Pro Football Focus recently finished grading the 2006 NFL season, which gives them a full decade of data (2006-2015) by which they can evaluate players. In light of that fact, they’ve decided to name an “All-Decade” team that doesn’t look at a standard (although arbitrary) decade starting at a zero year (like 2000-2009 or 1990-1999), but their decade of data.
One Viking made the first team, three made the second team and one player was sort of an honorable mention.
Well… not really. The author of the list had to be argued out of including Jim Kleinsasser at fullback. He was a favorite of theirs as a blocker, but he played about five percent of snaps at fullback and couldn’t reasonably be included there above Ovie Mughelli—who played as a fullback as his official position title and lined up there much, much more often than Kleinsasser.
Two of the second team decisions were very reasonable; it’s hard to argue with them choosing J.J. Watt and Justin Smith at defensive tackle over Kevin Williams or Vince Wilfork over Pat Williams at nose tackle. The Wilfork/Williams debate is much closer in my opinion of the two, but it’s worth reading their blurbs on DT and NT.
As the man currently in the process of redefining PFF’s grading scale—as well as his position (I could easily have included him as an edge rusher)—it’s perhaps not surprising that no one has a higher average grade than J.J. Watt’s 95.2, or an individual season grade better than his 98.8 (2013 season).
From 2009 to 2011, Justin Smith was setting the standard that Watt was about to overtake, ranking first twice and second once among interior defenders, while averaging a 91.0 grade during that time. Like Watt, he was outstanding against both the run and pass, tallying 522 QB disruptions.
One of the benefits of going back and grading earlier seasons is getting to see great players at their best, instead of at their current levels—such was the case with Vince Wilfork. While he still does a reasonable job in run defense, venture back to 2006–2007 for a taste of why offensive line coaches are still praising him. His “career” run-defense grade of 84.6 is inflated by those years, and it will be interesting to see where he finishes when we get around to grading 2004 and 2005.
One player who made the second team, however, was one you wouldn’t have expected to be anything but a first-team shoe-in: Adrian Peterson. Here’s what they said about selecting Marshawn Lynch above Adrian Peterson—
Conventional wisdom would have Adrian Peterson here instead of Marshawn Lynch, but maybe that’s more a function of the media’s dislike of Lynch than anything tangible. Whichever way you cut it (except with base numbers that derive as much from the quality of the blocking as they do from the actual RB), Lynch edges Peterson out in every category.
Consider this: Peterson, the bell-cow power-back, broke 423 tackles as a runner on 2,496 attempts, while fumbling 34 times. On 2,337 attempts Lynch broke over 100 more (530) tackles and fumbled only 20 times.
Additionally—although neither has done much as a receiver, admittedly—Lynch has been a superior pass blocker throughout the last 10 years.
The reasoning is understandable, but I think it falls short. They had to condense every argument to a few sentences, and it would have been boring to type “Player X had a higher cumulative grade than Player Y” for every selection (plus, that’s not how these lists are constructed), so it makes sense that they isolate a few statistics.
But if they took their own grading at their word a little more than an isolated statistic like “broken tackles per attempt,” they likely would have stuck with Peterson. Because both Lynch and Peterson were drafted in 2007, there’s an easy comparison between the two to be made.
The conventional wisdom is that Lynch was not particularly impressive in his time in Buffalo or early on in Seattle. Out of Lynch’s nine career years, four or five of them are not good. For Peterson, the argument has been that he’s been good—even the best—for most of the time that Lynch was in Buffalo and remained good as Lynch moved to Seattle, making Lynch’s candidacy based on a few years where he might have ranked above Peterson as a back.
Pro Football Focus’s grades do not generally disagree with this traditional line of thought regarding Lynch. He had one positive grade in Buffalo (+1.0 his rookie year) and three negative years. His half-season in Seattle and his full season a year after were also negatively graded, so his grades line up as +1.0, -8.5, -4.7, -5.3, -1.0.
(Note: I am using “grades” to refer to their internal system of assigning a player points based on every single play of the game, which gets translated into a +/- system where 0 is average, not their 0-100 rating system they display in public consumer data—which they now refer to as “grades” in their blurbs and on their website)
They were less bullish on Peterson than the general viewing public, however, and gave him negative grades (-6.7 and -2.9) for his second and third years in the NFL. Still, those grades are better than Lynch’s.
In fact, Peterson had a higher grade than Lynch in six of the nine years they have. Throw out the shortened 2014 season and Lynch’s unusual 2010 with two teams, and Peterson has a higher grade in five of the seven years there’s full data for.
In the two years Lynch has a better grade than Peterson, the average difference in grade is about 8.0 grading points. In the five years Peterson had a better grade, the difference is, on average, about 9.1 grading points.
Cumulatively, Peterson’s grade is around +75, while Lynch’s grade is around +62. While they both ranked #1 the same number of times (once) and in the top ten the same number of times (four times), Peterson’s next-worst years are where he ranked 11th and 27th, while Lynch’s are 32nd and 104th.
It seems like in their system, Peterson more consistently had positive grades across the nine years, had a better period of dominance over Lynch than vice versa, was better for longer, had a higher peak and a longer peak.
While it’s true that Lynch is a much better blocker and a marginally better receiver than Peterson, the bulk of his career shows that those qualities did not make up the deficit he had to Peterson as a runner until very late.
It’s an understandable argument, but it feels like only recency bias can drive putting Lynch above Peterson for a team that is decidedly not meant to do that.
While the “missed tackles” argument is also compelling, I’m not sure it does a good job of capturing the complete skillset involved in running the ball.
For example, Adrian Peterson has outranked Marshawn Lynch in every single year I can access data for (2007-2013) in yards after contact per attempt and in total, beats him by a significant margin, with 3.2 per attempt to Lynch’s 2.6. Peterson never fell below 2.9.
For context, 3.2 ranks #1 in that dataset for all running backs, and in most years ranks in the top five for backs with at least 100 carries. 2.9 ranks in the top ten. 2.6 yards after contact ranks around 15.
So while Lynch may have broken more tackles, Peterson’s broken tackles seemed to be much more meaningful—much less the fact that Peterson outranked Lynch in yards before contact (which is even more reproducible across years) and that “missed tackles per attempt” ignores the fact that it is a skill to avoid a tackle attempt as well—outrunning a tackle attempt is probably better than making the most of the tackle attempts that more consistently arrive.
Oh, and the one Vikings player who ended up cracking the first team? Antoine Winfield, at slot cornerback. It’s smart of them to include “slot” as its own first team space given how different those responsibilities are from the outside cornerbacks and it gives space for players like Winfield to shine. Their argument:
We’ve been arguing for some time now that lumping all corners together for ranking purposes isn’t a good idea. The position of slot cornerback is so different from an outside corner that it’s like trying to compare guards and tackles. For that reason, we are going to start grouping them differently, and this is a good place to begin. For a small guy, Winfield was one of the best run defenders we’ve ever seen, and running a screen pass to his side when he was in the slot was tantamount to giving up on the play. Teams would put guards on him, and he’d knife past and make the tackle for a loss with the consistency of a metronome. This wasn’t his whole game, though; he was excellent in coverage, too, as a career average grade of 84.7 in that facet of play will testify.
All hail Winfield, the history’s smallest, best cover linebacker.