Stealing Signals, Week 5, Part 2
Stealing Signals, Week 5, Part 2
Tilting, never playing for floor, and the Biggest Signals and Noise
For those of you who are newer around here, one of the things about this piece is how I have a bit of a love/hate relationship with it. The process of getting through all the analysis is very useful for me, but while there are people who are frankly better equipped to answer the bell for something like this consistently, week in and week out, it’s something for me where I find it excessively laborious some weeks, and at various times, depending sort of on my mental state and mood. And while I’m not self-important enough to act like I’m doing some kind of life-changing work here, because this very much is not rocket science, I have worked a lot of different types of jobs, and those are the weeks where I miss data entry, or more manual labor, because this isn’t something I can just brute force my way through by trying harder, if my focus just isn’t there, and sometimes it’s just not the day after watching 12ish hours of football consecutively. Sometimes you just need a break mentally.
Anyway, that’s the deal when you start to get these emails at like 2 a.m. ET, like you did last night. I’ll still answer the bell, and I hold myself to sort of an impossible standard and so I hope the work is still OK when it finally gets to you. But it probably doesn’t help that I wrote the extra TPRR piece last week, and then also decided to fire up a Part 2 to Input Volatility on Saturday that I called a “quick” post and went 1,200 words deep through three different topics, so I wrote up five posts last week with the two bonus posts in addition to three regular, long ones. And then I do the podcasts and everything else and it just becomes a lot, not necessarily to get through one week, but then the way the work compounds, because the week ends and then it starts anew, with all day research Sunday and then writing Monday and Tuesday.
These aren’t complaints; I love you guys, and I love what you do. I want to explain the lateness of last night’s post, and I also want to note that some weeks I’m not able to help with the random one-off questions I get. Long-time subscribers know that I don’t answer a ton of those anyway, just because it’s impossible to do it all, but when someone who hasn’t sent me a lot of individual notes has one specific question, I do try my best to make time to give a good answer that one time. And I got a few this week I did want to get to but it’s just not something that can happen some weeks, and so it’s sort of like tough luck and bad timing that you pinged me this week. I gotta keep the main thing the main thing, and these posts don’t write themselves, and again, it’s not hard work, but you also don’t activate useful analytical parts of your brain without some effort.
I envy the younger analysts I see with so much vigor this time of year, while I’m on Year 9 of this shit. I also think they are green in a lot of ways, and my work still has value, or as I’ve always said I’d just stop doing it.
I also went with this for the intro today because yesterday’s was long and led to the games being really tough to get through. I wanted to get right to the games today. There are two things I want to discuss, though, so I’ll do them quickly.
The offensive environment in Week 5. It was just so cool, and I alluded to it in the game writeups yesterday, but the less-mobile veterans are doing some cool stuff, and I feel better about QB play as a whole than I have in a while. Teams are also finding opportunities to take downfield shots, which we’d lost a little bit, but they are there a little more here in 2024 as defenses continue to evolve and do some different things. I’ve written about all of this in different ways this year and don’t have all the answers, but I absolutely think we’re seeing this sport evolve and the offensive “counterpunches” start to make themselves more well known, and it’s that evolution point especially because you do still see some of the really poor offensive games when some teams are uninspired, so even on the weekly level there will be these gulfs between the “haves” and “have nots” that are just so clear. I’m really in love with the sport right now even more than the fantasy-specific stuff, and to the point about yesterday being difficult to work through that I mentioned above, I was done with the intro early in the morning. That wasn’t difficult. I was eager to get up and write about that stuff. It’s stuff like parsing Tank Bigsby, and coming to the realization that everyone is going to be too in on this guy but he’s probably just a contingency play for real upside because Travis Etienne isn’t actually going to get displaced in the way other starters can — we see how difficult it’s been for guys like Bucky Irving and Chase Brown to displace far worse starters, because teams do still typically need a mountain of evidence a guy is bad to bench him, and Etienne hasn’t been that — but that sure, there is still upside for Bigsby because he did catch passes in college and could theoretically consolidate if Etienne missed time. Anyway, I feel good about that specific take, but I probably spent 20 minutes yesterday researching and thinking through it, and then another 15 trying to figure out how best to write it up so I get the message across, and that was only one portion of that game I wanted to cover. And that’s not like, “Oh geez that’s so difficult,” but just a note that it takes time to get through the whole league at that level, and that’s part of why you do see so many surface-level takes (I’d guess most analysts go deep on a few spots, and do have really strong takes on those, but that we all sort of have blind spots). And then I get these one-off questions about players I’m never really on and never really write about and it’s like, “Break this one down,” and honestly I just don’t have an opinion on, like, Zamir White. I have not thought about what he needs going forward enough to offer anything of use, and it’s why the article gets hard to write sometimes because I just really don’t have the bandwidth to update my priors in Week 5 on Hunter Henry, because broadly what I’m saying is if you want to do it right, it doesn’t really matter who it is, this shit is just dense (and you’re also usually sifting through like three gallons of noise, trying to figure out why people are saying some thing and what you’re missing, and usually you’re not really missing anything and people are just jumping the shark for attention, but that’s a whole other element).
I wanted to give a quick note on the Guillotine Leagues I’ve been playing that have been an absolute blast, and particularly one I got chopped from last night when Xavier Worthy scored that rushing TD (the other lowest scorer had him and I was nearly safe), because losing this team legit pissed me off. I’d made a shrewd bid on Jayden Daniels after Week 1, and another shrewd add of Kareem Hunt, who I decided to freaking sit this week because I also had added both Trey Sermon and Tyler Goodson and I convinced myself the high-floor play was to play them in tandem, because I had Daniels and Dalton Kincaid and six legit WRs, so I’ve just been trying to get by on RB points. And the reason I want to share my frustration at losing this team is how it’s also a lesson I’ve learned a million times — nothing good ever comes from playing for floor, even in a format where your only damn objective is not finishing last on the week. I knew Hunt had a much higher ceiling, and I’ve felt great about my early read there and hope a lot of you guys are rostering Hunt and feeling good about it, too, but I felt like there was a very specific outcome where Goodson played more than I thought, and specifically got all the HVTs from Sermon, and then also Hunt didn’t have as much of the backfield as he did last night, and my Sermon-Hunt tandem really performed poorly and it put me in jeopardy. Because the rest of my lineup was strong, I wanted to capture all the HVTs in the backfield in Indy, and I did that, and Goodson notably had a fine game, and the tandem was very good overall for what I expected and needed, giving me 27.0 PPR points. But you only get to start eight players in Guillotine Leagues, which meant that I was overconfident in my other six players being strong when I made this floor decision, and Hunt’s points on my bench would have kept me alive if I played him over Goodson, so my hyper-specific tactical maneuver — despite being an accurate read of the Colts’ RB split and total value (I would have felt like I made a very sharp call if you told me I’d get 27.0 PPR points from them before the week) — cost me this team, which is now just dead and everyone else gets to bid on my players and I feel like my soul has been ripped from my body. And for those of you wondering, it was Christian Kirk, Diontae Johnson, Deebo Samuel, and Dontayvion Wicks who were the WRs I went with, so all those Wicks drops, and the missed Kincaid connection, and McLaurin dropping the potential second Daniels pass TD, and Jahmyr Gibbs on bye, and Deebo getting no run even in a close game late as the Cardinals basically kept the 49ers off the field, all conspired to doom me, while Jaxon Smith-Njigba got some late volume and was clearly the name I should not have sat, but I’d decided to put him on my bench (along with Jordan Addison, and of course Hunt). All of that, and I lost by three points to a guy who started Tank Bigsby. On second thought, Guillotine Leagues suck. (Also, I’m going to fire up an in-season one probably next week, so if you’re interested in drafting maybe next Wednesday night and then playing against me the rest of the way, keep an eye out for more information about how to sign up next week.)
But the point of that story was never, ever, under any circumstances, for any reason, play for floor. Just never do it, even in the formats where it feels like you should. I’ve talked about this before, but it’s all a false sense of control. You can’t predict all the factors you think you can — like I thought I was obviously going to get WR points this week — and typically enough stuff levels out that it does feel like you were right, but that’s because what happened to me is rare, where I basically hit on this parlay from hell of all the down weeks and missed connections at once, on the week where my first-round pick was on bye.
But that’s the lesson that’s been true time and again in my experience, and when things do run a certain way I almost always wind up with the same conclusion — I got complacent and too convinced I could control some kind of floor analysis instead of just playing the best plays. Not only did I really like Hunt, he projected better than Goodson meaningfully, and while you could say I was let down by projections a bit when I trusted them to go Wicks over JSN for that last spot, I would have been fine if I had just played all the best plays I could and dealt with the ramifications. But I meddled, because I have a big brain and I know better, and even though my read on the situation I meddled in was hyper-specifically accurate for all the right reasons and that outcome was great, I still fucking lost because of the decision. Because all the variables are too great to parse for our dumb, tiny brains.
Let’s get to the games, since I started this by arguing I wasn’t going to write an intro, and I’m still in way too deep. 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.
Broncos 34, Raiders 18
Key Stat: Javonte Williams — 18 touches, 111 yards, 7 HVTs (tied third in Week 5), 6 MTF (tied fourth in Week 5)
The Broncos rolled over the Raiders, and Javonte Williams (13-61, 6-5-50) got going, forcing 6 missed tackles to bring his total to 11 over the past two weeks, second most in the NFL behind only Tank Bigsby after his Week 5 eruption. He also went over +10 RYOE for the second straight week, and while he’s still in the negative for the season because of a tough Week 2, he moved further in the right direction purely as a rusher with the +14 in that stat this week as well. But it was the receiving side that was most exciting with the HVTs and efficiency there, though I do have to note his green zone role is very frustrating. Twice, he got the first touch for the team after they got down in the green zone, but the first time Robert Spillane make a nice play to fill the hole right as Javonte committed to it, and then Javonte came off the field for Jaleel McLaughlin (6-22, 4-3-3-1) for the next two downs, and McLaughlin made a sweet play on third down to catch the pylon for a touchdown on a swing pass. The next time, Javonte got a carry from the 5-yard line and had a strong run through contact down to the 1-yard line, coming up just shy of the goal line for the second straight week. On the next snap, Bo Nix pulled out from under center like he was ready to go to a handoff, which I suppose could have went to the fullback, but seemed likely to be a carry from the 1-yard line for Javonte. Instead, the play was blown dead to defensive offside, and then from the half-yard line the playcall shifted to Nix just keeping it and going over the top for the rushing TD. Javonte’s peripherals are improving, and for the fifth straight game he played between 35 and 40 raw snaps (if you’re wondering why I’m using raw snaps there, it’s mostly because I noticed it when looking into Javonte jumping around from as high as 66% to as low as 51%, and I find the stability of his raw snaps fairly notable actually because Sean Payton obviously likes to spell his backs but is keeping Javonte in a very solid lead role, and his snap rate seems to just rise in lower-play-volume games to accommodate that). Jaleel continues to hit on some nice plays in his change-of-pace role, as well, but has not played more than 24 snaps in a game, though his 37% in this game was a season high. It sort of is as we expected, except we’re pacing for fewer RB targets than I thought was possible; the hope is we just get more passing as we move away from the rough start, and Nix gets more reps.
Courtland Sutton (5-2-32) led the downfield weapons in targets, and no one else had more than two in a game where we did get 10 targets to the main two RBs on just 27 pass attempts. Troy Franklin (2-1-20) remains the most interesting secondary name to me here, as he got a 33% TPRR and 65 air yards on a very limited role still here, and dropped a potential long TD.
On the Vegas side, Gardner Minshew was benched relatively early, with Aidan O’Connell ultimately throwing more passes, but completing fewer, and for 43 fewer yards. In other words, I’m not sure AOC did anything to argue he should start next week. Minshew at least got the ball near Brock Bowers (12-8-97-1) early, so he could elevate over a defender and then add a bunch of YAC for a fantastic touchdown from a guy who just looks like he’s so much better at football than everyone else on the field. He’s just an incredibly skilled player, and I think I wrote after Week 1 or Week 2 how he makes everything look easier than it is, which is probably the best compliment I can give a player.
Jakobi Meyers (9-6-72) also looks good sans Davante Adams, and then Tre Tucker (5-2-18, 2-8) got 5 targets and 86 air yards, and is clearly the third weapon here right now. DJ Turner’s (4-1-12, 1-5) 71% routes are also notable.
Alexander Mattison (15-38, 3-2-23) ceded some work to Ameer Abdullah (5-42-1, 3-3-9), who scored a short touchdown late but also added a long TD run in this game. A lot of Abdullah’s work did come late in comeback mode, and Mattison would presumably still be ahead in the early-game pecking order. This is of course one of the very least-inspiring backfields in football.
Signal: Javonte Williams — still in clear lead role, very solid PPR day in better offensive environment with 6 MTF and 7 HVT; Troy Franklin — still only 19% routes but 33% TPRR, 65 air yards, had a long TD opportunity but dropped it (setting the drop aside, he’s earning opportunities when he plays); Brock Bowers — continues his clear breakout and is a strong bet to finish as the overall TE1 (possible buy high still in some leagues)
Noise: Tre Tucker — 18 receiving yards (86 air yards in deep threat role, also got some rushing work as he is clearly part of the offense); Ameer Abdullah — 42 rush yards, TD, 44% snaps (TD came very late in comeback mode, and obviously Zamir White was out for this game, while Abdullah’s a vet who is unlikely to get every-down work)