2018 Seahawks rode the plus side of variance to success

Sunday I took a look another look at one piece of the Seattle Seahawks offensive playcalling in 2018, and how the offense only utilized Tyler Lockett as a rushing threat in a very narrow set of circumstances when it comes to down and distance. Today, I’m going to dig a little deeper into one specific part of the article by Matty F. Brown that was published Friday.

Specifically, what I’m going to dig into is the following tweet.

Before I jump right into the numbers, let me just state that according to the Pro-Football-Reference.com game play finder, the Seahawks did not convert to a new set of downs 75% of the time when the team called a run play on 2nd & 10. To back up that assertion, here’s a table of every single running play called by the Seattle offense on 2nd & 10 during the 2018 season.

So, looking at those 2nd & 10 plays from the 2018 seasons on which the Seahawks ran, we see that it’s actually 27 times the Hawks ran on 2nd & 10. Thus, the 18 conversions following a 2nd & 10 run represent a 67% conversion rate, which is obviously very good. In fact, it’s better than the conversion rate after throwing on 2nd & 10 in 2018, which was 55.8% (Author’s note: It appears Rossler’s percentages are different due to counting scrambles as called run plays, when by my opinion, they are called pass plays that failed into run plays. Thus, I’ve included them in the calculation of gaining a first down after calling a pass play on 2nd & 10).

In any case, the fact still remains that Seattle was more likely to convert a 2nd & 10 into a first down by running in 2018 than it was by passing. However, is that an actual trend or is it simply randomness?

As you can see in the above table, I’ve included the conversion probability for the third down of the distance created by each of the 2nd & 10 running plays called over the course of the 2018 season. By summing these together, we can see that the Hawks far exceeded expectations. Based on the numbers used, for the 24 third down plays which followed the second down runs that came with ten yards to go, there are two different conversion probabilities. The first is the league wide conversion rate for third downs of that distance and the second is the Seahawks conversion rate for third downs of that distance. Summing each of these columns, the Hawks would have been expected to convert

• 12.9365 first downs using the NFL conversion probabilities and
• 13.484 first downs using the Seahawks conversion probabilities.

As seen when adding up the number of “Yes”es in the First Down Converted column, the Hawks eventually converted 18 first downs from these 27 second and ten runs. These numbers include the three conversions which came on rushes of ten or more yards (one by Chris Carson and two by Lockett) on 2nd & 10, along with the probability of converting each of the 24 third downs.

So, the question then becomes what led to the offense converting four or five more first downs than would be expected? There are probably those who would argue that it’s because the team has Russell Wilson and he’s a really, really good quarterback. However, that argument seems to ignore the fact that Seattle converted below league average overall on third down, which is something that one might expect if that were the cause.

In any case, here’s a second table that includes all of the third down plays which followed a run on 2nd & 10 by the Hawks during the 2018 season in case anyone wants to attempt to discern any type of pattern. The only thing that really sticks out initially to me is that Wilson has a sack rate of 20% on third down after the team has run the ball on 2nd & 10, but that doesn’t have anything to do with converting, so I don’t have any further insight on what in particular allowed the team to outperform expectations. Perhaps it was Wilson. Perhaps it was Lockett and Doug Baldwin. Perhaps it was randomness.

Whatever it was, it was the upside of variance, and here’s to hoping they can catch that upside wave again during the 2019 season.