How Kelce Excelled, McKissic Struggled in the Slot
We have already looked at 2020's top wideouts in and out of the slot, but if we left it there, we wouldn't be looking at the full picture. Cole Beasley led all receivers with 308 slot DYAR, but he wasn't your leader in DYAR out of the slot, not by a long shot. That honor goes to Travis Kelce.
Just what is the difference between Kelce and Beasley? Or, perhaps more apropos, between Kelce and a larger slot receiver such as Larry Fitzgerald? Sure, Kelce has 2 inches and 40 pounds on Larry Legend, but both spent plenty of time working out of the slot and ran similar route trees. Does it make sense to talk about the best receivers in the league without talking about a player who was targeted well over 100 times lined up as a receiver, even if the roster insists on calling him a tight end? Kelce had more slot targets than Tyreek Hill did, after all; analyzing the Chiefs' performance in the slot without including Kelce is missing a huge chunk of the picture.
And, while we're at it, not all running back targets are created equal either. While there are plenty of running backs who clog up our tables with a zillion screen passes, there's a difference between a guy such as Alvin Kamara, who had over 80% of his targets come out of the backfield, and J.D. McKissic, who saw nearly half of his targets come from the line of scrimmage. A full picture of slot performance needs to include those running backs who split out wide too!
And so our annual look at slot/wide splits in the passing game continues, thanks to the charting efforts from our friends at Sports Info Solutions. Today, we'll finish off our look at offensive skill positions by taking a quick glance at running backs and tight ends.
Slot/tight targets made up 9.1% of running back targets in 2020—a little down from previous seasons, but well within normal tolerances. Those targets had a combined DVOA of 9.3%, compared to -1.8% when those players were lined up in the backfield. It's important to note, however, that these numbers aren't directly comparable to those of wide receivers; running backs are compared to other running backs, and opponent adjustments take into account the player's position on the roster, not on the field. That 9.3% slot DVOA for running backs isn't an apples-to-apples comparison to the 2.3% DVOA for slot receivers.
The gap in running back receiving DVOA is what we call selection bias. The Broncos were smart enough not to send Melvin Gordon out to play slot receiver, for instance, even if they were foolish enough to throw him a zillion swings and flat routes over the course of the year. All these caveats aside, however, if you have a running back who can run traditional receiver routes, they can provide significant value.
Only four of the 50 qualified backs from 2020 had more than one-third of their total targets on slot/wide plays, and only McKissic, Chase Edmonds, and Nyheim Hines would have hit the 25-target threshold on their slot and wide targets alone. Alvin Kamara had the most receiving DYAR among running backs last season; he basically got one target a game lined up as a receiver. Kareem Hunt was third, and he essentially saw one receiver target a month. Not all backs can handle a receiver-type role, and not all offenses have room for a running back split out there thanks to depth at the wideout position.
|Slot vs. Wide, Running Backs, 2020|
Generally speaking, there has been a correlation between running backs lining up at receiver and high receiver DYARs. In 2019, for example, we were talking about Christian McCaffrey, Austin Ekeler, and Tarik Cohen as the backs lined up as receivers, and they finished first, second and, uh, 33rd in receiving DYAR that year. Chase Edmonds and Nyheim Hines don't have the same kind of pedigree as the McCaffreys and Ekelers of the world; 2020 was Edmonds' first year qualifying for the main receiving tables, while Hines had never had a receiving DVOA over 3.5% before this season. But pedigree or not, they were two of the top receivers last year, finishing second and fifth in receiving DYAR, and both the Cardinals and Colts found plenty of ways to use them as actual receiving threats.
And then you have J.D. McKissic, which, what? 51 non-running back targets for the 28th-ranked running back by receiving DYAR, 30th by DVOA? McKissic's heavy number of targets were inflated by working with Alex Smith last season, because someone had to catch all of those dumpoffs. And that would be the end of the analysis if McKissic was just catching all those balls out of the backfield, but McKissic's 51 slot/tight/wide targets are the most we have seen from a running back since we started running this data five years ago. His 61 targets beyond the line of scrimmage last year were the most for a running back, beating Alvin Kamara's 48. Heck, he even beats Kamara if you just look at passes at least 3 yards downfield, 39 to 37. That's an absurd amount of usage for a player who, frankly, isn't that special—his numbers are supposed to come way down this year with Antonio Gibson taking on more of a three-down role, Peyton Barber hunting for targets as a pass-catching back, and Adam Humphries and Curtis Samuel coming in to take some of those slot targets away.
People are going to look at McKissic's 2020 numbers in the future and scratch their heads. How did McKissic tie for the 19th-most targets a running back has ever received? Why did Washington keep going back to him despite his -3.2% DVOA? Was his usage as a slot receiver just a symptom of Washington's broken 2020 offense, or was it a secret success, with McKissic playing the slot receiver role he was born to play? His DVOA on passes out of the backfield was -12.4%, so his slot/wide numbers had to be higher. Were they high enough to justify his usage? Let's take a look.
|Slot/Tight Efficiency, Running Backs, 2020|
|Minimum five slot/tight targets|
For McKissic, the answer is no; a -5.9% DVOA doesn't really justify 34 targets out of the slot. This is no surprise.
Nor is it a surprise that it's yet another Rams player who leads his position in slot DVOA, as we are apparently incapable of writing these slot/wide articles without talking about Sean McVay's team. Malcolm Brown's 130.9% slot DVOA is decently impressive, and doing it with a negative depth of target is even more so. Four of Brown's six slot receptions ended up as successful plays; two screens, a chip-curl, and a flat route, all of which picked up first downs. Considering these are some of the least valuable route types in the league, it's worth a small amount of praise that Brown was able to turn them into successful plays. His two unsuccessful plays were some dumpoffs on a pair of third-and-15s; the kind of give-up play that is a little more justifiable when you have seen the guy pick up third-and-longs multiple times over the course of a year.
But no, it's Chase Edmonds who was the best slot receiver among running backs last year, leading the league in DYAR and finishing a solid second in DVOA. Edmonds ran a more complete route tree out in the slot—seven curls, three outs, two drags, a seam, and a slant to go along with four flats and screens on his 18 targets. That's impressive, considering Kliff Kingsbury's Air Raid generally features a, shall we say, simplified approach to receiving routes; Edmonds' route selection as a receiver wasn't significantly different from that of Larry Fitzgerald or Andy Isabella. That probably helps preserve some of his value, even with James Conner coming to town to take on more of the actual running part of the running back role, though we should note that Edmonds had a negative DVOA in the slot in 2019.
In 2019, Tony Pollard led the league in slot DVOA, but we warned readers that most of his value came from one broken play against Detroit and shouldn't be taken as gospel going forward. Appropriately, he fell to dead last with -10 DYAR in the slot in 2020, mostly thanks to a couple of bubble screens/jet sweeps in September, with the Cowboys fooling no one by dumping the ball to him on second-and-long. Those two plays combined to lose 12 yards—not -12 DYAR, but -12 yards—and the Cowboys stopped using Pollard as any sort of receiving threat. I'd put him behind Reshad Jones, who managed a league-worst -60.4% DVOA on his five slot targets; his two incomplete passes were both charted as overthrows, and I'm not expecting Jones to track down an errant Tom Brady toss out of the flat.
For the fourth year in a row, Alvin Kamara had a positive DVOA out of the slot; he joins Austin Ekeler as the only backs to have positive DVOA in both 2019 and 2020 in this category. Kamara, Ekeler, and the injured Christian McCaffrey are generally considered the gold standard for receiving running backs, and the fact that they regularly have solid-to-great numbers when lined up outside of the backfield is a big reason behind that. Consistent good performance in this split is a pretty good way to evaluate the best receiving running backs in football.
|Wide Efficiency, Running Backs, 2020|
|Minimum five slot/tight targets|
For McKissic, again, the answer to "was he a secret success as a wide receiver?" is no, with a -1.3% DVOA not justifying 17 targets split out wide. Washington's offense simply was broken in 2020.
Myles Gaskin is your DVOA leader, but it should be noted that nearly 70% of his DVOA split out wide came on one 59-yard touchdown against the Raiders that really should have been stopped after 5 or 6 yards. Cut that one play out and his DVOA drops from 140.2% to 44.2% and he falls short of the minimum number of targets to be listed.
Myles Gaskin goes 59 yards to the 🏠! @MylesGAS
— NFL (@NFL) December 27, 2020
Instead, it's Nyheim Hines who's your leader by a wide margin, as splitting him wide became a regular site in Indianapolis for about a month. Hines had 11 of his 14 wide targets between Weeks 8 and 13, resulting in 48 DYAR and a DVOA of 78.3%. Why that month? It lines up a little with T.Y. Hilton's groin injury at the start and Johnathan Taylor's emergence as an every-down back at the end, though neither of those circumstances feel like they fully explain what went on there. In the end, targets to running backs split wide are rare enough that one solid month can lead the league. It also should be noted that Hines had a negative DVOA working out of the slot for the second year in a row; he is not someone I expect to be consistently near the top of receiving tables in future years. All in all, 103 of his 158 receiving DYAR came in a five-week period last season, so caveat emptor if you expect him to be a reliable receiver week-in and week-out in 2021.
How do you get a -79.0% DVOA when split out wide? If you're Jamaal Williams, you drop a screen, botch a broken play, and otherwise pick up nothing on a jet sweep and a couple of curl routes. Williams picked up 12 yards on his five wide targets, none of them going down as a successful play. The Packers offense looked rejuvenated in 2020. This was not thanks to Jamaal Williams.
Austin Ekeler and Duke Johnson are the only qualified two running backs with positive wide DVOA in both 2019 and 2020. Ekeler's opportunities were cut more in half without Melvin Gordon; Ekeler spent more time in the backfield rather than as a second running back split wide last season. Johnson's numbers are a bit artificially inflated by a 48-yard gain on a tunnel screen, but he would have finished with a positive DVOA even without that one play, for what it's worth.
Some tight ends are essentially slot wide receivers who are above-average blockers. Twenty tight ends saw at least half their targets from the slot, and eight saw at least a tenth of their targets split out wide. Two players—Jared Cook and Jimmy Graham—fell into both of those categories, really stretching the definition of tight end to it's very extremes.
|Slot vs. Wide, Tight Ends, 2020|
Note that not all of these numbers add up to 100; there were 66 targets to tight ends who were lined up in the backfield, led by the nine of Dawson Knox. The line between tight end, fullback, and H-back is sometimes a blurry one.
|Slot Efficiency, Tight Ends, 2020|
|Minimum 35 slot/tight targets|
No, Travis Kelce's 321 slot DYAR can't be directly compared to Cole Beasley's 308 due to position modifiers, but great googly moogly, 321 DYAR? In the slot? That's beyond ridiculous. Kelce only lines up as a traditional tight end 40% to 45% of the time, and for good reason; he's one of the most effective slot receivers in football, just one who happens to be one of the best run blockers at the same time and built like a Mack truck.
We somewhat facetiously asked what the difference between Kelce and a slot receiver was in the lede, and the answer is mostly size. Kelce is 6-foot-5, 260 pounds, larger by quite some margin than any qualified receiver in the league in 2020. In fact, the 10 heaviest qualified wide receivers all saw more than half their targets split wide, a group of players which is declining in recent years. Most players Travis Kelce's size can't run receiver routes. So when you get someone like Kelce who can, it's no surprise that opposing slot corners, linebackers, and safeties are just exceptionally out of their depth. It doesn't hurt that Patrick Mahomes is the one delivering those strikes to Kelce, but Kelce was also leading all tight ends in slot DYAR when Alex Smith was the guy throwing to him. Travis Kelce is very good; just another example of the groundbreaking analysis you've come to expect from Football Outsiders.
George Kittle and Jared Cook are also very good; the three finished have now finished 1-2-3 in slot DVOA in each of the last two seasons, and were 1-2-4 in 2018. We talked about how Kelce lines up wide so much, but Cook makes Kelce look positively H-back-esque. Cook spends less than a third of his time actually lined up as a tight end, with 82% of his targets coming lined up in the slot or out wide. Saying that Cook and Kittle play the same position is a bit ridiculous; Kittle is line up inline about 75% of the time and still matches and surpasses Cook's numbers in the slot. There's a question out there about whether Kittle or Kelce is the best tight end in football; my answer to that is that the two play different positions and wouldn't be quite as efficient if they were swapped one-to-one with each other.
T.J. Hockenson and Zach Ertz have now had negative DVOA in the slot in back-to-back seasons. Ertz may well be washed at this point in time, as all his advanced numbers fell off a cliff with Philadelphia's collapse in 2020. As for Hockenson, well, he made the Pro Bowl last season despite just 7 DYAR, and probably should have been replaced by Robert Tonyan. But do you know who definitely should have been replaced by Robert Tonyan? Evan Engram, who shows up near the bottom of the league in yet another advanced stat. Engram's 2020 Pro Bowl season continues to baffle.
And then there's Jimmy Graham, at the bottom of the DVOA table. We won't spend too much time talking about the fading last years of a player who, when at his peak, was one of the best tight ends in football, and we wish him well in his retirem—wait, he's back? Huh. Alright, Chicago, can you explain why? Maybe with the help of another table?
|Wide Efficiency, Tight Ends, 2020|
|Minimum four wide targets.|
Tight ends split wide really aren't a thing; note that four-target minimum to be qualified. We include it mostly for completeness.
Hey, Jimmy Graham! You do have life in you after all. And by life, I mean that seven of these 11 targets came in the red zone, with five resulting in touchdowns or first downs. Graham can still provide a size mismatch over an overwhelmed corner, and that's probably the extent of how he should be used at this point. But hey, it's a skill, and as long as he can continue to fight off smaller players for contest grabs, he can have a role on a roster.
Kelce once again leads the league in DVOA here, but Darren Waller pips him to the line in DYAR. Waller has now led this category two years in a row, in part because he's one of very few tight ends to receive double-digit wide targets multiple years in a row. 63 DYAR is a new record for tight ends split out wide, and maintaining nearly a 40.0% DVOA on 19 targets is nothing to sneeze at. Waller, Jared Cook, and Tyler Higbee are the only tight ends with positive DVOA split wide in each of the last two seasons.
Impressive in its own way is Hayden Hurst's 0% catch rate when split out wide. Yes, it was only on four targets, but any 0% is worth noting. To be fair, Hurst didn't have a great chance at any of his four targets—one was underthrown, one was a tipped interception, and two more were charted as defensed. Still, it's not like Hurst was finding ways to get separation, and maybe it's not the hugest surprise that Atlanta went out and drafted Kyle Pitts.