One of the most interesting and intriguing, yet always personally infuriating, plays that the Colts have run this year (both in preseason and against the Raiders) is the play-action pass with a pulling guard.
The play generally involves a single-back set and at least one tight end. Luck fakes a handoff to one side and the guard on the opposite side pulls to the play-fake side to simulate a running play, further adding to the ruse.
The concept is an old staple of the Stanford offense, and a well-established one. Chris Brown at Smart Football highly endorsed it last April (and provided a better explanation than I ever could), specifically talking about that Luck-led Stanford offense.
For the Colts, the play presents some benefits, but some real challenges as well.
By my count, the Colts used the play-action with a pulling guard three times on Sunday (out of 11 play-action dropbacks).
On those plays, Luck went 1 for 2 for 20 yards and touchdown, while running for eight yards on one carry. The Colts allowed pressure on two of the three plays.
The first play that the Colts used the aforementioned play was in the second quarter, which resulted in Dwayne Allen’s touchdown reception. The play is a good example of the benefits and drawbacks of the pull, especially for the Colts.
We’ll start with the negative that’s exemplified with the play: Samson Satele (yellow).
When a guard pulls, generally it’s up to the center to slide left or right and pickup the defender, depending on the defense. In this case, there was both an outside rusher for Anthony Castonzo to deal with and a defensive tackle directly across from Donald Thomas. When Thomas pulled, it left a wide gap for Samson Satele to slide through in a small amount of time, something that Satele simply doesn’t have the foot speed to do. The result was a fairly easy quarterback hit surrendered.
The positive side of that coin is that the play-fake actually works very well. both linebackers (red) take two steps forward before realizing too late that it was a fake. Allen (blue) is behind them before they can recover, and Luck gets it to Allen for the reception. With Allen’s strength, he aborbs a blow and walks into the endzone.
The other weakness that’s magnified by this play is the lack of Mike McGlynn’s quickness. When asked to pull, McGlynn simply isn’t quick enough to be able to get to edge rushers, as he’s asked to do here.
Now, on this particular play design, it’s not critical that McGlynn get there; Luck is sprinting right directly after the fake.
However, the play does indicate how slow McGlynn is to pull to the left. If you watch McGlynn pull on other plays, it’ll quickly become noticeable. He’s slow, and it can cause issues with protection.
On the one play that the Colts pulled a guard and did not allow a QB pressure, it was due to the defensive end not rushing the passer, but dropping back in coverage.
This play was really odd: Dominque Jones lined up at right tackle while Gosder Cherilus was at left tackle and Anthony Castonzo was technically a tight end on the left side.
The imbalanced line allowed the Colts to fill the gap on the left side where Thomas pulled out of easier: Cherilus essentially became the guard and Castonzo the tackle, but it left the team open to a devastating pass rush on the right side. Fortunately, the defensive end dropped back instead of rushing the quarterback.
Overall, it’s not a concept that I love, not with this offensive line anyway. It would likely be a bit easier to stomach if Hugh Thornton was pulling, and not McGlynn, but that’s not in the cards as of now. It simply introduces too many weak points into an already weak pass-blocking line, something I hate to see.
Nevertheless, the benefits are clear: the Oakland defenders bit on the play-fake on each play, with one directly leading to a touchdown.