The NFL is incredibly homogenous. Offenses and defenses are shockingly similar when compared to their college counterparts, and naming your ideal player at a position isn’t necessarily all that difficult. At quarterback, many people would look at Aaron Rodgers’ accuracy, mobility and intelligence and select him. At receiver, it’s hard to beat Calvin Johnson. Safety? Probably Earl Thomas. 95 percent of football fans would likely select a healthy Rob Gronkowski at tight end.
But if I told you to name your ideal defensive tackle, you would probably pause. Do you mean nose tackle? Probably Dontari Poe, or maybe Damon Harrison. Oh, not that kind of nose tackle. Well, I guess Haloti Ngata or the more recent incarnations of Vince Wilfork. Or maybe you meant an under tackle. Geno Atkins then, or Ndamukong Suh. Oh wait, did you mean 3-4 defensive end? J.J. Watt, obviously. Or, if we’re going to be strict about the "two-gap" thing, Kyle Williams may be your best bet.
If I asked you your ideal road vehicle, I may get similar results based on what you needed. Someone might want a Lamborghini, while another would prefer a Ford F-150. That Honda Accord might catch your eye if you're practical, and the Toyota Prius may appeal to the hippie in you. Maybe you have some off-roading to do (like a real hippie), so you’d like that Range Rover (OK, a yuppie hippie).
Depending on who you ask, those players and vehicles represent the top-tier of their class, and it’s all an appropriate answer to the question, in part split because of function. Getting the full capability of one of those for what you need will have you set for some time.
But what if you could get 85-90 percent of all of them?
Enter Kevin Williams, your RAV4 (or CR-V, but I liked the alliteration in the title). Like the RAV4, Kevin Williams is not the best of any one of those prototypes and may be deficient in one specific area (two-gap nose tackle for Williams; doing whatever it is you were going to do with that Lamborghini for that RAV4) but pretty much gets the job done and better than most of its counterparts everywhere else.
Instead of being an elite under tackle as he was in his heyday, Kevin Williams is a very good everything.
The trouble is, he may not want to be that.
Last year, the Vikings found themselves injured at nose tackle (also found themselves "bad" at nose tackle) in their 4-3 system, but had two starting-quality three-technique players in Kevin Williams and Sharrif Floyd. Moving a three-technique to one-technique is not an ideal solution (in fact, they fielded the worst defensive tackle in the league precisely because of that kind of move), but with multiple injuries, there wasn’t much choice.
He was significantly better than the starters at that position.
Hell, it was one of the best games he had in two or three years. His response to the revelation—one that may extend his career?
We just discovered that our used Range Rover was secretly a new RAV4, but it would really have rather stayed a Range Rover.
The coaching staff was mildly enthusiastic (which is a fairly energetic response for Leslie Frazier) but that enthusiasm didn't hold after a discussion with Williams about a potential move. While the starter and his backup at nose tackle healed up, the Vikings backed off of their mild enthusiasm and merely stated he would continue to "fill in" at nose tackle but not start.
Perhaps not ideal for either party.
So, what’s the deal? Is Kevin Williams not a team player?
That's not really accurate. He did whatever was asked of him and he did it exceedingly well. Terrifically. In fact, after he was asked to move from 3-technique to any-technique, his PFF score improved from -0.04 a game to +1.12 a game (nearly identical to Geno Atkins’ injury-shortened year, or Dontari Poe’s resurgent 2013 and better than Tony McDaniel).
The primary question is whether or not Williams signed with the team with the expectation that he play his preferred, primary role or play in an expanded role that may extend his career.
Given how resigned Williams was to retirement—he had stated multiple times that he’d be willing to retire if he didn’t like any particular deal (and refused to sign for vet minimum)—it seems like he’d have some leverage over how he’d like to be used.
Add to that the offers he fielded from a few teams, and it sounds like Kevin Williams is on a plan he prefers. Given that Tony McDaniel played 3-4 defensive end with the Miami Dolphins for three years, it makes sense to me.
For all the excellent SPARQ analysis done by you all, there doesn’t seem to be a particular prototype for three-technique defensive tackles. In fairness, the 2012 defense and the 2013 defense seem different in terms of gap-shooting and pass-rushing, which could allow us to treat Brandon Mebane as an anomaly, especially now that he plays a nose tackle role (from what I could tell, both Mebane and Branch played the B gap at times, but Mebane far more).
Even so, there isn’t a universal thread. Weights ranged from 283-304 pounds, ten-yard splits ranged from 1.63-1.77 seconds, bench reps ranged from 23-36, vertical leaps ranged from 27.5" to 38", broad jumps ranged from 8’11" to 10’1," short shuttles ranged from 4.47 to 5.04 seconds, three cones from 7.25 to 7.66 seconds, 40-yard dashes from 4.78 to 5.01 seconds, hand sizes range from 8.5" to 10" and arm length ranges from 32.38" to 34.5".
Of course, that’s not to say that most of the three-techniques eligible for a true pass-rushing role didn’t fit some sort of prototype at a metric, and they all had a pSPARQ score over 110 (112.1 to 137.6 because Clinton McDonald is insane), which is one standard deviation over the average.
They generally had good burst metrics (usually, sub-1.7 second ten-yard splits, 30" vertical jumps, 9’4" broad jumps) and length (33" arms or longer) with one exception in general for those metrics (and a different player was the exception each time for the exceptions).
In all honesty, I doubt it matters for Kevin Williams. He has a history of production, which means any evaluation of him does not require projection (in the same sense)—physical prototypes matter less for a productive player at his position than they do for an incoming rookie or reclamation project.
For what it’s worth, here’s Williams compared (by color!) to the extremely rough outline of a pass-rushing tackle who has played for the Seahawks since Pete Carroll took over (green is better, red is worse…r):
Given Kevin Williams' age, it's probably true that not much of this matters, aside from arm length—which I think we can safely assume hasn't changed much since the draft. Aging does things to a body, I'm told, and Kevin Williams is no longer the quick sort-of-defensive-end-sort-of-defensive-tackle the Vikings drafted with the seventh... or eighth... or ninth pick. Dammit. It was the ninth pick.
He has been quite productive. In the PFF era (2007-2013), he is the single-highest graded defensive tackle in their system, including one year as the top defensive tackle in their grades (2008), two years as the third-overall DT (2009 and 2010) and three years in the top ten (2007, 2011 and 2012). He’s a six-time Pro Bowler, a five-time All-Pro and is on the NFL All-Decades team for the 2000s. He joins Troy Polamalu and DeMarcus Ware as one of the only defensive players to be on that team despite being drafted in 2003 or later (of 22 players).
He officially has 66 career pass deflections, and is the PFF leader among defensive tackles for batted passes at the line of scrimmage, with 40 since 2007.
In second place is a tie between Barry Cofield and Marcus Stroud, with 17.
He leads defensive tackles in pass pressures, with 265. In second is Jonathan Babineaux with 52 fewer at 213.
Kevin Williams may be the best defensive tackle of the 2000s. It will be difficult to see him that way, in part because those sorts of designations are unfair to better players whose careers straddled the decade line (Warren Sapp) and in part because of his market and supporting cast.
He always had a great supporting cast, which hurts his perception as an all-time great. Normally those with a stellar cast get a boost from team success (Richard Seymour is well-regarded and deserves it, but does receive some level of recognition from playing for Super Bowl champions, as does Sapp), but not so for Williams, who has only been on a team with a win in the playoffs twice (2004 and 2009).
Regardless, he’s certainly in the discussion and his five first-team All-Pro selections are the most of any defensive tackle in the 2000s, with a tie for second place between Richard Seymour and Warren Sapp (three each).
As for some interesting additional context, Pro-Football-Reference lists Jared Allen as the player whose career best matches William in shape and quality (a calculation that has no regard for recency or team). That same list seems to think that there is a seven-year run of play that matches Hall-of-Famer Bob Lilly more than anyone else, too.
In my previous "previews," I may have left you with what you thought was hyperbole only to be somewhat disappointed (or with Percy, with a case of blue balls) but I assure you that Kevin Williams brings with him the experience of a Hall of Famer. Unlike Antoine Winfield, though, there are clearer signs of age with Williams.
His rate statistics as of late are merely "very good" instead of great, but he’s better than the Seahawks have had for some time. Sure, I’ve attempted to dazzle you with counting statistics instead of rate statistics, but that was more to explain who I’m writing about and who the Vikings are letting go of from a historical perspective than it is a way to convince you that you’ve got a great deal on a RAV4.
Something I like to do is create a PFF grade based on everything we know of a player and weight it based on recency to create a prospective grade that takes into account the last season but also some more information to correct for an off year and have a mean to regress towards. In this case, the most recent year is half the weight, the year before it one-quarter and so on.
In my rough looks, this combined approach has been more predictive than just the raw score of the previous year (I haven’t tested it, but at least the theory of it feels right). This doesn’t account for aging very well, but I’ll address that in a moment.
In doing so, Kevin Williams comes out ahead of Alan Branch, Jason Jones, Clinton McDonald and Tony McDaniel in PFF grades (+12.2 to +6.4, +7.1, +4.9 and +8.6 respectively) but falls behind Brandon Mebane solely because of Mebane’s prodigious score against the run last year.
In a recency-modified set of statistics (designed to mimic a year), Kevin Williams beats out all Seahawks defensive tackles in pressures produced, with 31, with Clinton McDonald falling in second at 25. Williams wins the sack battle against everyone except Jason Jones (3.6 sacks to 4.1), but doubles his QB hit total (8.2 to 3.7). Only Mebane comes close to the QB hurry total (19.0 to 18.7).
His 4.9 batted passes at the line of scrimmage are barely challenged, too. Mebane, Branch, Jones, McDonald and McDaniel produce 1.7, 1.0, 3.0, 0.0 and 1.0 respectively.
His run defense statistics (Run stop rate and run grade in particular) fall behind McDonald and Mebane, but are still well ahead of league average. With the Seahawks and their system, I think he’ll thrive in both categories, even if he won’t reach his career peaks.
In respect to aging, it is of course difficult to predict when exactly a player will fall off a cliff. As a Vikings fan, I’ve been spoiled by the long career of Pat Williams, who signed with the Vikings at the age of 33 and put together six solid seasons before retiring.
I looked at all of the defensive linemen who have retired and played at least one season between 2004-2014 at the age of 30 or more (175 players), those that had a near Pro-Bowl level year (an Approximate Value of at least 6, per Pro-Football-Reference) at the age of 30. There were 87 qualifying players, which seems like a high number until you take into account that DLs don’t usually last to 30 years of age unless they’re very good and that these are all players who have hit 30 years old over the span of more than a decade, not in a single season.
Of those players, 35 (40%) played at the age of 34. Compared to the general population of 30-year old DLs (those without the high Approximate Value), that’s very good. Overall, 46 players of the 175 played significant snaps at the age of 34 (26%). That’s unsurprising, given that we’d expect players playing at a high level slightly past their prime to last longer.
Those around Kevin Williams’ AV that year (his was ten, I looked between AVs of nine and 11) played until the age of 34 over half the time.
Naturally, we have more information about him than that. He played at or near a Pro Bowl level for the following three years, too. Looking at those with an AV of those whose weighted average between the ages of 30-33 (weighted for recency), there were 40 DLs who qualified at "near Pro Bowl level" or better. 33 of them (83%) played significant snaps the next year. Of those who played at that level specifically at the age of 33 (29 players), 25 of them (86%) played significant snaps the following year. For those that more closely match Kevin Williams’ AV, the numbers are similar—20 of 25 players (80%) with his recency-weighted AV played at the age of 34, and 8 of 12 with his AV (66%) specifically at the age of 33 played into the age of 34.
We’re much more interested in whether or not the aging data suggests that Kevin Williams will decline precipitously, not just whether or not a team would be willing to play him, so it may be more useful to take a look at which performers who have performed well continued to perform well.
Those with recency-adjusted scores similar to Kevin Williams performed at near a Pro Bowl level about 40 percent of the time. Those who put up his scores at the age of 33 repeated the score 65 percent of the time at age 34. There were only seven players who put up an elite score at age 30 and a near-Pro Bowl level performance at age 33, and four of them maintained that performance.
If you expand that to include those with high-level performances at age 31, the number drops to about 40 percent. At least one player who hit the elite-at-30 mark and the great-at-33 mark only put in a subpar score because of injury (Ted Washington), which skews what we’re looking for a bit because of the small sample.
There are a lot of words and numbers in the preceding paragraphs (made more confusing by hasty math and quick generalizations), so it's easier to sum it up like this: I think this means it is more likely than not that he will continue to put in high level performances, and a near certainty that he’s a better player than the bottom of a roster. Anecdotally, the fact that his play improved as the season went on is a good thing, and the fact that he will almost certainly have fewer snaps wear him down over the season should help.
So what can Kevin Williams do?
Kevin Williams doesn't have the speed he used to have, but he certainly has most of it and a lot of strength. More than anything else, however, Williams has savvy. I don't say that because we automatically assume all veterans have savvy, but because that's how Kevin Williams has been winning in the league for a while now, especially with former partner (but not brother) Pat Williams.
His ability to read a play and react to it is fairly astounding for a man his size and he adapts well to play fakes and misdirections as well as anybody.
This shouldn't surprise you. Here in Minnesota, Kevin Williams is consistently described with the kind of plaudits people like to reserve for their work-hard fan favorites. He has a high motor, plays the game "the right way," (aside from the StarCaps controversy that honestly made the NFL look worse than Williams), and has all of the subtle nuances mastered.
Sure, that's all true, but it can present itself in very meaningful ways.
I mentioned that Williams has a good rate of batted passes among DTs, but even if you expand the set to include 3-4 DEs (a group of players far more likely to bat passes) Kevin Williams has more passes batted at the line of scrimmage than any other interior defensive lineman in the past five years. J.J. Watt, of course, has more per year, but the point remains the same. For those with at least 2000 snaps, Watt ranks first in batted passes per pass snap, but Kevin Williams ranks second (of 77 qualifying interior defensive linemen).
His ability to get the pass out of the air isn't just limited to the immediate impact of forcing a third-and-long via incompletion—clogging up passing lanes can make quarterbacks hesitate or change their play. In the first GIF above, Aaron Rodgers arguably does that (I say arguable because there are any number of other factors that could have caused that, but the receiver in that direction was not covered downfield in my view).
He has a few pass deflections that don't come from play at the line of scrimmage, but the vast majority of them come from rushing the passer, not dropping back (like in the last GIF above). Much of this comes from a natural instinct to put his arms up if he's not getting to the passer, though of course his very long arms play a significant role. It's a fine balance between continuing to pursue the play and get to the passer and finding other ways to end the play.
There's something to be said about continuing to fight for pressure after losing the initial move and knowing when to do that is significant. As a result of that, he's been able to consistently get to the quarterback and create pressure despite his age.
He does this with a few different techniques and from a variety of positions on the line.
In terms of pass-rushing moves, Williams' favorite move is not easily identifiable with common pass-rushing parlance, like a swim, rip or bull rush. He locks one arm into the sternum of the offensive lineman and reacts. Some call it a one-arm bull rush, others a one-arm post move. I've seen it called a stiff-arm rush and a long-arm by people. Most often, I've seen it called a stab.
At any rate, it's a pretty simple move that requires hand quickness and strength, as well as an intuition that guides what to do next. He doesn't always dip out of contact, and sometimes if he wins at the snap will be able to walk into the QB. The distance he keeps from offensive linemen allows him to read the play more effectively as well. It allows him to make sure he's attacking off the center of gravity and maintaining leverage, and also is difficult to stop without opening up to a good counter that puts an elbow in the OL's back.
Williams' game often relies on winning first contact. He isn't usually the first player off the snap of the ball, but he's reactive enough. More importantly, his tools let him get his hand in the chest of a guard or center well before they can do the same to him.
He likes to bull rush as well and has maintained a high level of strength in these past several years. He won't often use a swim move (he does use it, though), but I've seen a lot of different things from him; dip-and-rip (more often on third down), chop moves and other speed rushes. Generally speaking, he likes to keep his rushes as power moves, but is certainly capable of using speed rush moves if he needs to. His quickness lend itself well to this approach, but it seems as if he prefers to stand up the offensive lineman rather than simply going low—perhaps to better read the play or swat the ball if need be.
Williams has missed a few more tackles in recent years if only because he was more adept at taking down ballcarriers with one arm in the past than he is now, but this isn't for the most part an issue. He missed more tackles per run tackle attempt in the run game than any Seattle lineman in the past two years (in each individual year and combined), but did a fair bit in the run game regardless because of his ability to slow down runners in the backfield and clog run lanes.
Also, I should note that because missed tackles are low-frequency events for DTs, it is difficult to gauge a talent (or lack thereof) in this area; Williams' missed tackle total in 2012 was "high" at three. In 2011, he was among the better DTs in the NFL at completing tackles. Nevertheless, his relatively high ranking in this statistic in 2013, 2012 and 2010 implies that it's an issue. It's not as big a negative as it is for linebackers, but it will be something worth watching (Also of note, DTs on teams with good linebackers tend to have fewer missed tackles on their resume because tackles that would have been missed count as assists as the LBs clean up the play).
Other than that, he is extremely good in the run game for the reasons listed above. His speed in pursuit is paired with intelligent tackling angles and quick reaction times. His high motor leads to him making more plays than many others in his position as a three-technique.
Williams probably should stay at three-technique, but he certainly has proven he can play as a one-technique in order to extend his career. The signature value of a 4-3 nose tackle tends to be an ability to consistently anchor against double teams, and Williams did that often as a three-technique. This year, his play against double teams wasn't as strong as it historically has been, but he knows how to change his game as the distance to the quarterback gets shorter.
I would be a little concerned, but not overly so, about the way he approaches those double teams:
I would argue that the first of those GIFs is acceptable play against a double team, while the the second is a bit more worrisome (though he does end it pretty nicely, I suppose).
When playing a double team, Williams prefers to get skinny through a gap instead of simply winning against his opposite on the line. He far prefers to attack than defend (which is of course the counterintuitive nature of line play) but does set low in short down situations in order to simply block a gap or prevent yardage.
He'll consistently keep his feet moving in order to prevent being washed out of plays and he's rarely out of position in a way that begets the kind of yelling that other Vikings defensive linemen incurred from Minnesota fans. If he has been moved out, it more often than not means that the offensive line committed too many people to blocking him, and there should be advantages gained elsewhere.
Nevertheless, his penchant for attacking when his job may be to occupy blocks can cause some integrity issues on the line. To me, this makes him an excellent backup nose tackle, but not a primary option if you have a choice.
The most interesting thing about Williams is that he probably is a very good 5-technique, and in an alternate universe I would be interested in seeing how he developed as a player under that system. When locked up against an offensive lineman, he can play with excellent control and almost always wins the one-on-one battle. His arm length and general capabilities moving individual lineman in order to make a play as it develops are assets here, too.
I think his lateral movement is solid, and when he doesn't have to worry about gap integrity, he can move across the line extremely well. Given his success rate on twists and stunts, playing in a two-gap system would be better for him than it would be for most one-gap specialists.
I'd have no issue with Williams taking on the Red Bryant role in the defense if necessary despite being 20 pounds lighter because he can take the pounding of strongside work and the tight end/tackle double team would be far easier for him than the center/guard double team he's shown adequate (but not amazing) play against.
Naturally, having him on the field in the Bear front would be a great use for him as well.
Given that he's been projected to play in run downs and that Carroll mentioned that they wouldn't ask him to do "anything different than what he's done," I wouldn't be surprised to see him as the three-technique on the weak side instead of as the five-technique on the strong side. McDaniel's experience as a five-technique is indicative here of that sort of move in particular fronts. It is a little curious, however, that Carroll did mention Bryant when asked about Kevin Williams' role (though he did take care to specifically use the phrase "three-technique").
His role seems to easily be a rotation three-tech likely to see about 500 snaps, and as a backup option at five-technique with only emergency work at nose tackle if things go quite wrong. There is a good chance that he will only be asked to do one thing in order to speed up his game, so seeing him as a gap-filling specialist on run downs may be his future and should probably make him more effective than he has been the last two years as a general three-technique tackle.
If he should only be asked to disrupt the inside and complicate things for pulling guards or leading fullbacks, then he'll be worth it in the run-heavy NFC West.
I don't think he is as good as Antoine Winfield was when he arrived at Seattle camp, but Winfield was competing with a secondary known to be good. Williams is arriving in a situation where he isn't necessarily the top interior lineman, but it's one where the corps is without a lot of known talent. The enormous uncertainty among the defensive linemen means that a roster projection is next to impossible and Williams may get washed out, but I think it's more likely than not he makes the roster and makes a significant contribution.
You've got your RAV-4. He's fueled up for one year and you've found a way to make him more efficient and remain versatile. The question is whether or not you want him to be good at one thing or competent at a lot of things. Both have value.