2017 was Darrell Bevell’s most unorthodox year

Any offensive coordinator will absorb fans’ rancor when things go wrong. Over the years Darrell Bevell took shit from the Seattle Seahawks community for rushing at the wrong times, for passing at the wrong times, for playing too conservatively and too creatively.

Any given play call can fail to gain a first down, but the criticism that rang out probably more often than any other was how Bevell’s sequencing of play calls were too predictable. The formula was familiar:

Ha ha du classique! You know it when you see it.

But we were curious to find out just how often Darrell Bevell in practice resorted to this particular pattern of plays, so we examined the Seahawks’ play combinations from every new series of downs in the 2017 season, and in earlier seasons of his tenure.

The implication of “run-run-pass” spoken as a self-evident critique is that back to back rushes on first and second down typically yield so few yards that the offense is forced to pass on the ensuing third and long. This assumes the third down call is entirely situational, which is probably more true than most downs given the priority to avoid fourth down—but whatever degree that it is true subtracts the validity of predictability of the third down call from the critique because that decision is viewed as a necessity. The problem with run-run-pass, for most who complain about it, is the opportunity cost of the first two runs. For that reason, and to simplify the combinations, we decided only to inspect the sequences of first- and second-down calls.

There are also, as we have written before, all sorts of different ways and directions to rush and pass that complicate the real accuracy of lumping all variations of running and passing plays into the same separate baskets, or treating all similar sequences of rushes and passes the same. But we are going to limit the complexity of this investigation to match the complexity of the critique.

Anyway, it turns out two consecutive runs were the least frequent combination of calls Bevell made on new series of downs in 2017—accounting for less than 20 percent of the sequences. This past year, a run followed by a pass was the most common order of plays (36 percent), then pass-to-run (22 percent) came almost tied with pass-pass (21.6 percent) before run-run brings up the rear.

These figures discount the 99 plays called on first down that produced an immediate new set of downs. As you may expect, those plays are highly skewed toward passes (which are far more likely to gain 10-plus yards at once), although you might be equally surprised to learn that even 27 first down rushes gained another first down. Another big chunk of those results (20 percent) came from penalties on the defense, but I assume most of those were defensive holding or other types that result from pass play calls. So it’s worth noting that even though the two-down sequences involve more run plays on first down, Bevell called passing plays on first down more often overall.

As for second down, the data show passes came more frequently on those plays as well. That perhaps can add some additional context to charts like this one:

Although Ben B’s study demonstrates the results of second down rushing, it’s worth considering how many of these second and 10 situations were the result of first down passing, and consider how different sequencing of the plays might change predictability of the second down calls or their expected value.

For the purposes of this study, we wanted to capture the most typical playcalling scenarios—so we excluded quarterback kneeldowns, for example (which are run-run by default), and also situations when Seattle was leading or trailing by more than eight points because uneven play scripts have a tendency to produce more extreme playcalling results (more passing when behind, more rushing while ahead).

However, it occurred to us that part of what bothers Seahawks fans might be an undue frequency of run-run called specifically while needing to catch up. Seattle was trailing by multiple scores already in the first half four times in 2017, and these situations might have annoyed fans more than the “typical” game states. But for the record, run-run was even less of a factor at those times with just 15 such combinations in 79 possible sequences (19 percent). You might say that similarity to the standard rates is evidence for calling those combos “too often” in catchup mode, but it is still far from frequent—and second down rushing as a whole was even less common with pass-run making up only 16 percent of combinations.

Indeed, in 2017, the only way run-run-pass could be considered “classic Bevell” is for the way it resembled Bevell’s playcalling practice during the Golden Age of Seahawks football, from 2012 to 2015. Here is a chart comparing first- and second-down combinations for all years when Bevell combined with Russell Wilson:

As you can see, run-run has not been the leading type of combo since the last time Seattle had a sturdy rushing game in 2015, and even then it was in decline relative to its massive peak in 2012. It was the third-most used option in 2016.

But back in 2012 the disparity is startling: The sequence types are widely distributed along the y-axis, led by run-run as the most extreme proportion on the table (37 percent) and the other run-first combination still accounting for more than 27 percent of play sequences (not pictured: a lower percentage of passing plays also gained first downs on first down). This makes sense considering how the Seahawks were bringing along a rookie Wilson by relying on his arm as little as possible until he adjusted to NFL schemes and speed. (Wildly, run-run was even more prevalent in spots when the Seahawks faced a deficit that year, amounting to 53 percent of all combinations called while behind by more than a score. They really wanted to restrict Wilson from slinging.)

But although pass-pass sequences took an even more radical dip in 2013, and the distribution rank remains 1) run-run 2) run-pass 3) pass-run 4) pass-pass all the way through 2015, you can already see these trends start to converge toward a more passing-oriented offense before 2016 when the lines finally cross. In that year, Bevell not only called run-pass more often run-run but also pass-pass made a big surge into the second most frequent combination. Meanwhile, rushes that burst for 10-plus yards on first down took up an even smaller slice of those plays not counted in these charts (17 percent).

So the years 2012-15 form the reasonable foundation for an expectation that Bevell calls run-run frequently and perhaps stubbornly, but a steady evolution toward more dependence on passing plays also takes place. What makes 2017 so interesting according to this chart is that those courses go totally haywire. Although the run-run sequence continues its crawl to the bottom, pass-pass also dives again in favor of the mixed combinations pass-run and especially run-pass.

In that sense, the year that got Darrell Bevell fired was his least-predictable, most unorthodox season of playcalling, both from the viewpoints of his traditional habits and the trajectories that were beginning to flip those patterns. Obviously, playcalling is still mostly a situational proposition and you have to deal with the talent and execution at your command as well as whatever down and distance. Bevell’s screwy patterns this past season probably reveal the challenges he faced with a lousy rushing game and poor pass protection more than any grand design. We don’t intend this as a defense of Bevell or an argument he should have kept his job or alternatively that his firing was deserved. The readers can reach their own conclusions on the meaning of these data.

But the findings do seem to cut against some of the narrative grain. Rather than being an unadaptive coach who refused to recognize the changes in performance undermining his offense, this analysis seems to show somebody trying to balance responding to new challenges with maintaining a philosophical commitment—and who in so doing lost the plan altogether.