Dangerous Territory: Play-calling when pinned deep

Play-calling deep in friendly territory is risky business, and the negative consequences of any mistakes in these situations only heaps added pressure on the offense to move the ball downfield. A bad play or two can spell disaster for a drive, leaving the offense in third-and-obvious-passing situations deep within their own territory. In addition to this, a fumbled snap, a poor QB/RB exchange or an interception can exacerbate the situation, sapping the confidence of the offense.

So how exactly did teams address the challenges of playing deep from their own territory? Let’s look at the 2012 season data.

The short answer to this question, which may not surprise you, is that play-calling trended away from spread and empty backfield formations in favor of traditional single-back formations. The changes in offensive scheming manifested themselves in two ways. First, running took on added significance. Second, teams reduced the number of formations they used.

When I began looking at the numbers behind play-calling in these “against their own end zone” scenarios, I went in assuming that coaches and coordinators would be more cautious than in other scenarios because they have far more to lose if the play is broken up here instead of in the open field. Based on my assumption, play-calling should have been more conservative, relying heavily on the run and a short passing game to avoid giving the opponent good starting field position. Below is a short chart showing the number of plays run, both inside and outside of the offense’s 10 yard line, broken down by running and passing.

The first thing I found was that rushing attempts increased by nearly 7% when teams were closer to their own end zone. To put that into perspective, a similar increase in rushing plays outside of the 10-yard line would mean an additional 1,214 plays, or a new run/pass ratio of 8609:6208. This relative conservatism in play-calling when backed up to the end zone makes sense if you consider the potential negative consequences of passing. If a pass falls incomplete, the offense has not made any forward progress but has still exposed the quarterback to sack or injury, and this also does not include any psychological consequences of pressure, an interception or tipped ball. Running, on the other hand, is a safer alternative and even a short gain can provide a quarterback or punter valuable breathing space in which to operate.

Beyond the simple examination of running and passing, however, I also wanted to see what kinds of variations would be found in formations. Again, I was not surprised to find that outside of the 10, formations had far more variety than inside of the 10. What did surprise me, though, was just how significant the disparity was. Inside the 10, teams used 14 different combinations of receivers, running backs and tight ends. Outside of the 10, that number burgeoned to 26 combinations! Simply stated, when playing deep in their own territory, offensives reduced the variety in their play-calling, choosing to use fewer and more traditional schemes.

In order to provide some examples of the conservative play-calling I mention above, I have linked to some videos to illustrate how different teams lined up to execute plays deep in their own end zones. In each case, the team uses a play that we would call their bread and butter, but none of the teams attempt to use a gimmick formation (e.g. Oregon does not use a 3RB set with two TEs inside the 10, but they do go outside).

The first two examples are from the Alabama-Texas A&M game.

In these scenarios, Alabama uses a shotgun single-back formation with four WRs (1RB, 4WR Set). The four WR set and the shotgun formation lure A&M’s overly-aggressive defensive ends deeper into the backfield as pass-rushers trying to sack McCarron. With the DEs now in the Alabama backfield, Eddie Lacy has more room to run the draw play. In the second scenario, now having secured a few extra yards from Lacy’s run, Alabama chooses to use the exact same formation, this time deciding to pass. The A&M defensive ends, having been caught out of position on the last play, do not crash off the ends, instead choosing to play more conservatively along the line, giving A.J. McCarron ample time to check down his receivers and dump the ball off to his RB, Lacy, before the play is eventually broken up. By relying on the conservative running play, Alabama is able to move the ball away from the end zone and set up the second play.

The next two examples are from the USC-UCLA game.

In these scenarios, USC uses an I-Formation with two receivers out wide and the other lined up off the TE (2RB, 2WR Set) in order to pound the ball out of the shadow of the end zone. In the second example, USC again lines up in an I-Formation, now with Robert Woods no longer lined up directly off the TE. The only difference between the short gain and the long run is the success of the FB in knocking his LB out of the play, giving the RB enough space to break into the second level of the defense. In both scenarios, USC chose to keep the ball on the ground, running inside behind two all-conference linemen and the fullback. Also playing to USC’s advantage, they purposefully ran away from UCLA’s defensive stud of a linebacker, Anthony Barr.

The last two examples are from the Oregon-Oregon State game in order to show how a spread offense handles situations like this (and Oregon fans seem to post more videos than other fan bases).

In the first example, Oregon lines up in a spread shotgun formation with two RBs and three WRs. Relying on their zone read gameplan, Marcus Mariotta hands the ball off to his #1RB who runs right up the middle. This is a very atypical play for Oregon especially when using DeAnthony Thomas, typically choosing to leverage him in their riskier, though incredibly potent, outside run game. In the second example, Oregon again chooses to line up in their spread shotgun formation. Much like in our Alabama example, Oregon now has a few extra yards of space between the QB and the end zone so they choose to pass. When the ball is snapped, the linebackers choose to play a shallow zone, possibly spying the QB because of Mariotta’s effectiveness in moving the ball with his feet. Because the linebacker is playing shallow, Mariotta is able to hit his receiver over the middle for an easy catch in the gaping hole between the second and third levels of the defense. Now, it could be argued that these plays are not very conservative, but when compared to Oregon’s typical running style of bouncing outside with the wide receiver setting the edge, I would argue the two shown in the video are, in fact, a more controlled risk.

For the sake of comparison, here are four consecutive plays and formations called by Oregon. These four plays are all from the Oregon-USC football game in November. Notice the lateral running game that is absent in plays called within the 10.

It’s probably no surprise that offensive coordinators are more conservative deep in their own territory, where any mistake can swing a game in the opponents favor. The question is how that manifests on the field. The most obvious sign of cautious play-calling in our charted data was the reduction in the number of formations used. While many of the 26 formations mentioned were used sparingly outside of the 10-yard line, the fact that only about half as many existed within the 10 says a great deal about risk aversion. Not many coordinators are going to line up their offenses with three RB’s when deep in their own territory because any lost yardage can be dangerous. But many did attempt to use three RB’s outside, where a loss of yardage had less of a negative effect on the team. The second factor was the increased reliance on running. Though not a dramatic increase, the simple fact that running showed an increase demonstrates that coaches wanted more space between themselves and the goal before they opened up the playbook.