The Curious Case Of The Cowboys Offense

The Dallas Cowboys came into the 2011 season with a high-powered offense behind a questionable offensive-line, and with a new coach and scheme for a questionable defense. Yet leading into Week 7 with a 2-3 record, it seems the Cowboys offense has let down the fans more often in the defeats than has the defense. I began to wonder what kinks must be worked out before the Jason Garrett offense can resemble its dynamic predecessor...and in the process found myself tumbling down a rabbit hole into a dark and curious tale.

I am still unsure of what this all means and whether any real clarity, some kind of fairytale metaphor or moral, can be drawn as a conclusion. Certainly the 2011 Cowboys offense must become more consistent. While both wins have been come-from-behind victories with clutch fourth-quarter performances, all three losses have come after the Cowboys had the lead and couldn't close out the game.

An effective ground game would obviously help. The Cowboys currently rank 28th in rushing average and 27th in rushing yards. But while clearly a culprit, I wondered to what degree the absence of effective rushing actually affects Garrett's offensive designs. The other factor that has been an obvious culprit has been the inconsistent performances of Tony Romo. These two trains of thoughts are what led me down the rabbit hole and into a wonderland of stats, personal deductions, opinions, and theories that led me to even more questions and curious possibilities.

If you want to see how deep the rabbit hole goes, click on the blue link...

Theatrical and literary references aside, the difficult task remains the structuring and presentation of all the information and ramblings into a coherent story. Let's begin where every offense starts.

The Quarterback

This is not about questioning Tony Romo's ability. He is a very capable quarterback who, like everyone, is susceptible to mistakes. This is not about questioning his leadership, which becomes more apparent every year, nor his competitiveness, which has been proven plenty of times. Personal favorites include asking about the pass after the broken collarbone in 2010, and playing through a punctured lung and fractured rib in 2011.

Certainly the offensive-line and receivers are instrumental to a quarterback's success, but I would claim both have been more than adequate this season. Though without a rushing attack, Tony Romo has had better pass protection than last season and talented receivers. Certainly the absence of Miles Austin was felt in the games versus Washington and Detroit. Certainly Romo's injury has been a pain. But there seems to be more going on.

Last week against the Patriots, it seemed to me that Tony got timid after throwing an interception on the opening drive. Romo seemed to be playing it safe for most of the game, though he did play looser at the start of the second half. In what would be a scary proposition, it appeared to me that Tony Romo might have been questioning his inner gunslinger.

Now, I still believe Coach Garrett must wrangle the gunslinger in certain situations, but it is vital that Tony Romo doesn't doubt his abilities as a gunslinger. He must make better decisions in certain situations, like not forcing the difficult throw when the team has over a twenty-point lead, but he must believe he can make every throw. Worrying too much or questioning one's ability and not pulling the trigger when necessary could prove disastrous for Romo...the difference between remaining a good quarterback instead of becoming a great quarterback. It seemed against the Patriots Tony was not testing the defense downfield very often. In fact, if you read the NFL play-by-play, there are very few passes not described as "short pass". While some of those short throws led to broken tackles or yards after the catch, Romo did not often attack downfield or try making difficult throws in tight windows.  

In the last three games, the average gain per pass completion has been considerably lower than in the first two weeks. I thought perhaps the rib injury was the biggest factor, but it seems unlikely considering Romo's performance against the 49ers, where many deep completions occurred after the injury. (Stat shown is only Romo's avg/comp and does not include any passes by Jon Kitna)

Of course, it could be Garrett changing his play-calling to compensate for the injury or even that the swelling has been more troublesome than the pain at the moment of the injury. Perhaps Garrett has changed the play-calling to a certain extent, but Romo is still the one making the reads and pulling the trigger. Finally, it could have been the opponent's defense forcing short completions and Romo merely taking what the defense allows...but isn't that where the gunslinger takes more than what is given?

Because of the difficulty in tracking every route in a passing attempt and the pattern of every completion this season, I decided to see how often Tony threw to players that weren't wide receivers or Jason Witten, a probable sign of Romo checking down to shorter routes. I separated Witten from the "Other" category because he often lines up in the slot as a receiver, and often runs routes downfield and/or in tight coverage. It appears Tony checked-down quite often the last two games.

Against New England, Romo completed short passes to "others" nine times, and quite a few of the completions to wide receivers were on short routes as well. I also find it interesting that the Patriots managed to keep Witten to only four receptions (lowest this season). While many defenses key on Witten, the reason why most of his receptions are contested and through tight windows, fewer receptions are also a sign that Romo avoided making those tight throws in the middle of the field. Also interesting to note, in the Detroit game, six receptions to the "others" occurred after Romo's second interception near the end of the third quarter. Meaning in just over one quarter, (and after eventually losing the lead) Romo threw check-downs to the "others" as many times as he had in the entire games against the Jets, 49ers, and Redskins.

We will never know for sure without being a member of the Cowboys organization and in the huddle and/or film room, but it seems Tony Romo may be letting his 2011 interceptions affect his play. It is a very troubling thing to consider.

To Run Or Not To Run - That Is The Question

Perhaps I was not quite done with the literary references, but the fact remains, the Cowboys have one of the worst rushing offenses in the league. The question remains: how much does the lack of rushing impact the offense, and should Garrett compensate by trying to run less? Personally, I think running the ball is vital to any offense (even those that pass first) and the Cowboys must find a way to improve the effectiveness of the ground game. It also appears to be an important aspect to the Jason Garrett offense.

As I began thinking about the explosive Garrett offenses of the past, I wondered if the running game was a necessity. As Romo began his career and quickly surpassed many Cowboys greats in various passing statistics, did Garrett really use or need the running game?

This may come as a surprise, but it appears Garrett's explosive offenses have a running game of virtually the equal caliber as the passing attack. When comparing where the team averaged in league rankings for passing/rushing average and passing/rushing DVOA, I was surprised at the minor differential.

While the pass/rush averages differ by more than two, it's surprising to find Garrett's offenses average in the Top 10 in rushing average. Meanwhile, even more surprising is the passing and rushing DVOA rankings are within one spot. While most Cowboys fans recognize and remember the high-powered passing attack in Garrett's offenses, it appears the Cowboys rushing rankings are nearly as impressive.  

Don't let this trip turn bad; this is not a debate about which is more important, passing or rushing. Different offenses rely on the two dimensions of an offense for different things. But with the Cowboys currently ranking dead last in rushing DVOA, it is clear Garrett and Coach Houck must find a way to mesh the abilities of the mobile Yuglies and the versatile running back rotation to improve the ground game and help the 2011 offense become more consistent. Does it really matter if the Cowboys can't run the ball? Well...never before has Garrett's offense ranked below the middle of the pack in rushing DVOA or rushing average.

(DVOA rankings from FootballOutsiders.com, and pass and rush averages/rankings from NFL.com)

This curious case of the Cowboys offense continues to provide for a strange trip.

The Cowboys have not been considered a dominant rushing team in quite a long time, yet it appears the Cowboys have always ranked in the top half of the league in rushing average and DVOA since Garrett has been the play-caller. It is also interesting to see that the rankings for rushing average and DVOA have a smaller differential than the Cowboys passing average and DVOA rankings.  This is one reason I like studying the DVOA rankings. The advanced statistic known as DVOA (defined here) tracks every play and considers a variety of variables to provide a broader picture of passing and rushing efficiency compared to league averages. While I prefer to look at averages per attempts over total yards, again because I consider it a better stat for understand a team's efficiency and effectiveness, DVOA can provide even more insight. (Author's note: I also like to compare rankings as opposed to just the averages or percentages to get a clearer picture of efficiency when compared to other teams in the league).

Looking at the Pass Avg. versus Pass DVOA rankings, it appears that when the Cowboys passing offense is elite by league standards, the averages and DVOA are very similar. However, when the Cowboys passing averages were not as impressive, the DVOA rankings were considerably worse. This could be a sign that while completed passes were still longer completions than most of the other teams in the league, the adjusted value over average take a hit because of negative plays in the passing game. Seeing that the rushing averages are also lower in those seasons makes me again wonder about the possible correlation in Garrett's offense between effective rushing and an efficient passing game.

Could the hampered rushing attack in 2008 and 2010 be a reason for the offenses lowered passing DVOA? Vice versa?  A bit of both? While only looking at statistics will never provide a divine truth, it does seem apparent that the Cowboys passing efficiency and rushing ability are connected. It would then lead to the idea that the poor rushing attack in 2011 is also putting a strain on the 2011 passing DVOA. Could the interception rates and rushing averages be interconnected? Is Romo more confident when the offense is clicking both through the air and on the ground? Could defenses be easier to pass against when the Cowboys are rushing well? Could the offensive-line make less mistakes and the offense be forced into fewer sacks and turnovers when the rushing attack is working well?

As I said, my trip through the rabbit hole left me with more questions than answers. It left me wondering about the curious case of the Cowboys offense under Jason Garrett, known for passing yet ranked almost as well in rushing. An offense that clearly has more ability than has been on display in 2011. It remains to be seen if Garrett will continue to try and find an effective rushing attack while the young offensive-line (about to see the third different starting group in seven games) continues to fight through growing pains. It remains to be seen if this will help or hurt the passing game. It remains to be seen if Tony Romo can continue to attack downfield while also limiting turnovers with smart decisions.

It remains to be seen if these ramblings, questions, and theories prove to have any validity. But hopefully the Sunday matchup versus the poorly ranked St. Louis Rams defense will be a chance for the Cowboys 2011 offense to find greater rhythm and consistency.