Seahawks advanced stats: Russell Wilson and Passing Plus-Minus

During the offseason, Football Outsiders publishes a wealth of interesting quarterback statistics from the previous season. These include play-action splits and pressure splits, which have not been released yet for 2016, and passing plus-minus, which I will be discussing here. The usefulness of these stats is that they use their tracking data to provide context to each quarterback’s performance. For example, pressure splits describe how often quarterbacks are pressured and how they perform when under pressure relative to when they are not pressured.

The yearly passing plus-minus articles are written by Scott Kacsmar and debuted last year (covering the 2015 season). This year’s was released in late May and I haven’t seen much discussion about it, perhaps because going deep into QB stats isn’t as fun when Wilson’s performance in 2016 suffered due to his injuries.

The idea of passing plus-minus is simple: to adjust completion percentage for the difficulty of a pass: i.e., target location. One would expect Sam Bradford to complete a higher percent of his passes than Cam Newton because Bradford throws a huge portion of his passes close to the line of scrimmage where they are more easily completed. Here is a chart from the first passing plus-minus article illustrating how it is easier to complete passes that do not travel as far in the air:

Note that these do not count: "Thrown Away," "Tipped at Line," or "Quarterback Hit in Motion." Again, passes thrown close to the line of scrimmage and to the middle of the field are easier to complete than deep passes and passes to the sidelines.

How does Russell Wilson’s completion percentage look after adjusting for target location?

C%+: Wilson’s completion percentage in 2015 was 6.2% higher than the average quarterback after adjusting for target location. Plus-minus: Wilson completed 27.4 more passes than expected after adjusting for target location, which led the league in 2015.

A quick digression into some non-Wilson tidbits. The five QBs ahead of him in 2016, from #1 to #5, were Drew Brees, Matt Ryan, Sam Bradford, Kirk Cousins, and Aaron Rodgers. Brees has led the league in passing +/- five times since 2010 (Brees in 2011 is the best season on record in the time that Football Outsiders has data for this, since 2006). Two of the worst three seasons on record as of 2015 were Jaguars QBs (Gabbert in 2011 and Bortles in 2014). Andrew Luck’s rookie season is the 10th worst +/- recorded.

Back to Wilson. It is no surprise that he fared worse in 2016 than 2015, but I find him being ranked sixth in 2016 somewhat remarkable given the injuries he played through and complete lack of support from the running game or offensive line. Ranking sixth in this statistic is also surprising given that Wilson ranked #15 in DVOA, #14 in DYAR, and #16 in passing expected points added. When I reached out to Scott Kacsmar, he guessed that sacks are the most likely explanation for the discrepancy between passing +/- and the other advanced QB stats.

And indeed, Seahawks QBs were sacked on 6.9% of dropbacks, eighth worst in the league. Another piece of evidence suggesting sacks are a drag on Wilson’s numbers are his ranking #13 in 2016 in NY/A (which account for sacks) and #7 in yards/attempt (which do not).

What have we learned from +/-? Even though he was injured on three separate occasions, when Wilson was able to attempt a pass in 2016, he was still one of the 10 best at completing passes, given target location. If he manages to stay healthy and not get sacked a lot in 2017, both of which are somewhat dependent on offensive line play, expect him to return to being one of the very best QBs in the league.