A Look At Duke's Playcalling

Duke’s offense has suffered from slow starts and periodic droughts all season long, and fans on social media are beginning to focus on the playcalling as the source of the problem.

Zac Roper has been the Blue Devils’ offensive coordinator since the 2016 season, which corresponded to a dropoff in team scoring.

In fairness to Roper, projected starter Thomas Sirk went down with a season-ending injury late in 2016 training camp, and the team had to start a redshirt freshman. Of course, that freshman then developed into the sixth pick in the NFL draft, and the scoring still didn’t climb back to previous levels.

There was also a dropoff, though not as steep, in total offense.

This year, as the offense struggles to move the ball, the clamor has grown louder that Roper is not up to par as a playcaller in the ACC. The final play against UNC was perhaps the low point.

Duke lined up to run a trick play at the goal line, leading UNC defensive coordinator Jay Bateman to call time out, because Duke had run the same trick play, from the same formation, at the goal line to beat Army a few years earlier, when Bateman was coordinator there.

Despite the fact that they were calling the same trick play against a coordinator as the last time they played him, and despite the fact that UNC called time out to discuss it after recognizing the formation, Duke stuck with the play and threw a game-ending interception.

We decided to try to take a look at Roper’s performance as coordinator. Have the Blue Devils’ struggles on offense been due to his playcalling?

It’s obviously a tough thing to isolate. Quarterback Quentin Harris can check out of called plays or make decisions based on what the defense shows him. So, is a bad play the result of Roper’s call or Harris’ decision? Or a failure from one of the other players on offense?

We focused on an area where the coordinator likely has more of the responsibility—the start of games.

Many teams script the first 15 or so plays of the game, a practice that began in the 1980s with the San Francisco 49ers. They also script the first play of many special situations, such as third (or fourth)-and-short or goal line plays. We saw that in Duke’s opener against Alabama, when they went to the triple-option after getting a turnover in Alabama’s red zone. Afterward, coach David Cutcliffe said the team planned to go to that scheme the first time they were in scoring position.

While, we don’t know for sure that Duke scripts its first 15 plays, it seems likely that the Blue Devils would be sticking close to Roper’s game plan in the early going.

So we took a look. Here’s Duke’s per play average for the first 15 offensive plays (excluding and-short and goal-line plays) of each game this season.

There’s obviously a great deal of variability. But it’s tough to see the “slow start” trend in the data. That’s because the “slow start” often comes after the true, scripted start. For instance, against Notre Dame on Saturday, Duke’s first two snaps of the game went for 10 and 6 yards. Then the offense slowed down, but the early success, presumably designed by Roper, goes into the first-15 numbers. Similarly, while the offense “did nothing” against Pitt, the first drive of the game marched down the field and produced a field goal.

In fact, in the first 15 plays, the offense actually does slightly better (4.81 yards per play) than afterward (4.74).

The offense is also much more balanced in the first 15. Duke has called 69 passes and 66 runs in the first 15 plays, a 51-49 split. After that, it’s 221 passes, 305 runs, a 42-58 split.

The running game doesn’t do quite as well in the first 15—3.7 yards per carry, compared to 3.9 afterward.

Harris’ performance at quarterback is fairly consistent.

The special-case scripts also seem to be successful. In the first and-goal snap of each game, Duke has gotten a touchdown one third of the time. Similarly, on the first “and-one” third or fourth down snap of each game, Duke has averaged 1.9 yards and gotten the first down 56 percent of the time.

So, if Roper’s game plan is relatively successful, what happens afterward? Does Harris play too conservatively, audibling out of pass plays into safer runs? Or do the coaches scrap the plan when the team falls behind and do a poor job of improvising?

It’s tough to pinpoint whose fault it is when the offense hits a dry spell, but it seems that Roper at least gets the offense off to a solid start.