Managing Expectations: Realities of running in the NFL

Last Thursday I authored a short post that was simply several polls asking for fans thoughts on the running game and testing their knowledge about certain aspects of the running game of the 2017 Seattle Seahawks. My plan when I put that post together was to gain a better understanding of the expectations of fans regarding the run game and to address the results of the poll in a post that looked at the fan responses, however, the results of one of the questions was so noteworthy that I felt it required an entire post of its own.

Specifically, the question asking what fans consider an acceptable stuff rate for an NFL team, the results of which were as follows:

As a reminder, a stuff is when a rushing attempt goes for a loss or no gain, and according to the poll, of the 747 fans who took the time to vote, just a hair under a third of respondents believe that 10% is an acceptable stuff rate.

That is incorrect.

10% is not an “acceptable” stuff rate, as it is an off the charts phenomenal stuff rate. How off the charts is it? Since 1996 no NFL team has recorded a stuff rate that good. Not the 2005 Seahawks with Shaun Alexander running behind Walter Jones and Steve Hutchinson (21%). Not the 2014 Dallas Cowboys with DeMarco Murray running behind Tyron Smith, Doug Free, Travis Frederick and Zack Martin (18%). Not the Minnesota Vikings with Adrian Peterson in his prime or the Kansas City Chiefs with Jamaal Charles in his prime nor any other team. That’s twenty two years of NFL data and over 600 team seasons, with no team performing at that level. The closest any team has come was the 2002 Denver Broncos, which posted an 11% stuff rate, and in 20 of those 22 seasons the best team in the NFL posted a stuff rate between 12% and 16%.

That takes us to the next option in the poll, which is a 15% stuff rate. Almost half of all those who answered the poll chose this answer, but once again it is an answer that falls on the extreme high side of expectations. In 4 of the past 22 years a stuff rate of 15% would have led the league, and in three more it would have been close enough that it may have led the league depending on what came after the decimal point. In short, in almost a third of seasons over the past two decades a 15% stuff rate would have been at, or extremely close to, leading the entire league.

This brings us to the third option in the poll, 20%. This is probably the most correct response to the question. In the 22 seasons for which Football Outsiders has data, the league average stuff rate has fallen between 18% and 22%, with 19 of the 22 seasons coming in the narrow range of 19% (ten times) to 21% (five times).

Thus, in the NFL fans should expect to see one out of every five running plays end in a loss or no gain, and so it’s congrats to the 131 respondents (18% of 747 responses) who chose this option.

That leads us to the final option of 25%. I suppose one could argue, rather easily actually, that with the Seahawks posting an abysmal 29% stuff rate over the course of the 2017 season that 25% would be an acceptable rate, and it would be hard to counter that argument. However, that wasn’t my intention with the question, though I can’t argue with that line of logic, but let’s move on to the related question regarding the stuff rates for individual Seahawks backs this season.

In that poll we saw the following results:

On this question, 24% of fans were correct in choosing J.D. McKissic who posted the lowest stuff rate of all Seahawks running backs in 2017, but I would like to focus on something else. That something else is the value of the eye test, and the impression of fans when a player delivers such as Mike Davis did when he was given a shot towards the end of the 2017 campaign.

While Davis led the team in rushing attempts that gained 10 or more yards (9), he was also the absolute worst on the team when it came to his stuff rate, and it wasn’t even close (Author’s note: C.J. Prosise did have a stuff rate of 36.4%, which was somewhat close to Davis’ team worst 39.7%, but Prosise had only eleven rushing attempts on the season which I have excluded as a small sample size). Here are the numbers for the five Seattle running backs who recorded at least 46 rushing attempts on the season.

Thus, it appears that many fans may have fallen prey to what I call the slot machine impression when it comes to their thoughts on Mike Davis. With a slot machine, there are just enough moderate wins delivered to keep the gambler pulling the lever in hopes of a big payday, while distracting the gambler from the fact that they are losing money on the majority of the pulls. This is Davis’ style of play - he delivers just enough double digit runs in order to give the impression he is effective and those double digit gains outweigh the impressions left by the fact that he recorded a loss or no gain at a rate nearly double the league average.

Later on this week I’ll take a look at the results of the other questions that were posed in the original post, but having reasonable expectations is definitely important, so this was something that I definitely felt warranted attention all to itself.