Stealing Signals, Week 10, Part 1
TNF and early Sunday games
The biggest thing on my mind this week is still the two-deep shells, as I look at data on a weekly level, and there are far more quarterbacks than just Patrick Mahomes and Josh Allen who have noticeable drop-offs in passing aDOT over the past month. Mahomes and Allen both got back to their downfield ways in Week 10, but several other quarterbacks did not.
This is a pretty poor visual unless you click through and blow it up, but it’s essentially weekly passing aDOTs for a variety of the league’s quarterbacks. I’ve talked about this phenomenon with Shawn Siegele on Stealing Bananas, and I shared some thoughts on Twitter last week which were mostly met with responses like, “Teams need to run the ball better.”
That’s definitely some of it, but I have yet to see a satisfying explanation for why so many different quarterbacks and offenses around the league have taken such sharp downturns in downfield passing, and all around the same time. Yes, defenses are running more two-deep shells, and yes, they are playing more defensive backs and all of those things. But they’ve been doing that for years now — there was the draft, I don’t know, three or four years ago, when more defensive backs were drafted than ever before in league history. That moment felt like a sign of the times, after teams had shifted to more wide receivers on the field. These are not new 2021 trends, and defenses that feature two deep safeties in particular have been studied by offenses for decades.
So what I’m not satisfied by are these explanations that amount to saying this is the natural ebb and flow of things. I’m also not saying all these quarterback aDOT shifts are going to stick forever, but man, there’s too much happening to too many offenses — and it’s been too successful for defenses — to not want more information.
To me, it seems like there has to be some new schematic wrinkle or focus that many of the league’s defenses are taking up, and have started to pass around during this year. I keep hearing about two deep with man under, which basically means when a player runs a vertical route, he has a guy with him in man coverage and help over the top. If it’s Cover 2 with zone underneath, you can run multiple vertical routes in a way that stresses the safeties — say one down each sideline and one that splits them — and if the underneath defenders don’t follow downfield, you have a numbers advantage. In man under, the underneath defenders are following downfield, and so now you have a lot of defenders downfield when you run those, say, three vertical routes.
But again, that’s not new; that didn’t become something novel here in 2021. It’s difficult to beat down the field, but it makes things easier underneath, where you wind up with one-on-ones that, if won, could become big gains with yards after the catch. We’re seeing a lot of these offenses not only struggle to get the ball downfield, but also just outright fail overall, and I’m talking good offenses against defenses that until a few weeks ago were outright bad.
The aDOT shifts are not super complicated if you simplify things to the two deep stuff I discussed above, but how can that be all that’s going on here? I can’t stop feeling like something is being missed in the larger discussion, like how the middle linebacker getting depth defined the Tampa 2 in the early 2000s. There are some really sharp Xs and Os people out there, and this trend is being discussed everywhere, so it’s probably the case that it’s been discussed in a way that would satisfy my curiosity more, and I’ve just missed it. Certainly if it’s so easy to spot that all the defenses around the league have easily copycatted it in-season, then there have to be observers out there spotting it, too.
But I’m pretty lost, and I’m not afraid to admit that, in part because while I do think some people understand this well, this reeks of one of those football things where 95% of people commenting on it are lost, too. This is kind of a tangent, and one I’ve probably addressed before, but football analysts love to speak more confidently than they are, and we’re conditioned as humans to be attracted to that confidence, and to want to believe it. There’s a whole lot of “fake it ‘til you make it” in my industry, and that’s fine to a degree, but people inevitably start to believe their own BS. To be perfectly clear, there are also a whole lot of people who are incredibly sharp, and do clearly know something in depth when they speak confidently about it. But it’s a competitive industry, and the default tends to be to never acknowledge what you don’t know, at least not publicly. Privately, over my seven years in this industry, I have seen a ton of those very publicly confident analysts express a ton of doubt. (As I write that, I’m thinking of a few analysts I never worked with but know well, so I would suggest assuming nothing about the places I’ve been or the people I’ve seen this from.) But it’s likely the case that the most publicly bull-headed analysts are the ones compensating the most, essentially afraid of being found out, and it’s definitely the case that the most intellectually humble are the ones whose opinions I respect the most.
Anyway, my confusion stems purely from a data perspective. This many different quarterbacks completely shifting the distribution of their passing to far more underneath throws than downfield looks — all around the same few weeks in the season, and to such a stark degree — feels bigger than the answers I’ve seen. The new-age dual-threat QBs can’t read defenses? Tom Brady can read a defense. I know the Bucs have had injuries, but he’s posted passing aDOTs at 8.5 or lower in four straight games after being at 8.9 or higher in four of his first five, and the Bucs just lost to Washington. Those are very cherry-picked cutoffs on the aDOT point, but you can make similar observations about a dozen other quarterbacks around the league, as the above tweet indicates. Then there are some quarterbacks who haven’t noticeably shifted down, and their numbers are bouncing around, but there’s basically only one regular starter whose aDOT is noticeably rising right now — Jalen Hurts, and that’s mostly because the Eagles have shifted so many short passes to runs; their pass rate and overall pass attempts have cratered, so their distribution being different is fairly easy to understand.
How about the idea individual players are lacking in execution, like Mahomes or Allen or Lamar Jackson? OK, fine, maybe. But it’s Mahomes and it’s Allen vs. Jacksonville and it’s Jackson vs. Miami and it’s Justin Herbert and it’s Brady and it’s all these dudes, all over the past month. So honestly, no, I’m not really buying that it’s just a matter of execution, and just a coincidence that none of the league’s rising stars, or even some surging vets like Brady or even a Derek Carr, seem to have answers.
Those explanations might explain certain dips in certain weeks, or shifts for certain teams, but we’re talking about essentially all of the most successful vertical passers in the NFL, dating to prior seasons, but especially including those that were crushing early in 2021, all not only failing to have answers, but all being forced to full-on shift the distribution of the number of passes they are even trying to get down the field. Again, this is why I can’t shake the feeling there’s more to this, somehow.
I’ve also seen some comments about small sample bias, and these are not small samples. This is a big, league-wide thing, impacting a ton of teams, that has shifted data quite a bit across the board. Maybe I’m the idiot here, which wouldn’t be the first time, and is maybe the betting favorite. But if you’ve read or heard something lately from that maybe 5% of people who can really break this down, and show subject matter knowledge such that they’d, say, be able to answer basic follow-up questions — which, yes, I’m saying most people commenting on the “two deep” boogeyman seem unlikely to be able to do — I’d love it if you passed it along.
Kansas City and Buffalo were successful on offense this week, but the big discussion during Monday Night Football was the Raiders playing single-high a bunch, and I’m not real sure about what the Jets were doing but the Bills definitely took it to them. But maybe there was something here, and hopefully those games were the start of some changing trends, where the offenses start to figure it out.
Let’s get to the games. Data is typically courtesy of several RotoViz apps, NFL fastR via Sam Hoppen or Michael Leone, Pro Football Reference, RotoGrinders, Add More Funds, or PFF. Week 1 included a guide on some important acronyms to know for Stealing Signals like HVT, TPRR, TRAP, and WOPR. I also wrote about wTPRR here.
For snap shares, the default for the plusses and minuses is the player’s season average.
Dolphins 22, Ravens 10
RB Snap Notes: Myles Gaskin: 61% (+6), Devonta Freeman: 58% (+0 vs. W9 high), Le’Veon Bell: 42% (+10 vs. high)
WR Snap Notes: Albert Wilson: 55% (+23), Marquise Brown: 75% (+0), Devin Duvernay: 60% (+5), Rashod Bateman: 55% (-8 vs. low), Sammy Watkins: 32% (return)
TE Snap Notes: Mike Gesicki: 81% (+9)
Key Stat: Marquise Brown — 13 targets, 7.8 aDOT (second lowest single-game aDOT this season, behind 7.6 in Week 9)
Entering this game, the Dolphins ranked 26th in the NFL in defending explosive pass plays, while the Ravens were third in explosive pass rate on offense, so this is one of those matchups I’m talking about. The Dolphins just gave headaches to the Bills a few weeks ago, and then here they were able to force Baltimore underneath and make them extend drives. Baltimore did well early, driving for a couple field goal tries, and they could have had a long touchdown if not for a bizarre Sammy Watkins (3-1-7) play where he seemed afraid of the goalpost or perhaps the stands just beyond the end zone in Miami. At any rate, he basically quit on a very catchable long TD that proved to be much more costly over the course of the game than it even seemed at the time.
Rashod Bateman (8-6-80) had another solid game, and while his routes dipped back down to 67% after peaking at 80% last week, Watkins returned to just 33% routes after being over 70% in each of the first four games of the season. Watkins also lost a key fumble late that was returned for a touchdown, so there’s not much of a case to be made he deserved more playing time. Somewhat surprisingly, though, Devin Duvernay (4-4-28, 1-19 rushing) ran routes on 63% of dropbacks, his highest rate since Week 5 when Watkins left early and Bateman was not yet active. The wide receiver rotation definitely stood out early in the game, and I think the Watkins gaffes were a good example of this idea I’ve mentioned before as it related to Jonathan Taylor and Marlon Mack, and Shawn and I have talked about a lot on Bananas, where teams can’t afford to have lesser-than players taking snaps from good ones. Bateman is already a contributor; the Ravens arguably lost this game by mixing in Watkins, because in a game where you’re struggling, all it takes is a couple of bad reps to swing things.
Marquise Brown (13-6-37) was pretty immune to the rotation, running routes on 90% of dropbacks, but his 7.8 aDOT was just barely ahead of Week 9’s season low of 7.6, and the deep shots haven’t been there for him, which hurts his efficiency as a big-play guy. The routes and targets were too strong to ignore, though. Mark Andrews (8-6-63-1) had a big day and was also very involved from a routes perspective, settling in at 94%.
Devonta Freeman (10-35, 4-3-23) was again the clear lead back, with Le’Veon Bell (3-1, 0-0-0) mixing in and Ty’Son Williams active but not seeing an offensive snap. Williams is droppable if you have him in deeper leagues; that he’s not able to even get on the field while Latavius Murray is out is a pretty bad sign for any hopes this year.
Jacoby Brissett was hurt right after half, then was trying to get back in the game, but the Dolphins decided to roll with Tua Tagovailoa. I’m not sure what happened there other than the Dolphins maybe thinking it was unfair to put Tua in a tough spot with limited reps in that game, but once he was in there, the cat was kind of out of the bag and they wanted to see how he’d respond. And he played well, throwing for over 150 yards on just 13 attempts.
Jaylen Waddle (6-4-61) had a solid day, and ran routes on 98% of dropbacks. Isaiah Ford (4-4-84 on 22% routes) and Albert Wilson (5-4-87 on 42% routes) made the most of limited roles, but aren’t players to chase right now. Mike Gesicki (7-0-0) was the only Dolphin other than Waddle to run routes on more than 50% of dropbacks, as it was the same main two as it has been with DeVante Parker out, and then a huge rotation behind Waddle and Gesicki.
Myles Gaskin (14-31, 2-1-14) continued to lead the backfield, but also continued to struggle. He’s been far less efficient this year than last, and really the bulk of that has come over the past month when his workload has been strongest. Over the first five weeks, he was averaging 4.9 yards per carry and a poor but not horrible 5.0 yards per target; in five games since, he’s at 2.7 YPC and 3.2 YPT. Maybe Tua being back will open some things up, but those are really poor numbers and it’s the kind of thing that tends to lose you work.
Signal: Ty’Son Williams — active but no offensive snaps; Myles Gaskin — really poor efficiency over an extended stretch (2.7 YPC and 3.2 YPT in past five games)
Noise: Isaiah Ford, Albert Wilson — four catches for over 80 yards each (both under 50% routes); Mike Gesicki — zero catches on seven targets; Ravens — WR rotation (we’ll see how things progress, but this was the first game with both Sammy Watkins and Rashod Bateman in the lineup, and it was clear Bateman was more of a contributor)
Washington 29, Buccaneers 19
RB Snap Notes: Antonio Gibson: 63% (+11, highest since W1), J.D. McKissic: 38% (-8, lowest since W1), Leonard Fournette: 65% (+5, highest since W4), Ronald Jones: 2% (-7 vs. low)
WR Snap Notes: DeAndre Carter: 67% (-3 vs. high), Tyler Johnson: 69% (+2 vs. high)
TE Snap Notes: Ricky Seals-Jones: 45% (injured), John Bates: 64% (+40 vs. high), Cameron Brate: 62% (+1 vs. W8), O.J. Howard: 58% (+12 vs. W8)
Key Stat: Buccaneers — season-low 47 offensive plays
It’s always interesting to see what teams do coming out of byes, and in this Washington upset win both teams were in that situation. One clear trend was both No. 1s solidifying their roles. Antonio Gibson (24-64-2, 2-2-14) benefited from positive game script, and his routes were only back up to 35% of dropbacks, which is higher than any of his three most recent games, but lower than any of his first five. The early-season bump in routes was cut by injury and now with J.D. McKissic (2-4, 4-4-35) just playing very well, it isn’t likely to come all the way back. This snap share suggests Gibson is healthier after the bye, with two positive notes being Jaret Patterson barely playing and Gibson getting all five of the team’s green zone touches. McKissic’s routes fell to 43%, his fewest since Week 4, which is in some ways due to the positive script but also a reminder his biggest performances this year have been boosted by Gibson’s injury.
Like Gibson, Leonard Fournette (11-47, 9-8-45) played his highest snap share in several weeks, and racked up 9 HVT, while Ronald Jones fell to a new low of just one offensive snap. Again, script could play in here to some extent, as Fournette and Giovani Bernard (1-4, 3-1-4) tend to play more on passing downs and the Bucs totaled just 12 RB rush attempts. But for as uninvolved as Jones has been all year, he’d gotten double-digit snaps in six of eight games prior to the bye, and played fewer than nine snaps just once, way back in Week 1, when he got benched for fumbling. Fournette’s hold on the backfield seems to have only strengthened.
Tom Brady had a rough game, throwing two first-half interceptions, the first of which was somewhat bad luck, while the second was just a bad pass. Brady’s passing aDOT was a season-low 6.1, and I mentioned in the intro how that continued a recent trend for him. In my mind, one answer for Tampa could be to make Jones more of a focal point as the better pure rusher than Fournette, but it appears that ship has sailed.
Washington continued to use DeAndre Carter (6-3-56-1) in a more prominent role even with rookie Dyami Brown back after Brown missed Week 8. Carter ran routes on 65% of dropbacks, his third game at that number or higher, and in all three he has seen at least six targets and posted at least 50 receiving yards. He’s not someone to add in most leagues — he’s a 28-year-old with little production throughout his career — but he’s scored in two straight games and is someone to keep an eye on or consider in deep formats.
Terry McLaurin (8-6-59) led the team in the major receiving categories, while Ricky Seals-Jones (4-3-30) left early after a decent start. It looks like John Bates (3-3-25) would get most of the TE routes if both Seals-Jones and Logan Thomas are out. No one else had more than two targets.
Chris Godwin (8-7-57) led the way in targets and receptions for Tampa, while Mike Evans (3-2-62-1) hit for a long TD but was otherwise not very involved as the more vertical weapon. Tyler Johnson (5-3-17) continued to man the No. 3 role with Antonio Brown out. Cameron Brate (3-1-6-1) ran routes on a season-high 79% of dropbacks as O.J. Howard wasn’t targeted.
Signal: Antonio Gibson — coming off bye, had highest snap share since Week 1, all five of team’s green zone touches, but routes weren’t as high as the first few weeks; Leonard Fournette — coming off bye, had highest snap share since Week 4, while Ronald Jones played a season-low one snap; DeAndre Carter — stayed in a pretty full role with Dyami Brown back, and Carter has earned at least six targets in all three games with at least 65% routes; John Bates — seemingly next man up at TE for Washington if both Seals-Jones and Thomas are out
Noise: Bucs — season-low 47 offensive plays; Washington — season-low 53.6% expected pass rate (were in rare positive script that led to 32/34 pass/run ratio)
Lions 16, Steelers 16
RB Snap Notes: D’Andre Swift: 93% (+21, +15 vs. high), Najee Harris: 87% (highest since W3)
WR Snap Notes: James Washington: 87% (+30 vs. W9), Ray-Ray McCloud: 70% (+34 vs. W9)
TE Snap Notes: T.J. Hockenson: 94% (+10, +1 vs. high), Pat Freiermuth: 62% (-9 vs. W9)
Key Stat: Lions — (-25.7%) PROE (lowest single game by any team this season)
This game suuucked. At least it got the result it deserved, a tie. Detroit ran the ball well early, so they just ran the ball a ton, setting a new single-game low for any team all season with a -25.7% Pass Rate Over Expected. D’Andre Swift (33-130, 6-3-5) got an absurd amount of carries, and was over 100 rushing yards by the third quarter, while both backups ripped off long TD runs in their brief opportunities. Jermar Jefferson (3-41-1) seemed to suffer an ankle injury on his long touchdown, and then later Gordon Igwebuike (2-56-1) had an impressive 42-yard score. The Lions had their shot to win in overtime but missed a field goal, and then the Steelers could have had a couple shots if not for two overtime fumbles, and anyway it was Jared Goff vs. Mason Rudolph and the kind of game that was meant to be a tie.
James Washington (6-2-15-1) ran routes on 94% of dropbacks and Ray-Ray McCloud (12-9-63) was at 83% with no Chase Claypool. Diontae Johnson (13-7-83) led in targets, like usual, and Pat Freiermuth (9-5-31) continued to come on, although it was his late fumble that cost the Steelers their last shot at a field goal try in overtime. McCloud’s dozen targets are maybe notable, although Ben Roethlisberger was out and Rudolph threw 50 times. Some of McCloud’s targets probably came from Najee Harris (26-105, 4-4-28) in the shorter area of the field, and I’d suspect Harris to keep doing his thing when Roethlisberger is back. I have “Mason Rudolph sucks” in my notes and I didn’t elaborate and don’t remember why I wrote it but I’m passing it along.
T.J. Hockenson (1-0-0) had a forgettable day, but his routes were strong and the Lions only threw 25 times for 114 yards, so it’s not like they thrived without him involved. Amon-Ra St. Brown (6-4-61) and Kalif Raymond (6-4-31) were your big receiving stars for Detroit.
Signal: James Washington, Ray-Ray McCloud — big routes for both with Chase Claypool out
Noise: T.J. Hockenson — one target, no catches (Lions threw 25 times for 114 yards, clearly still need him to be a focal point of a functional passing game); Lions — massive negative PROE of -25.7% (lowest by any team all season)
Titans 23, Saints 21
RB Snap Notes: D’Onta Foreman: 35% (+14 vs. W9 debut), Adrian Peterson: 33% (+0 vs. W9 debut), Jeremy McNichols: 27% (-18 vs. W9), Mark Ingram: 85% (+51 vs. W9 NO high)
WR Snap Notes: Marcus Johnson: 63% (+15), Nick Westbrook-Ikhine: 37% (-10), Tre’Quan Smith: 91% (+11 vs. W9 high)
Key Stat: D’Onta Foreman — 78 yards on 13 touches, team-high 3 HVT
The big news here was in the backfields with both the stars out. D’Onta Foreman (11-30, 2-2-48) didn’t run particularly efficiently, but had an explosive play in the passing game, and that was more than could be said for Adrian Peterson (8-21, 1-1-(-1)). Both early-down backs got one green zone touch, and Foreman gained playing time while Peterson stayed even from last week. I think you have to like Foreman right now over Peterson, and while Peterson likely won’t disappear entirely, Foreman just looks more explosive and like the better bet to be a solid fantasy piece. Jeremy McNichols (4-7, 3-1-1) saw his routes dip all the way down to 23%, with Peterson leading the team’s backs at 30% routes and Foreman coming in at 20%. That type of thing is in flux and could shift substantially next week.
On the other side, Mark Ingram (14-47-1, 7-4-61) got a workhorse role, and looked pretty solid. He hit on a wheel route (RB air yards!) for a late big play, and so the Saints went back to an Ingram wheel route for their potential game-tying 2-point conversion attempt, which failed miserably. It called to mind the Frank Gore debacle to have Ingram wheel routes as a prominent offensive staple down the stretch here in 2021, and I couldn’t help but laugh.
Marcus Johnson (6-5-100) ran routes on 60% of dropbacks with Julio Jones out, and had a solid day. My guy A.J. Brown (4-1-16) did not, but this was a slow game with just 27 pass attempts, and Marcus Lattimore is good, and it happens, and look I’ll make every excuse in the book if you don’t stop me. Expect a bounce back next week against Houston. Geoff Swaim (5-4-26) caught four passes for the third straight week, and I will continue to pretend like he didn’t.
Taysom Hill did a little bit of everything for New Orleans, going 2-for-2 passing, rushing three times, and catching one of two targets. That he’s seemingly healthy but not starting at quarterback is interesting, but the Saints have lost both of Trevor Siemian’s starts, so maybe Hill will get a shot soon.
Tre’Quan Smith (7-4-44-1) was up to a season-high 90% routes and looks like the clear No. 1 at this point, with Marquez Callaway (4-2-37-1) operating as the No. 2. Deonte Harris (4-3-84) remains in a part-time role at 48% routes, but continues to be better on a per-route basis than the two guys running the most routes. Adam Trautman (6-5-32) might be the most interesting Saints pass-catcher because of tight end eligibility, but even over the past three weeks where he’s been vaguely productive, he’s only at a 22% TPRR and 0.46 wTPRR, and extending that time period at all only makes him look worse. Smith is the other one I’m monitoring, and this was a step in the right direction for him to see the seven targets and keep solidifying his routes, but his 13% TPRR and 0.33 wTPRR since returning in Week 7 leave a lot to be desired.
Signal: D’Onta Foreman — led the backfield in touches, gained snaps on Adrian Peterson, and was more efficient for the second-straight week
Noise: A.J. Brown — one catch
Cowboys 43, Falcons 3
RB Snap Notes: Wayne Gallman: 43% (+30 vs. high), Mike Davis: 37% (-19 vs. low, blowout script), Cordarrelle Patterson: 28% (injured)
WR Snap Notes: Michael Gallup: 53% (return)
Key Stat: Falcons — (-15.8%) PROE
The Cowboys trounced the Falcons, riding a 29-point second quarter to a massive 36-3 halftime advantage and making the second-half basically all garbage time. That made most of the playing time trends hard to dig into, really, but it’s interesting to note Dak Prescott’s aDOT was among his lowest of the year at 5.8, and he’s one QB who didn’t seem to be negatively impacted by just taking what was there underneath. Prescott threw for 296 yards and two scores on just 31 attempts, and also rushed for a score.
Michael Gallup (5-3-42) made his return and saw two quick targets. He later had a pretty bad drop on a wide open slant that should have been a 15-yard gain or so, and also made a really nice catch on the sideline at the 1-yard line that was a near-miss TD and set up one of Ezekiel Elliott’s (14-41-2, 3-3-15) two short-yardage touchdown plunges.
CeeDee Lamb (7-6-94-2) was the other centerpiece early, and because of the blowout and a minor bruise or something, he only ran routes on 61% of dropbacks. Amari Cooper (4-4-51) played with some of the backups through the third quarter and ran routes on 92% of dropbacks; his role was a little more concerning with Gallup back, and Dalton Schultz’s (2-1-14) targets were pretty nonexistent despite 83% routes. Cooper didn’t see his first target until late in the first quarter, and had a couple catches in the second, but then had his big play of the day — a 22-yard reception — very late in the third quarter. Schultz’s only catch and both of his targets also came in the third quarter. It could have just been a matchup thing, but Gallup looked good back in Week 1 in his only other time this year, and if he does emerge, that will definitely strain Schultz as the guy who has been the No. 3 target all year, and could cut into Cooper or Lamb. In this one, it was Cooper who looked more like an afterthought of those three wide receivers, and Lamb who looked like the No. 1 who is going to get his.
The Elliott split with Tony Pollard (11-42, 7-6-56) leaned more toward Pollard this week, with Elliott below 50% snaps for the first time all season and Pollard at a season-high 42% snap share, but it’s tough to parse how much the massive second quarter and blowout script played in. I mean, that’s the obvious answer, but Pollard’s big share was somewhat real in the sense that it was mostly Corey Clement, not him, playing in true garbage time with Cooper Rush at quarterback throughout most of the fourth quarter. At the same time, Elliott’s low snap share was almost certainly the result of the game flow. And then the duo played together some, with Pollard catching a little tip pass from Elliott after a handoff and nearly taking that misdirection play for a sweet touchdown, except it got called back for holding. There’s probably nothing new here, and it’s just been two weird games over the past two weeks (they were blown out by Denver in Week 9), but it’s something to monitor and Pollard definitely rebounded from one of his lowest snap shares in Week 8 coming out of the bye.
Atlanta basically gave up in the second half, feeding Wayne Gallman (15-55, 2-1-21) carries. The same notes about Elliott likely apply to Mike Davis (4-18, 1-0-0), but Gallman might still be in line for a bigger role next week considering Cordarrelle Patterson (4-25, 2-1-14) suffered an ankle injury and is likely to miss some time. The Falcons play on Thursday night this week, so Patterson is almost certainly out for that one, and Gallman I think becomes someone worth stashing in deeper formats to see what the split looks like with Davis and without the presence of Patterson in the offense, which opens a pretty big hole.
Matt Ryan wound up throwing just 21 passes, with Josh Rosen adding six, as the Falcons didn’t run a lot of plays early and then threw in the towel. Kyle Pitts (7-4-60) had a strong game all things considered, and will need to be even more of a focus with Patterson out. Olamide Zaccheaus (7-2-22) got some volume though Russell Gage (3-0-0) and Tajae Sharpe (1-1-0) ran more routes. There are some bad wide receiver groups around the NFL right now, but Atlanta might have the worst.
Signal: Michael Gallup — looked good in return, was involved early, and his presence seemed to hurt Dalton Schultz most after Schultz had carved out a role as the No. 3 passing weapon; Kyle Pitts — 7-4-60 game despite 27 total passes and 131 passing yards, and Pitts will now need to be more of a focus for Atlanta with Cordarrelle Patterson out
Noise: Ezekiel Elliott — two lowest snap shares of the season over past two weeks (huge negative script in Week 9, huge positive script in Week 10); Falcons — (-15.8%) PROE (they just ran it the whole second half in the blowout loss)
Patriots 45, Browns 7
RB Snap Notes: Rhamondre Stevenson: 55% (+21 vs. high), Brandon Bolden: 27% (lowest since W6), D’Ernest Johnson: 88% (+19 vs. high)
WR Snap Notes: Donovan Peoples-Jones: 85% (+5 vs. high)
TE Snap Notes: Hunter Henry: 82% (+1 vs. high)
Key Stat: Rhamondre Stevenson, D’Ernest Johnson — 8 HVT each (tied for fifth most in Week 10)
Neither team threw much in a slow game where both offenses essentially got only eight possessions. The Patriots scored a ridiculous six touchdowns on their eight and also added a field goal, punting just once. The Browns had a long touchdown drive to start the game, but couldn’t keep pace the rest of the way.
This was another game where the fantasy notes were clearly about the running back rotations. Rhamondre Stevenson (20-100-2, 5-4-14) was involved early and often for New England, handling the Damien Harris role and getting all four green zone touches. He scored twice, but really could have had a third touchdown on a swing pass near the goal line that he flat dropped; there was a defender nearby and it’s not clear he would have scored, but I think he probably would have. He did catch his other four targets and has mostly been sound in that regard this season, while he ran 32% routes. Brandon Bolden (3-32, 3-3-38) was in his passing-down role, running routes on 43% of dropbacks and handling a few rush attempts behind Stevenson. The Patriots get the Falcons on Thursday night, so whether Harris gets cleared in time will go a long way to determining whether Stevenson is another strong start this week, but the way he’s playing he’s likely carving out some type of role in the backfield regardless.
With Nick Chubb out, D’Ernest Johnson (19-99, 8-7-58) got another start, and he got a workhorse role with basically all the other backs out, too. The Browns got trounced, but not because of Johnson, who had a strong start running the ball, and then also added the seven receptions.
Hunter Henry (4-4-37-2) had a nice day with two touchdowns, and he’s up to seven on the year as a clear red zone threat. Kendrick Bourne (4-4-98-1 with 3-43 rushing) had a nice long touchdown as well, on a well-thrown ball from Mac Jones, who is playing at a high level. Jakobi Meyers (4-4-49-1) got the first receiving touchdown of his career, and Nelson Agholor (4-2-21) also was at four targets, as things were spread evenly. We even got a N’Keal Harry (1-1-26) flash with a nice play on the sideline.
Things were similar for the Browns, with the unconcentrated targets. Jarvis Landry (5-4-26), Austin Hooper (5-4-25-1), and Donovan Peoples-Jones (5-1-16) all saw five targets, while David Njoku (4-1-11) saw four and Harrison Bryant (2-2-18) got a pair. That’s about what you’d expect given the routes and the way this offense functions.
Signal: Rhamondre Stevenson — all four green zone touches, basically had the Damien Harris role but also his normal routes involvement for a strong 8 HVT as the total package
Noise: Kendrick Bourne — 4-4-98-1 with 3-43 rushing (has flashed at times, but is still the clear No. 3 at just 68% routes in a run-first offense)
Bills 45, Jets 17
RB Snap Notes: Zack Moss: 48% (-3), Devin Singletary: 38% (-11), Matt Breida: 14% (+2 vs. high), Michael Carter: 52% (lowest since W5), Ty Johnson: 32% (-10 vs. W9), Tevin Coleman: 22% (return, +5)
WR Snap Notes: Gabriel Davis: 52% (+1 vs. high), Cole Beasley: 16% (injured), Corey Davis: 84% (return), Jamison Crowder: 83% (+3 vs. high), Elijah Moore: 56% (-4 vs. W9)
TE Snap Notes: Dawson Knox: 84% (return)
Key Stat: Stefon Diggs — 208 air yards (season high, most in Week 10 through Sunday)
Buffalo came out throwing vertically, and Josh Allen did bounce back and notably had his highest passing aDOT of the season at 12.9. On the first play from scrimmage, the Bills went play-action downfield to Stefon Diggs (13-8-162-1) and that was a sign of things to come, as Diggs wound up racking up his most air yards this season at 208, which was his first game over 100 since Week 4. Even late in this blowout, they were throwing downfield to him, as he caught a 43-yard pass in the fourth quarter, almost like they just wanted to get him some numbers. I’m certainly not complaining.
No other Bill had more than three targets behind Diggs’ 13. Cole Beasley (2-2-15) was injured early, which opened the door for more routes for Gabriel Davis (3-3-105), and he caught a few downfield looks. Emmanuel Sanders (2-2-27 with 1-24 rushing) caught his two targets. Dawson Knox (1-1-17) had an end zone target on a scramble drill where Allen flipped the ball toward him wide open in the back of the end zone, and if it was on his frame it would have worked, but it was a little to the side and Knox couldn’t corral it. It didn’t go down as an official target because a late defender took a shot on Knox and was flagged for hitting a defenseless receiver, but Knox did have an end zone target here that won’t show up in most stats, and it was a very near miss. He was in a route on 74% of dropbacks, and if Beasley misses any time, Knox will be right back to a very solid option.
One interesting wrinkle with the Bills’ vertical passing is they didn’t throw much in close, instead rushing for four scores. Matt Breida (3-28-1, 3-3-22-1) caught an early touchdown and then rushed for a score in the third quarter, in his most extensive action of the year. He only played eight snaps total, but he brought an explosiveness the other backs have lacked at times, and I have to be honest as someone who has a lot of both Zack Moss (7-27-1) and Devin Singletary (7-43-1, 1-1-6) that I’m concerned Breida will carve out a role here as a third piece of a committee. Both Moss and Singletary also added rushing scores, as did Isaiah McKenzie, but Breida getting involved has me wanting to keep the Bills’ backs on the bench next week until we see if Breida’s Week 10 builds into anything more.
Mike White had a rough game, throwing four interceptions as the magic ran out. He was once again great for the RB receiving lines, though, despite the Bills entering the game as the defense that had allowed the fewest RB receptions in the NFL. Michael Carter (16-39-1, 6-4-43) led the way and Ty Johnson (2-2, 8-5-36) also saw big targets, but both were hurt by the return of Tevin Coleman (4-29, 3-2-4), who played 22% of the snaps. Carter’s routes fell to 38% while Johnson was at 35% and Coleman took 19%, so this is another backfield to be a bit cautious of right now, especially if White doesn’t start Week 11.
Elijah Moore (6-3-44-1) found the end zone late from Joe Flacco, but he ran routes on just 52% of dropbacks as Corey Davis (7-5-93) returned to a full role and Jamison Crowder (6-3-20) was at a season-high route share. It’s hard to understand why Moore wouldn’t cycle in with these guys more, but he basically split the third WR role with Keelan Cole (5-1-18) as Davis and Crowder were the clear Nos. 1 and 2 in terms of routes. With the way Moore is playing, I do think he’s likely to find his way into more routes soon.
Signal: Dawson Knox — routes on 74% of dropbacks in return, had a near miss on an end zone target that was wiped away by penalty (he got hit as a defenseless receiver); Tevin Coleman — returned to a 22% snap share, 19% routes, which cut into both Michael Carter and Ty Johnson; Corey Davis — returned to 90% routes and seven targets
Noise: Matt Breida — 50 yards, 2 TDs (only played eight snaps, so the production is pretty noisy, but I do think he could carve out a role in a way that is bad news for Zack Moss and Devin Singletary)
Colts 23, Jaguars 17
RB Snap Notes: Jonathan Taylor: 84% (+10 vs. high), Nyheim Hines: 28% (lowest since W7), James Robinson: 59% (return), Carlos Hyde: 27% (-52 vs. W9)
WR Snap Notes: T.Y. Hilton: 61% (return, +11 vs. high)
Key Stat: Jonathan Taylor — 8 HVT (tied for fifth most in Week 10)
This was my very funny joke that is so funny I also talked about it on Bananas and am now sharing it here.
Jonathan Taylor penalized for having an actual cannon on the field from which to shoot himself out of every single playHe’s a monster, though. Jonathan Taylor (21-116-1, 8-6-10) is legitimately one of the biggest storylines of the season, and we’re looking at an elite running back that I expect to contend for 2,000 yards rushing in the future, but who can also create plays in the passing game. His snap share was up at a season-high 84%, his routes have been stronger for the past five weeks than any of the first five, he has at least one green zone touch in every game since Week 4 and all of the team’s 17 RB green zone touches in that span, and the Colts have generally just committed to him being the centerpiece of their offense. It’s glorious.
Michael Pittman (5-5-71) had another efficient game, while T.Y. Hilton (5-1-5) was back in the lineup at a 72% route share but was inefficient. Zach Pascal (2-2-9) maintained his high route share despite the lack of production. It’ll be interesting to see if Pittman can keep seeing big target shares if Hilton can stick in the lineup for an extended period, because even at Hilton’s age he’s probably the biggest threat Pittman has had to volume.
James Robinson (12-57-1, 5-4-27) made it back from his foot injury, but wasn’t back to a full role, playing just 59% snaps. Carlos Hyde (2-4, 2-1-0) played some other snaps but you wouldn’t know it from his stat line.
Jamal Agnew (5-0-0, 3-79-1 rushing) had a sweet 66-yard touchdown run, showing off that kick returner ability with the ball in his hands. He still drew some targets again, too, but didn’t haul any in. Dan Arnold (7-5-67) led the team in the major receiving categories, as Trevor Lawrence struggled, and in particular continued to struggle to get the ball to the outside receivers. Marvin Jones (6-2-35) and Laviska Shenault (8-3-15, 2-6) are not usable, and for those of you who have been holding out, it’s definitely OK to move on from Shenault.
Signal: T.Y. Hilton — season-high 72% routes, five targets (seemed pretty healthy)
Noise: Jamal Agnew — 66-yard TD run, but also no catches on five targets, so it kind of cancels out
It was +10000 to hear the quote, "it’s definitely OK to move on from Shenault" From Gretch this season, tough scenes
I do not want to see D'Andre Swift carry the ball 33 times ever again. Good lord. That seems like the worst possible use for a player of his abilities and a great way to get him hurt. Fortunately, they won't be playing against Mason Rudolph next week. Odd as it sounds, I think it would be better for Swift if Williams were healthy to take those taxing between-the-tackles carries. I'm not sure what these coaches are thinking. You also saw similar last night from Shanahan, giving Elijah Mitchell 27 carries, including the final few meaningless drives. It's madness.