Stealing Signals, Week 3, Part 2
Stealing Signals, Week 3, Part 2
Disingenuous takes, plus Biggest Signals and Noise of Week 3
Part of the introduction yesterday leaned into this idea of disingenuous touting, but my god did I see a lot more of it in the past day. And while I do wish I could write something different in the introduction, there are reasons this has been on my mind a lot over the past week or so, and I can’t write intros about anything other than what is on my mind, so you’re getting a lot of the same concepts this week. I do sort of regret that, but it’s kind of like either I talk about what I think is relevant and is most on my mind, or else I don’t write an intro, because that’s how they work. If you don’t love when I get a little more negative, I’d suggest moving past the intro this week, and I promise I’ll get myself onto something better next week.
There’s a lot of things I’ve written about in the past couple years that have been pretty prescient, and I’m proud of that. Last week, I mentioned the intro on “motion is the future,” but I didn’t link to it because I was needing to work quick. That’s the link for anyone who likes reading back on stuff; I came across it this morning searching for something else, and this quote I included cracked me up:
Stealing Signals and Stealing Bananas have been emphasizing the importance of motion and misdirection and play action for like two or three years now. I feel like it’s super old news and I haven’t even discussed it much this year as a result, and then I see a chart like this and realize it’s still this hack that only a handful of teams are utilizing and they are obviously seeing results from it.
Last week, I took the condescending tone of “I don’t know how else to talk about this, this is taking so long,” so obviously my commentary that it was already old news last year is where the humor is in it for me. Unfortunately, we went through another offseason with at least roughly 50% uninspiring coaching hires, and we’re back here again. Anyway, that was last week’s intro, not going to go too deep into it again.
This week is a thing I’ve written a ton about and I’m unfortunately really bummed to have been so right on it. It’s the kind of thing that makes me question doing this content anymore at all, or how I do it, and what kinds of stuff I interact with.
It’s really, really tough to understand how much variance there is in all this. My whole point in this introduction is going to be that we ascribe way too much meaning to everything. But the issue is it’s getting worse with our connectivity and our society moving to bite-sized info everywhere. Football is perfect for this with its weekly schedule, and there have always been these descriptive-masked-as-predictive commentaries, but it’s just on steroids these days.
There’s a feedback loop where people want the analysis to match the results, and so it goes to this idea of consensus I was talking about yesterday, where claiming anything isn’t going to last makes you a contrarian, or a denier basically. Instead, we just continue to move further in this direction where it’s about who was right or wrong based on the very most recent data point; all fantasy football content is anymore is just claiming that the scores that have been achieved in the past week are all that matters.
But it flies right in the face of what we know about the sport in a macro sense. I wrote an intro after three weeks last year and looked up the top performers in the first three weeks the year prior, and it was laughable. It’s legitimately the same shit, every time, where fantasy analysts care far more about social capital than reality.
I’m not gonna look it all up again, but I just did a quick look at the first three weeks of 2023 and Kirk Cousins, Tua Tagovailoa, and Justin Herbert were the top-scoring QBs, Davante Adams was WR4 (he finished WR10, but the rest of his season more or less made people think he was washed), Tutu Atwell wasn’t far off Puka Nacua (but Atwell’s three games of 15+ PPR points were the only three he’d score all year), Tony Pollard was RB4 with three straight games of 15+ PPR points (he’d hit 15 in 5-of-13 games the rest of the way), etc. Honestly, it does look like the first three weeks were a little more indicative of the full season last year than what I found when I dug into 2022, but the point was never that we’re not actually seeing some useful things. Of course some of these trends matter.
What’s frustrating for me is not the obviously bad analysts. It’s become increasingly easy for me to spot people whose entire analytical model revolves around this game of “who was right” where they bounce weekly from one take to the other based on what happened in that week’s games. Again, when variance is the rule, that creates for some really crazy gymnastics, but the door’s always open for some new backward-looking take that’s like, “That dude was always going to smash!”
My issue boils down to how I’m seeing it with otherwise good analysts, who have been increasingly lining up to give up and lean into this crap because the masses will adore them for their charts that say nothing and their snappy one-liners that obviously remove all context. And this stuff is truly making people dumber, and then one of the ways it impacts me is I get asked about it and have to start from an expectation that I’m wrong and try to rewire the thought, which honestly is exhausting. To be clear, I always, always say in these rants that I don’t know shit, and I truly mean that, and I hope that’s the reason you’d listen to me at all, because to be clear I’m saying nothing other than don’t listen to snake oil salesmen, not that my good-faith attempts to find answers are going to be perfect. I just know they won’t be complete and total bullshit in service of likes and views and clicks from the hoards of people who are only engaging with that content because it fits their narrative. That’s the whole feedback loop, right? The people clicking are the ones looking for confirmation bias.
But also, as I write over and over again, they need the other answers to reflect the results they’ve seen, even if it’s just three weeks where we barely got teams to throw a little bit. It stifles any creative thought. After three weeks, you will be mocked for suggesting something other than what’s happened might happen. You’ll also be mocked for any contrarian preseason takes that weren’t accurate. “How could you think something other than the consensus would play out in this situation?” the disingenuous ask, while ignoring all of the other situations where the consensus was blatantly wrong. But that’s the thing that makes them cling to the thing that was “obvious” harder. It’s the only plank of wood floating down a river of uncertainty and they are drowning. It’s a coping mechanism to keep them from discussing the horrendous other takes they had, but the coping mechanism becomes all they can do, and then they pass that off as real analysis, and again it is just making the audience — who eats this shit up — dumber. Me writing this up sounds like sour grapes in some way, despite me being in a great position as an analyst through three weeks.
I’m just annoyed that we’re regressing as an industry, and that there can’t be any nuance in a field where there absolutely has to be nuance. Everything we know argues that’s the deal. We gotta be able to speak in the gray area. But black-or-white thinking always wins the audience. None of this is new, obviously, but after years of this, it’s truly — to my eye — getting much, much worse. There are analysts who do this only sometimes, and they used to drive me nuts five years ago, and now I’m like, “He’s one of the good ones,” because the window has shifted beyond that person and they have been consistent.
I am extremely hesitant to give a specific example here, because the way we think, it feels like this whole intro must be about that one example. The below example was absolutely not the catalyst, and separately I got in a whole conversation about Jayden Daniels on the timeline this morning, as well, and how Kliff Kingsbury clearly held him back despite Daniels finding success through two weeks, but after an eruption in Week 3 when they did finally incorporate the vertical passing, people are going backward to say that the previous criticisms of Kliff not incorporating vertical passing was just to bring Daniels along slowly — a hilarious irony because, of course, he looked so much better when that stuff was in there.
It’s of course all just taking the Week 3 result and then trying to explain everything up to this point based on that, ignoring that the Commanders got crushed by a Bucs team that just lost to the Broncos, and likely would have lost to the Giants if not for a kicker injury, to say that the offense has actually been good all year. And some of that makes some sense until you think for just a couple seconds about how throwing downfield isn’t inherently some thing that inexperienced QBs can’t do. That’s just an assumption being made to service the narrative that what Kingsbury did up to Week 3 must have been right. By contrast, Shane Steichen has had the less-experienced (younger, played three fewer college seasons as a QB) Anthony Richardson throwing downfield creatively since Week 1, and it has been an effective way to put him in position to succeed, although the unfortunate reality there is he looks right now like he’s probably only going to be as good as his surroundings.
But one of the things Kliff clearly should have been doing in the first few games is at least designing a few half-field reads for deep shots, even if coached as very unlikely to be completions unless there is a clear one-on-one, almost like a baseball pitcher throwing an intentional ball to set up the next pitch, because it keeps safeties honest and can open up space underneath, but it also gives Daniels the upside of impacting the game in a way he’s clearly capable of, both from his prospect profile and the Week 3 results. Anyone watching his downfield throws Week 3 and acting like he couldn’t have done that in Week 1 or 2 and needed to be brought along slowly is just doing gymnastics, obviously, and there was this whole pushback that 125 and 122 air yards passing in the first two games were necessary stats to protect Daniels, and Well Actually the offense was good anyway.
To be clear, the take I feel strongly about after Week 3 is that Daniels made the offense look half decent the first two weeks despite poorly thought out design — as always, two things can be true — and once the design made a little more sense, he looked like a goddamn superstar, because they gave him the chance to impact the game with vertical passes, the plays last night that most clearly determined Washington’s ability to win that game. One person correctly pointed out to me that I was harping on a thing that didn’t really matter at this point, and that’s totally true if they keep the vertical passing in the offense, but man this idea circulating that Daniels couldn’t throw downfield by means of developing him is just so, so dumb, and I’m hammering this point because of the mentality it emphasizes, not the specifics around Daniels. Showing vertical passing would have protected him more. It makes him way harder to defend. And throwing a couple downfield passes isn’t rocket science a rookie can’t handle until his third game. That’s all just in service of the trends that exist, which must be right, because they are what happened.
Anyway, that wasn’t even supposed to be the example, but I got in that conversation publicly, so I added it. But the example is a little more straightforwardly disingenuous, and I just want to reiterate that these are legitimately everywhere and this is just one, and I’m not trying to come at anybody. But I came across a take that was shitting on Bijan Robinson’s fantasy results to date, specifying back to the beginning of his career. And this is obviously a guy who played in an offense with Arthur Smith last year — which meant a virtual split of touches — and has now been in an offense through three weeks that has a QB who is working back to health and has been extremely suboptimal (-9% PROE or lower in all three games).
The point I’d make here is not that Bijan will definitely be fine, because there are reasons for concern, and I’ll discuss in my writeup of his game below how this week was not ideal for him. The point is that after a down game, this is the take, when it’s plainly obvious that the 20-game sample we have to date is not reflective of what he could possibly be. And I’ll go a step further and say if you are a fantasy football analyst, it’s in my opinion absolutely your obligation to not actively misinform on that point for social clout, if you want to be taken seriously. If you think critically for literally three seconds, you will be like, “Oh, yeah, I mean those 20 games probably aren’t going to be reflective of who he really is,” for all of the obvious reasons.
Acknowledging uncertainty is base-level stuff here and has to be done, but we know why it’s not, and it won’t be. Again, the response to these horrendously involved breakdowns of minor comments should be, “It’s just a joke, man, lighten up.” Which, yes, definitely. I sound like an insane person in this whole intro. And I’m certain I’ve done that exact kind of thing before. I’m absolutely a hypocrite here. There’s just no question.
But that doesn’t change the macro point, which is that this stuff does great in terms of earning the social clout, and so it has become the consistent and dominant theme of all content. The masses want what they want because the masses are idiots. Or more practically, this is a hobby, and there’s comfort in the consensus, and most people don’t want to learn lessons or think critically in their off time.
But that’s why I believe pretty strongly it’s incumbent on the people whose job it is to do this stuff to resist the urge, and not actively misinform. But of course there’s always a battle for eyeballs and ears, and so telling people what they want to hear is lucrative. I’m the fucking dork telling people not to chase clicks and take themselves more seriously while their current strategies are working just fine, thank you.
It’s not going away, and I’m going to just tilt it harder. Companies are actually going to hire bigger and bigger bullshit artists into bigger and bigger spots, because in some ways they take the heat so other people don’t have to. But when I talk about “disingenuous touting,” this is what it is. It’s making some point that will earn social clout when it takes literally three seconds of critical thinking to recognize why that point doesn’t really hold water.
My old mentor “Fantasy Douche” (yes, I had a mentor named that) used to say that you could look at last week’s box scores and figure out 90% of the takes that you were going to see. There are so many analysts just creating charts and “advanced” bullshit that do nothing but find ways to regurgitate what we already know happened if we just look at the points, yards, and touchdowns. It’s entirely for show, because it’s what the audience wants. The charts might as well be a list of the fantasy points that have been scored, except then you’d need to acknowledge some players that might not be as advantageous to acknowledge, so instead it’s just looking for the stat that allows you to highlight the thing you were trying to highlight, that everyone knows happened. And then finding a way to clip it and share. “Jauan Jennings was good in Week 3.” It’s so goddamn stupid.
But again, the masses want those explanations that fit with what they’ve seen, because no one wants to be told why the stuff that’s happened isn’t going to continue happening. Even though that’s literally what always occurs. But then next week, we’ll just pivot. It’s a lot of chasing your tail because there’s no accountability, and you win by having thick skin to anyone who might call you on your bullshit and just continuing to throw more takes at the wall so that there are those that survive and you can cling to those as evidence you’re sharp. It’s no wonder these people also tout shitty RBs; their whole content philosophy ignores efficiency, and is only about volume. “Being good at things actually doesn’t matter.”
And as the newer generations get better and better at going viral versus saying anything meaningful — and don’t get it twisted, as much as this is an “old man yells at clouds” rant, it’s absolutely the case that the younger generation sees social capital in this way, where it almost literally does not matter what you say as long as it’s a banger — it’s legit awful for a fantasy industry where accountability does lack because analytical skill is so opaque. So the people who win are just the people who find increasingly acrobatic ways to regurgitate the box score results and claim anyone who sees anything beyond that is the actual fraud. I got smashed this morning because I wouldn’t admit that Kliff Kingsbury has actually done A Good Job so far this year, and I mean I was already preparing to write (and will below) that I think there’s some stuff I missed in what he’s done before. But the results don’t actually speak for themselves in Weeks 1 and 2 when you had that dynamic of a weapon that we saw in Week 3 and you were parking the Lambo in the garage.
But that doesn’t matter: It becomes an argument about process, and then eventually the future results don’t even matter. All that matters is if you can prove you’ve won an argument based on the past results — judged by moronic consensus opinion, with no ability to parse how the future might be different — you’ve already won.
Anyway, I just need to cultivate this stuff better. I felt like I had over the course of some years, and I’ve gotten feedback from many of you guys that suggests I basically don’t need to be writing intros like this, because I do have this space, and it’s successful, and you guys are very much not the problem, and if it makes me this unhappy I should just focus my work here. It’s absolutely part of my thought process, for what it’s worth, and I am just so fortunate to have this space. There’s just always that constant push and pull to try to expand and find other people who would like to be part of this community, but to do that you have to wade through some dangerous waters, where the takes could pull you under.
But I keep writing about this because — like I said — these intros have always been reserved for what I think is most actionable, and while I’m reiterating the same shit over and over, I do still think one of the very most actionable things in fantasy right now is simply understanding that an alarming degree of the analysis is claiming everything that’s occurred is all that matters when the golden rule that we know is nothing lasts in the NFL. It’s truly complete nonsense to try to reconcile the micro takes and the macro realities, and if you feel like I harp on it too much, I’d actually suggest that limit more or less does not exist.
I’m repeating myself, but again, if you think critically about it for literally three seconds, it’s truly completely insane, and the only possible answer you can come to is analysts get the wrong feedback from their audiences, lean into that, and no longer actually seek truth and accuracy. And based on how hard all this stuff is, and how random the results can be, you can kind of understand why that happens. But again, these are people who are ostensibly experts in their field, and they are ignoring the most glaring and obvious known variables that we learn week after week where we need to leave room for the unexpected, and they just pivot and do it again and pretend like the last week didn’t exist.
Alright, I’ve chased my own tail enough times in this intro. Now I’ll go write all the things I think are true, but to be clear, I’m frustrated because I literally don’t feel like I have the space to be contrarian anymore. It’s way, way more likely I’m going to be mocked for a good contrarian take than it would work against me to have a bad consensus one. It makes it hard to do my work, and yet I have unburdened myself with this intro so that I can do it. I need to speak my peace and then I’ll get to work, it’s just how I’m wired. Hopefully I have something useful and interesting today.
You can always find an audio version of the posts in the Substack app, and people seem to really like that. You can also find easier-to-see versions of the visuals at the main site, bengretch.substack.com.
Data is typically courtesy of NFL fastR via the awesome Sam Hoppen, but I will also pull from RotoViz apps, Pro Football Reference, PFF, the Fantasy Points Data Suite, and NFL Pro. Part 1 of Week 1 had a glossary of key terms to know.
Panthers 36, Raiders 22
Key Stat: Diontae Johnson — 38% TPRR, 0.94 WOPR (highest in Week 3)
After PROEs of -8.5% and -12.7% in Weeks 1 and 2, Carolina came out throwing in Week 3, posting a +5.9% PROE as Dave Canales clearly felt different about what he could call and the team could execute with Andy Dalton under center. It’s sort of hard to overstate how massive that shift is, and if it sticks, we’re looking at what could have been immediate Signal that the Panthers went from perhaps the biggest “have not” in the NFL to a clear “have.” Dalton had a massively lower time to throw (2.17 to Young’s 2.68 on the season), and his 18% pressure rate reflected that (Young’s was 31%, and the Raiders have a good defensive line). Diontae Johnson (14-8-122-1) dominated volume and is a clear winner, especially with Adam Thielen (5-3-40-1) suffering an unfortunate hamstring injury on a long TD reception that will cost him some time. But the downstream impacts are perhaps more interesting, and we’ll get into those in the next bullet. First, I want to note that Dalton played legitimately great. Early on, he ripped a 3rd-and-14 in-breaker to Diontae for a first down, which is notable in that it came in a clear pass situation and the defense was obviously going to be dropping. That requires good processing and delivery; good quarterback play. The very next play, he split two zone defenders with a bullet to Xavier Legette (3-2-42) on a curl sitting down behind them, and it opened Legette up for YAC and he had room to run for a 35-yard gain on the play. He capped that drive with a touch pass at the goal line for an over-the-shoulder TD for Diontae. The rest of the game featured some big passes, including the TD to Thielen up the seam, but that was the sequence when it became clear this offense would look quite a bit different after the QB change.
Outside Johnson, Legette looks like a clear add. Thielen’s injury should open up snaps, and it’s possible it goes to some slot guy I’m not even aware of yet, but I think we’ll probably see them do some creative stuff to get the young guys on the field more. That includes Jonathan Mingo (4-3-18, 1-2), but while he had a little catch-and-run at one point, he’s not really a dude, and that gives Legette a shot to matter. Tommy Tremble (3-3-29) seems to have earned the strong starting role over Ja’Tavion Sanders (1-0-0), so I’m not sure there’s anything at TE, either, which further emphasizes the volume Diontae could see, and how Legette’s path to being the second meaningful weapon is wide open. Legette did have a rough play on a 2-point conversion where he got an open target but jumped higher than necessary and it carried him out of the end zone.
Miles Sanders (7-17-1, 2-2-13) ran more routes than Chuba Hubbard (21-114, 5-5-55-1) and got the short TD, but Hubbard still dominated the overall volume, and his +33 RYOE dwarfed Sanders’ -10, emphasizing what the YPC numbers show that Hubbard looked much, much better in the run game. It’s not ideal that Sanders stole four HVTs, but Hubbard had six, including a slight edge on green zone touches in a split there (3 to 2). There were just way more Team HVTs with Dalton under center, which is very positive, and Hubbard looks like a very solid RB option going forward, with Sanders a plausible plug-and-play option given the routes and green zone work. The other clear winner here was Jonathon Brooks, who suddenly looks like he could eventually return to a solid RB situation after concerns it would be a disaster.
In a blowout loss, the Raiders looked rough, with Brock Bowers (4-3-41) and Davante Adams (9-4-40) both having quiet games. There’s nothing of major concern here, and it’s probably a testament to Bowers that his down game looks like something I might be vaguely happy with if I saw it from another TE. Jakobi Meyers (9-7-62-1) had a nice game, and Tre Tucker (9-7-96-1) caught a very late TD from Aidan O’Connell. Tucker looks like a guy who will get some deep shots, and he caught a 54-yard pass in the first quarter from Gardner Minshew, but he also got extended work late here, with a total of three receptions from AOC in garbage time.
Signal: Panthers — +5.9% PROE with Andy Dalton after -8.5% and -12.7% in first two weeks; Diontae Johnson — 38% TPRR, 0.94 WOPR, and will benefit strongly from the QB change; Xavier Legette — should see a bump in routes with Thielen out, path to being the second meaningful weapon in the passing game is wide open with Mingo and Tremble as other big route runners; Panthers — 10 Team HVTs (Chuba Hubbard led with 6, and was the more efficient runner and should lead, but Miles Sanders also had 4 thanks to 38% routes and 2 green zone touches, plus Jonathon Brooks’ situation just improved upon his return)
Noise: Zamir White — 22% snaps (he’s rotational, and it’s not great, but this was more script-related, as his final touch came on first drive of second half in blowout loss); Tre Tucker — 9-7-96-1 line (caught an early deep shot, and he’s a deep threat who will do that some, but paired that with three late receptions in garbage time from the backup QB, including a TD)