ends and receivers. How do they communicate these changes to the players in such a rapid fashion that the players get in and out on time?

Thanks
Dennis Dusek
La Vernia, Texas


Great question, Dennis. And it shows the complexities of how the NFL really operates. What he is asking is about specific players inside personnel packages. For instance, "12" personnel (1 RB, 2 TE) can mean Murray, Witten, and Bennett or it can mean Murray, Witten, and Phillips. To you the fan, this might not mean anything. But to a defense, they will treat each tight end differently in coverage ideas. They know that Bennett is a guy that could perhaps run by a linebacker (although we seldom see it - just imagine that he can), whereas Phillips is a player that you would not worry about shaking loose from a LB. Or, "21" personnel (2 RB and 1 TE) could mean Murray and Fiammetta or it could mean Murray and Jones. Well, a defense better understand that while Fiammetta is never going to do anything downfield, Felix Jones can take his man on a wheel route and turn it into a big play if you are not properly defending him.

The offense has sub packages in each personnel grouping on the sideline. Garrett will yell out "21" if it is standard "21", but "21 mini" if it is Jones and Murray. He will have different names for each group and that will identify which WRs or TEs he wants on the field. And this gives the defenses one more thing to worry about. Hope that helps.

========


Hey Bob,

I thought you would enjoy this one: Dallas is first in the league in sacking the QB on the road and last in the league in sacks at home. I am astounded by this stat. How in the heck is this Rob Ryan machine not generating sacks right here at home with the crowd on their side and all of the other advantages of a home game? It amazes me that no one has brought this up before (when I read this, I had a vivid image of Stafford sprinting out of the pocket to his right and throwing the ball away to avoid the sack. It felt like he did that about 10 times in the first half of that game--never took a sack).

roger light


Through 11 games, the Cowboys have 30 sacks. This places them in 7th place in the NFL in team sacks (Baltimore 1st, 38) and 4th in the NFC (Minnesota, Washington, and NY Giants have more). But, yes, quite oddly, they have 20 in 5 road games (1st in NFL), and just 10 in 6 home games (26th most). His email was sent before the Thanksgiving game, so the Cowboys are no longer in last place at home as they bagged 4 more against Matt Moore.

Why is this? That is a wonderful question. I do wonder about your theory that the crowd "is on their side" as I find the new stadium to be about as quiet a home stadium as there is around the league on game day. I certainly know they had the best intentions when they build the gigantic edifice, but I believe the high prices and the high comfort puts people in a slightly "upscale" frame of mind. Add to it the magnetic tv screen above the field and somehow the Cowboys have created a fan experience that seems to hypnotize their own fans into a docile state.

To compare it to a true home stadium sound advantage of a New Orleans or similar enclosed stadiums around the league seems a silly waste of time. That being said, the pass rush should not be that related to the sound when it comes to simply equaling the production on the road where they get no help whatsoever. To average 4 sacks per game on the road and less than 2 at home is a confusing number that may just be an example of sample sizes being so small. It is certainly worth keeping an eye on.

========

I understand the well-documented impact of Tony Fiametta. I’d like to understand your thoughts on the impact of Miles Austin? In particular, upon Austin’s return, do you think they’ll reduce their use of “12” personnel because they’d rather have Austin, Robinson and Bryant on the field to stress the defense, or does the lack of a fullback still require them to use 2 tights? Or could they use more “02” personnel groupings to get the 3 wides out there, but then lose Murray on those plays?

I appreciate your coverage of the Cowboys … the most substantive coverage out there without all the “noise.”

Best regards,

Bill


Thanks, Bill. This is one that I would love to ask Garrett about. Many of the best offenses in the league use "11" personnel (1 RB, 1 TE, and 3 WR) as their standard offense. The Cowboys, since I have been charting them in 2008, have almost never used 11 personnel from under center. They do use it extensively on 3rd Downs and the 2-minute drill, but that is exclusively out of the shotgun. Of course, in this situation, the Cowboys run about 11% of the time, and that is almost always a play like a 3rd and long draw play where they are just trying to get a few yards for the punter.

Why do other teams love "11" personnel? Because, it forces the defense to play nickel (5 DBs) or dime (6 DBs) immediately on 1st and 10. And this then means that their running game has far more space because to run nickel, a defense has to take off a linebacker and replace him with a guy who doesn't weigh 200 pounds. The Cowboys have never liked this, and I have never honestly understood why. In 689 offensive snaps this season, they have run "11" personnel 13 plays total. They obviously don't believe in it, even though as you astutely point out, it seems the Cowboys have the perfect personnel for such a strategy. It may show us that although the OL is playing better, he just doesn't trust them yet. To play "11" as your base, you better believe that your OL can protect your QB. "11" will require plenty of "scat" protection, which means just your 5 OL protecting with no help from TE or RB. So, the answer might be that simple.

========

Bob,

What in the wide world of sports were the Dolphins doing on that play in the 1st Quarter with the wacky formation? And how is that legal?

Thanks,

Ted


Ted, I am so happy you brought it up. This play is something I could study all day.



The play can be seen on video here and shows that the Dolphins were lining up a center and then 7 players to his right. Brandon Marshall goes in motion from the left sideline to the right and that leaves only Matt Moore and Reggie Bush in the backfield. By nature of the rules, the Dolphins are required to have 7 on the line (which they do) and each end is an eligible receiver. As I understand it that means that the Dolphins center is also an eligible receiver (I promise I have never seen that before). Here is another look at the formation:



It sure looks like the Cowboys just ignored the wacky formation and lined up the defense on the middle of the line and not the football. I wonder what happens if the Dolphins try to attack DeMarcus Ware with an eligible center, Reggie Bush, and Matt Moore? It seems the Cowboys are all shifted so far that if they can figure out a way to outflank Ware, this could be a big play to the weakside, but of course, that requires them to attack the most athletic player in the game.

Anyway, Sean Lee blew the play up, and the Dolphins never tried it again, but I was thoroughly fascinated.

========

Bob-

The games that Fiametta have played are the games the Cowboys have run well. You and a lot of others have noticed that. But you've always spoken with great respect of Football Focus and Fiametta grades rather poorly with them. Thoughts?

JDRB


This is an excellent discussion to touch on. Fiammetta is a name that is getting a ton of play this season, and I feel that I am responsible for helping to get it started by talking about the great production of the run game and the offense when he is on the field.

But, is it all because of him? My friends at ProFootballFocus.com have him ranked as the 18th best FB out of 31 in the league. That is hardly great evidence that the Cowboys have an elite player here. But, I think it is possible that we are having two different discussions. Are the Cowboys playing better with him on the field? Yes. Is it precisely because of him? Maybe not. There are so many moving parts during a given football play that I think it would be absurd to say that any success out of "21" personnel can be linked back to one, largely anonymous player. However, there must be some connection because when John Phillips tries to be plugged in on many of the same plays, the plays are not as productive with everything else seeming to be the same. With Fiammetta, it is possible that the defenses are reacting differently. Perhaps they are more likely to sit on a run play and move a safety into the box, but with Phillips back there, they assume it could be a play action pass.

I would be careful to say all of this is because of Fiammetta. I would not lock him up to a long term extension because of our statistics that show the team runs best with him there. But, I also don't want to ignore the evidence. He seems very competent as a lead blocker. But, he is also not blocking 3 players at a time.

The best example of individual performance versus team performance is the play of a Left Tackle on a Pick 6. The play was a disaster and might have lost the game. But, the left tackle did his job perfectly and might have been given the maximum grade for his work. The point being that when things go right, someone still failed at their job, and when things go poorly, someone did great. It is simply the fabric of football.

2 comments:

Doctorjorts said...

In the old swinging gate formations, the center would frequently be the last player on the line of scrimmage, and therefore an eligible receiver. I wonder what the Cowboys would have done if Brandon Marshall lined up as the center?

scottmaui said...

Bob, this is a great Cowboys post. First off great questions, and great answers.

1) Love the passing chart! Simple idea that I wonder why I've never seen before. Can't wait to see more of it, compare games and QBs, etc. I can imagine it can tell us a lot. So much more information easily conveyed than just numbers or the simplified number of passes to each large field regions they sometimes show.

My one suggestion would be to somehow convey position of the throw in relation to the snap as well as the field. e.g. In this graph, we can't distinguish a pocket pass snapped on the left hash, from a slide/spin out of the pocket to his left on a play that started in the middle or right hash. I don't know how to convey that graphically, other than to make a separate graph that adjusts the origin point of the pass in relation to the position of the snap instead of the position on the field. Only then could you get a sense of where he's throwing from in relation to the pocket, which is a whole different set of information than where he's throwing from on the field. And useful in some different ways.

re the 11 personnel, I think one big reason the Cowboys haven't used it that much this season is because they've seldom had 3 healthy (and competent) WRs playing at the same time. There may be other factors like the versatility of the TEs that he likes, but his ability to call 11 personnel has been hamstringed uh literally. Between Dez's thigh, Miles 2 hamstrings, and Laurent's adjustment period I'm not sure but have they all be on the field together for two full games even?

That Dolphins play was the ultimate in unbalanced line. Eligible center? Scooby Doo "Hhuuhh??" {rubbing eyes} moment. The fact that Sean Lee not only made the tackle but completely blew up the play and Bush was pretty hilarious. It's like, okay, maybe that's not the new wildcat!

The Fiametta thing really is interesting too. He's such an enigma. The vegan thing and mysterious illness on top of a run game that for whatever reason seems to work much better with him in there despite however he might grade. I mean folks forget Felix had 115 yards behind Fiametta in week 3 vs the Redskins. Murray's best games were with Fiametta. The strength of the running game actually seems to correlate more with him than the back. There's so many moving parts to the running game that it's very hard to isolate how much each part is responsible for the success or failure of the whole... the back, the oline, the lead blocker, the defense, and the game situation. So it's very hard to separate the signal from the noise and see if there's something the grades are missing, or if there's just a happenstance from sample size. The way I've heard it described and I do see some of this compared to Phillips, Fiametta is a very fluid blocker who can adjust to clean up the trash if another blocker fails or the play doesn't unfold as planned. But the biggest thing is maybe he and Murray just have a good rapport so Murray can read his body language and know where the block is going to be and how to cut behind him to find the lane, like which side he's going to shield off instead of just stoning his guy in the middle. Just speculation but maybe there's something to that, intangible but real. Whatever it is, I'm just delighted we have a functional run game now, with or without Fiametta, and hoping he can come back soon and make it even noticeably better!