The recent focus on the draft has shifted attention away from the fact that the Cowboys still haven’t named a play-caller. Some find this to be an important determination that still must be made, while others argue that it really doesn’t matter who calls plays. Barry Horn recently pointed out that Tony Romo checks out of a lot of plays at the line-of-scrimmage, meaning the original play-call often isn’t even run. It sure seems like Romo is going to have more freedom than ever in 2013; with a greater number of audibles probably on the way, do the play-calls even matter? Here’s why I think they do. . .

Romo’s Audibles

While Romo has been provided more pre-snap freedom in recent years, he still doesn’t audible all that often. I track his checks each year, and he’s never called an audible on more than 10 percent of the Cowboys’ snaps in a single season.

More important, the majority of Romo’s audibles—traditionally around two-third of his checks—are “Kill” calls. Jason Garrett frequently calls two plays into Romo, who then says both of those plays in the huddle. The offense lines up and plans to run the first play called, but if Romo doesn’t like what he sees in the defense for that particular play, he has the freedom to “kill” the first play, alerting the offense to run the second. That’s what the quarterback is doing when you hear him yell “Kill, Kill, Kill” on television. So while Romo might check out of more plays in 2013, chances are most will still be “Kill” calls, meaning the new play will still be one given to him by the play-caller.

Shifting Odds Through Play-Calling

The goal for any play-caller is to maximize the offense’s chances for success in any given situation. While it’s always important for the offense to execute, sometimes it simply isn’t probable; if a play-caller dials up the worst possible play against a particular defense, it almost assuredly won’t work no matter how well the offense executes. Similarly, the perfect call in a certain situation decreases the number of things that need to go right for the play to be successful, ultimately leading to a higher expected success rate.

As a side note, that’s why it’s silly to automatically label a play that didn’t work as a poor call or one that did work as a great decision. If a play-caller puts his offense in an optimal situation in which they’ll convert a first down 99 percent of the time but it just happens to fail on one particular play, the call was still a good one, regardless of the outcome.