Texans Have Been Working the Play-Action to Death

[CloakNNNdagger]

And it will certainly be revived this year. An interesting article that has 2011 patterns that may surprise even some Texans fans.

12 Jun 2012
2011 Play-Action, Offense

by Rivers McCown

Continuing our look into the many, many tables to be completely revealed in Football Outsiders Almanac 2012, today we'll look at play-action passes. Our game charting project tirelessly catalogs the plays that are run over the course of an entire season, marking (among other things) play-action looks. A massive thanks to those involved, from someone who has been working for the project since the 2007 season: your effort really helps to further the statistical revolution in football.

We've seen a pretty settled pattern of play-action passes for the past couple of years. NFL teams ran play-action 17 percent of the time during the 2008 season, and that number has been gradually ticking up from 18 percent in 2009 and 2010 to 19 percent last season. Last year, the Ravens ran play-action more than any other team, but this year they actually cut back on it and fell out of the top 10. This year's champion is another of the clubs you'd expect: Houston.

The Texans went from a team that loved the play-action to a team that needed to run play-action once T.J. Yates took over at quarterback. In fact, of Yates' 150 non-penalty dropbacks last year, 59 of them took place off play-action; nearly 40 percent of his throws. That's a big part of the reason why their 33 percent figure is the highest we've recorded over the past five years. However, they were running it more often even with their regular offense: 93 of Matt Schaub's 308 dropbacks were play-action, roughly 30 percent.

There's a clearly-defined tier of teams that run the most play-action. Houston, Denver, St. Louis, and Washington have all finished in the top 10 in play-action pass percentage the last two seasons. That's a Shanahan team, a Shanahan-disciple team, Shanahan's old team, and ... the guy who replaced Shanahan in Denver. (Pat Shurmur was the St. Louis offensive coordinator in 2010, though.)

The team that ran play-action the least last year? That'd be the Buffalo Bills. Not really a surprise given Chan Gailey's pre-disposition towards the spread offense -- faking a handoff does require a little bit of time, after all. Buffalo was tied for last with Detroit in 2010, as well.

The list of teams that has been in the bottom 10 in percentage of play-action passes in each of the last two years: Bills, Lions, New York Giants, Tennessee, and Arizona.

The most effective play-action team by DVOA last year was New England, but they actually didn't run much play-action at all, and the increase over their normal passing DVOA wasn't very pronounced (just 19.8%). The team that improved the most using it? Denver went from a -28.9% DVOA on regular passes to a 39.2% DVOA off of play-action. It's almost like they had a quarterback that was known for not being an effective pocket passer back there.

The team with the worst play-action DVOA was, believe it or not, Jacksonville. Of course, the Jaguars had one of the worst passing offenses in the NFL last year anyway, but they ran play-action 22 percent of the time to try to take some pressure off their rookie quarterback. Blaine Gabbert, who mostly operated out of the spread in college, led the team to a -60.9% DVOA off play-action, a little over 35 percentage points behind than the second-worst team, Tampa at -25.1%. Anyone who watched Gabbert last year knows that footwork isn't exactly the most polished part of his skill set, so it would make sense that putting him in motion could lead to some poorly-placed balls.

However, the biggest difference was not in Jacksonville, but in New York. The Giants finished with the fourth-best passing offense in the NFL, but said offense actually went from excellent to poor with the threat of a handoff. Big Blue had a 41.3% DVOA last year on pass plays without play-action, but that dipped all the way down to -21.1% on the fake. The Giants didn't run the ball much at all, and didn't run much play-action when they did. One anecdotal reason for this is that the Giants were so much more effective with three receivers on the field: as we showed in our preview of the strategic tendencies tables, the Giants tended to be most effective out of 11 personnel (i.e. one back and one tight end), and tended to use that more than any other formation. Just 18 percent of their plays out of that formation were runs. Naturally, without much on tape of them running from the formation, New York didn't have much reason to play-fake out of it either. The Giants ran a play-action pass with three or more wideouts on the field just nine times all season.

Below is the play-action data from 2011 for all teams, in descending order starting with the teams that used the most play-action. The columns in green represent performance with play-action on actual passes only, with sacks, scrambles, and intentional grounding removed.

NFL play-action data from 2011 [see chart at bottom of link page (http://footballoutsiders.com/stat-analysis/2012/2011-play-action-offense)

[Playoffs]

And it will certainly be revived this year. An interesting article that has 2011 patterns that may surprise even some Texans fans.Interesting read. Really surprising how much the Giants changed their offense play selection. I think we'll need to do some of that, to shift back towards the norm. But do we have the tools & will Schaub trust them? Or better, will Dennison even think to do so?

[CloakNNNdagger]

Interesting read. Really surprising how much the Giants changed their offense play selection. I think we'll need to do some of that, to shift back towards the norm. But do we have the tools & will Schaub trust them? Or better, will Dennison even think to do so?

When you try to be too unpredictable, you can:kitten: become predictable.

[b0ng]

This shouldn't really be news to anybody who has watched the Texans regularly for the last 5 seasons.

[GP]

As long as you've got a great RB and at least 1 or 2 weapons at WR/TE....you can run the playaction 30 or 40% and be successful with it.

DL are affected. LBs too. As are CBs and Safeties. They're all watching the drop back and wondering if it's gonna' be run or pass.

Once your run game is nullified, by injury to your good RB or by just not having a reliable run game at all, then the playaction pass doesn't work as well. There's no reason for defenses to respect the run; which means they'll play the pass and go after the QB.

The Colts last year--they had no run game and therefore their QB was always under pressure.

[chenjy9]

IMHO there is nothing wrong with overusing play action, because the whole point is to get the other time to bite down hard on it and hit them with that huge pass. As long as the team has the proper tools to run a successful play action offense (competent QB, competent WR, and competent RB) it will work and keep teams guessing.

[b0ng]

As long as you've got a great RB and at least 1 or 2 weapons at WR/TE....you can run the playaction 30 or 40% and be successful with it.

DL are affected. LBs too. As are CBs and Safeties. They're all watching the drop back and wondering if it's gonna' be run or pass.

Once your run game is nullified, by injury to your good RB or by just not having a reliable run game at all, then the playaction pass doesn't work as well. There's no reason for defenses to respect the run; which means they'll play the pass and go after the QB.

The Colts last year--they had no run game and therefore their QB was always under pressure.

Hell the playaction worked pretty good when we were running Chris Brown, Steve Slaton, and Ryan Moats in 2009. When it comes to drawing up PA passes, I don't think there is anybody better at it than Gary Kubiak. It's a shame most people on this very board could pick better DC's than him though :(

[thunderkyss]

As long as you've got a great RB and at least 1 or 2 weapons at WR/TE....you can run the playaction 30 or 40% and be successful with it.

DL are affected. LBs too. As are CBs and Safeties. They're all watching the drop back and wondering if it's gonna' be run or pass.


So then we look at Denver & Jacksonville. Great RBs..... eh. subjective. Good enough to move the chains, definitely.

But when you're not scared of the QB keeping the ball & doing something with it.... does it really matter who the receivers & RBs are?

[thunderkyss]

Hell the playaction worked pretty good when we were running Chris Brown, Steve Slaton, and Ryan Moats in 2009. When it comes to drawing up PA passes, I don't think there is anybody better at it than Gary Kubiak. It's a shame most people on this very board could pick better DC's than him though :(

I call it the Madden syndrome.

Greg Williamson, Vic Fangio, Monty Kiffen, Anyone who was ever the DC in Baltimore.....

It's easy to throw names out there. But to get them to come to Houston, much less sign on the dotted line is a different thing altogether.

[Insideop]

IMHO there is nothing wrong with overusing play action, because the whole point is to get the other time to bite down hard on it and hit them with that huge pass. As long as the team has the proper tools to run a successful play action offense (competent QB, competent WR, and competent RB) it will work and keep teams guessing.

Don't forget the O-line. They have to "sell the run" on the passing plays, and I think I heard last season the Texans O-line was better than anyone at doing that. Just hope the right side of this season's O-line, whoever they are, can do the same!

[infantrycak]

Don't forget the O-line. They have to "sell the run" on the passing plays, and I think I heard last season the Texans O-line was better than anyone at doing that. Just hope the right side of this season's O-line, whoever they are, can do the same!

I think running the ZBS helps with selling the run. It, particularly on the extreme stretch, looks very distinct. The Texans often take a much more aggressive angle off the snap than other teams.

[GP]

So then we look at Denver & Jacksonville. Great RBs..... eh. subjective. Good enough to move the chains, definitely.

But when you're not scared of the QB keeping the ball & doing something with it.... does it really matter who the receivers & RBs are?

Absolutely. It's why MJD is racking up yards with Jags but not seeing his t-e-a-m go anywhere with it...because they have a consistent problem at QB (and WR). They have a TE who drops passes too much.

The thing that hurt TJ Yates, IMO, was that he wasn't able to do what Schaub had been building on for many years: That instinct of knowing where to go at the snap...that knack for finding Owen Daniels on a seam route or deep zone read, etc. Or knowing when he had Walter who had snuck out from coverage due to Foster play fake or AJ and OD drawing CBs off him.

That's the thing that I wonder we could see Yates get better at with OTAs and a full camp this year. Which means reps with the 1's (building that knack that a Kubiak QB has to develop).

But still, the bread and butter is the run game. Without it, a guy like Yates becomes instantly mortal. With it, Yates has time he normally wouldn't have the luxury of having on a pass play.

Don't forget the O-line. They have to "sell the run" on the passing plays, and I think I heard last season the Texans O-line was better than anyone at doing that. Just hope the right side of this season's O-line, whoever they are, can do the same!

Good point. I didn't mention it, but you're 100% right.

It takes the whole organism for things to work. Your o-line has to be performing at a HIGH level. Super high. If they're not, the whole thing bogs down.

Gary has got a competent o-line, which I think will be even more competent with the absence of Winston. Everybody who comes to Houston now knows they can crack the starting lineup if they play hard enough and consistent enough. Jacoby is gone. Winston is gone. Two players who were being paid nicely who really were not earning it (to the degree they were being paid, at least). IMO.

[badboy]

Any play that works is a good play. Glanville could still a coach if that one goofy play had worked near his own end zone.

[TheMatrix31]

Good. Play-action the **** out of anyone and anything.

[Norg]

play action only works if teams respect your run game and the oline is in sync if not they will just blow the play up

i for one would like to see them practice more of a 4 WR set spread them out and go type of plays

and the oline acctually practice on picking up blitzes better and pass protection esp when your out numberd

and matt acctually knowing how to slide to the left or right in the pocket

[ckhouston]

With the dominance of our running game, we should see PA open up a few big time down the field plays this year.

[NastyNate]

I think running the ZBS helps with selling the run. It, particularly on the extreme stretch, looks very distinct. The Texans often take a much more aggressive angle off the snap than other teams.

THIS! PA isn't predicated on your run game so much as it is on the lateral movement of your linemen. You could put Chris Ooglyboogly back there and not have any drop off in PA production based on the effectiveness of the line. ILB's hone in on a guard's first step to read a play, if they continually move laterally it becomes much more difficult to diagnose.

[Texan_Bill]

THIS! PA isn't predicated on your run game so much as it is on the lateral movement of your linemen. You could put Chris Ooglyboogly back there and not have any drop off in PA production based on the effectiveness of the line. ILB's hone in on a guard's first step to read a play, if they continually move laterally it becomes much more difficult to diagnose.

Sweet!! When did we pick him up???


.............Championship!!

[ckhouston]

PA isn't predicated on your run game

If you arent able to run the ball, then PA doesn't work. If they arent worried about you gaining any yards on the ground, then why would they bite on the PA? Agreed that the line is also a key, but first you have to establish the run as a threat ... then you can pick them apart with PA.

[b0ng]

If you arent able to run the ball, then PA doesn't work. If they arent worried about you gaining any yards on the ground, then why would they bite on the PA? Agreed that the line is also a key, but first you have to establish the run as a threat ... then you can pick them apart with PA.

Except our playaction worked fine in 07 and 09 when we had non respectable rushing attacks.

[The Pencil Neck]

If you arent able to run the ball, then PA doesn't work. If they arent worried about you gaining any yards on the ground, then why would they bite on the PA? Agreed that the line is also a key, but first you have to establish the run as a threat ... then you can pick them apart with PA.

The facts don't support you.

The believability of the PA fake sells it. If you make your play action look exactly like your normal running play, then the defense will read run and react as though it's a run. And it's going to do that regardless of how good your run is. That's what happens if you have good play design and consistent action by your entire team.

When we had bad running games, our play action fakes still worked because the plays were designed to sell the run. As opposed to those weak-assed fake draws where the QB doesn't even really look at the RB that so many teams call PA.

[Rey]

If you arent able to run the ball, then PA doesn't work. If they arent worried about you gaining any yards on the ground, then why would they bite on the PA? Agreed that the line is also a key, but first you have to establish the run as a threat ... then you can pick them apart with PA.

I think when you have a great running game your play action works better, but with the style of offense we run with the stretch action our play-action will be effective regardless.

But when you have a guy like Arian back there people are going to get open more. And then what is the killer about our offense is when the passing game gets going the LB's can't flow as hard and that backside DE has to honor the bootleg and cant crash down the line and that opens up the cutback lane.

That's whats so great about this offense. It just opens things up for itself. That's not to say great players aren't needed, but the offense in itself allows great players to be even better.

[The Pencil Neck]

I think when you have a great running game your play action works better, but with the style of offense we run with the stretch action our play-action will be effective regardless.

But when you have a guy like Arian back there people are going to get open more. And then what is the killer about our offense is when the passing game gets going the LB's can't flow as hard and that backside DE has to honor the bootleg and cant crash down the line and that opens up the cutback lane.

That's whats so great about this offense. It just opens things up for itself. That's not to say great players aren't needed, but the offense in itself allows great players to be even better.

The play that comes to mind is the first play of the Tampa Bay game last year. TB comes in talking about how they were going to shut our running game down. First play? Play action bootleg. 80 yards to Jacoby Jones, TD. I love watching that play.

Later in that game, play action to Arian and then when he sneaks out to the flat, Schaub hits him. 70+ yards, TD.

What a bittersweet game that was.

[El Tejano]

I always thought that Schaub and Rosenfelts worked the PA to perfection. They even fake out the camera guys!

[Insideop]

The facts don't support you.

The believability of the PA fake sells it. If you make your play action look exactly like your normal running play, then the defense will read run and react as though it's a run. And it's going to do that regardless of how good your run is. That's what happens if you have good play design and consistent action by your entire team.

When we had bad running games, our play action fakes still worked because the plays were designed to sell the run. As opposed to those weak-assed fake draws where the QB doesn't even really look at the RB that so many teams call PA.

That's exactly what I was trying to say about giving the O-line some credit for the PA. When the O-line doesn't "tip their hand" so to speak, and sets up the same every time, it's hard to tell whether it's another run or a PA pass coming. And defenses look for every little sign or tip they can to tell if it's going to be a run or pass. I heard, more than once last year, that the Texans O-line was one of the best at at disguising what play was coming. Of course it always helps to have a couple of stud RB's that make the running game so good.