Stealing Signals, Week 2, Part 2
Late games, plus Biggest Signals and Biggest Noise of the week
It’s dual intro day, where I had an idea, but then something else more pressing popped up, so you get both.
First, there’s this paradox I sometimes reference that I’ve been spending more time contemplating lately. Teams in the NFL are doing everything they can to disguise their schemes, and their tendencies, and to make it so their opponents can’t easily gameplan for them. Meanwhile, we in the fantasy football space have the temerity to believe we can boil their offenses down to simple trends and use past data to predict what’s next. Those things don’t reconcile!
I actually thought of this while watching Chris Olave run a shallow cross on a third-amd-short early in the first Monday night game last night. Olave’s routes are typically running into the intermediate to deep areas of the field, so while it’s at all a crazy concept to just have him run a drag, it also probably wasn’t a coincidence it was a third-down design where the Saints were hoping to catch the called defense napping a bit, losing account of the guy they wouldn’t necessarily expect to be passing off through underneath zones on that play. And it was successful for the conversion.
That’s a super minor example, and it’s just one play. But we’re also not talking about huge samples that we’re making our reads off of, and there are times where teams find something in a matchup and pull that lever a handful of times. You’d think roles and underlying usage would be pretty consistent, but even that stuff can be impacted by the ways teams might gameplan from a run/pass or personnel perspective based on the opponents’ strengths, individual matchups, player health, or other factors.
And even when it’s not a wholesale gameplan adjustment, little stuff can be skewed with an example like the Olave one, where — if my big assumption about this being a specific third-down play is accurate, which it very well may not be — stats like his aDOT for that week can be impacted by the team specifically drawing something up outside of their normal tendencies.
As always, I’m just hammering home that our confidence level in most things should be low. We’re trying to set ourselves up to benefit from a number of different possible outcomes, not see the future. That said, some of the best analysis you’ll see on a weekly basis is where people really nail matchup-specific stuff, and essentially predict what the gameplan is going to be, and how an offense might attack a perceived defensive weakness, or something like that. You also see — more broadly — people understanding that sometimes a guy who hasn’t been used much lately, but is clearly in the long-term plans, might actually be “due” in a gambler’s fallacy sort of way, because this isn’t a role of the dice, and the teams might actually be intentionally looking into that sort of thing (i.e. getting a guy going, or else just doing something they haven’t shown on tape yet, which is essentially what I was doing with my Christian Kirk commentary in Input Volatility last week).
Anyway, what we’re trying to do in this game is identify what these offenses are, and how they operate, and then what they might do for curveballs, or how they might evolve as their opponents adjust. But the point of this intro is that’s also what their opponents are trying to do, and they know that, and they are trying to not be predictable. It takes a lot of hubris to think you can figure those things out with just a small amount of research. It’s one of the reasons I don’t like doing late-week content here, because I can’t dig into it as deep as I’d like when I’ve already done Stealing Signals in the early week (we have human limitations, obviously), and I know half-assing it isn’t going to move the needle.
There’s just so much we don’t know, especially in small samples. And as part of that, it’s important to keep a focus on a longer view, where we’re not overreacting to something like Breece Hall having a very quiet Week 2. Around here, we try to draft players for whom the weekly specifics aren’t necessarily make-or-break, because the idea is they are the cream that will rise to the top regardless of the twists and turns in their team’s season.
My second intro for today is just about the RB position. I didn’t talk as much this offseason as I did last year about the “RB Frankenstein” concept, and how you can try to replicate full-season RB production of legit top-10 workloads — basically everything but the true Legendary RB seasons — through spot starts and sharp navigating of what’s available on the waiver wire.
And one of the big pushbacks is everyone is trying to add RB talent, even and perhaps especially if they draft RB-heavy early and have hit some roadblocks and are now thin. So in stretches of seasons where there is maybe only one good RB, or only a couple ways to play things, it gets tough.
But I also want to talk a bit about what’s come available here in 2023. Obviously Kyren Williams was a big hit in Week 2, but also Justice Hill and Gus Edwards posted usable weeks, and look like they’ll continue to for the near future. We also got Zack Moss posting a low-end RB1 week, and some people did make that start, though it’s not something I was ready to dive in on personally (I guess I did note Deon Jackson’s work last week was unlikely to stick).
Jerome Ford also had a top-10 RB week, but obviously not in a way that was expected. He becomes a major pickup this week, which we’ll discuss more below. We’ll also talk about New Orleans, with Jamaal Williams getting banged up, and then I wanted to talk a bit about a spot from yesterday that I didn’t really elaborate on, because it seems like David Montgomery might miss this Thursday’s game (and perhaps more?).
In that scenario, I think Jahmyr Gibbs is set for a big workload, and he could be an absolute smash in the short term. But Craig Reynolds looks like the guy who might play alongside him — all four of Reynolds’ touches in Week 2 came in the fourth quarter, after Montgomery left, as Monty’s last touch came late in the third. Reynolds is one of those waiver wire plays where if you miss everything else, you still have this option that probably won’t be heavily targeted on the wire that is startable in a bind.
Although, maybe he does get targeted thanks to the Montgomery injury and some skepticism on Gibbs’ role. There are still a huge number of what are essentially “streamable” running backs across the fantasy landscape right now, including rookies like Tyjae Spears and Roschon Johnson already looking playable in a pinch, later-round targets like Raheem Mostert and Brian Robinson both posting top-five RB weeks in Week 2 and being obvious starts going forward, and still other rotational backs like Jaylen Warren and Chuba Hubbard averaging between 9 and 10 PPR points per game through the first two weeks (a nice perk on top of their obvious contingent upside).
But ideally, you’re not needing to go that thin with it. I was actually just looking at some of my best ball stuff, and my top-scoring BBMIV team is actually in 114th overall right now with 317 points, and it’s a Zero RB build. The interesting note here is it has Warren as the RB3, but it hasn’t used his score either week — nor has it used a score from the RB1 or RB2 (who are going to ultimately sink this team, because it’s my one BBM share of Dalvin Cook and then A.J. Dillon) — because it’s gotten points from Tyler Allgeier, Mostert, and Edwards. It’s different in best ball, obviously, but the point is to argue that if you’re in a position where you have a Warren scoring in the 9-10 point range and he’s not even your optimal play, that’s ideal.
But you have to stay diligent. A big part of this in seasonal is recognizing it’s a carousel. Alvin Kamara was suspended, then Kendre Miller got banged up, so Jamaal Williams looked good, and then by Week 2 Tony Jones is scoring two touchdowns. Of course no one had that, but it serves to show how quickly RB depth can be accessed by NFL teams. In Miami, De’Von Achane hasn’t been ready, while Jeff Wilson hit IR, so Salvon Ahmed has found himself in the RB2 role already, and posted 7.1 PPR points this week. That’s obviously not huge, but it’s Week 2, and it’s a RB that was drafted in roughly 0.0% of the millions of fantasy drafts done in 2023, and he easily out-scored Najee Harris and Alexander Mattison.
One last thing to emphasize here. Guys like Kyren and Moss are already projecting for as much or more points on a weekly basis as the guys like Rachaad White and James Conner. This is why in August we emphasize RB ceiling at all costs. Again, we’ve played two weeks, and if you have a White or a Conner type, you’re looking at their upcoming schedules and probably thinking about sitting them, while broadly you’re praying they hold onto the role they have and keep giving you some usability deep into the season. But for backs that just aren’t special, everything is just downhill from here; the early-season role is the peak, and even then, you might not have ever had it, which was the case with Harris (not to brag, but I obviously feel pretty awesome about the combo of full fading Harris and making Warren my highest-rostered RB across my Underdog portfolio).
I realize in the most competitive leagues, a lot of these RBs are rostered, which raises the price on the ones who aren’t, and means it isn’t easy to ride the RB carousel. But in a lot of leagues, if you missed on Kyren and Justice last week, you might have shots at Moss and Kendre this week. And if you’re in the highly-competitive leagues, you’re looking at who you can stash for next week (Pierre Strong? Back to Jeff Wilson if he got cut, with a look to Week 5?). The are always new opportunities if you’re paying attention.
People will defend having paid a fourth-round pick for Harris or Mattison, or especially a sixth- or eighth-round pick or whatever it was for someone like White or Conner. But the point isn’t that these guys have zero value. It’s that Kyren and Moss and Mostert and on and on are right with them in scoring — or in Najee’s case, just flat outscoring them already — and are also going to project to score right with them in Week 3. I keep emphasizing the future projections because this isn’t fluky spike scoring and the market will expect it to continue in a predictive sense. By comparison, you’re not going to see Josh Reynolds projecting to out-score Ja’Marr Chase in Week 3, just because that’s happened so far (although you might see Puka Nacua up there, which is a little different of a conversation).
Alright, now that I’ve hit on more big-picture thoughts on how to play fantasy than most columns will cover this week, we can finish the introduction and get to the games, plus the recap at the end of Part 2 with the Biggest Signals and Noise of the week, as well as some awesome visuals from Sam Hoppen on key stats I think we need to keep an eye on, and some takeaways on those.
Data is typically courtesy of NFL fastR — also via the awesome Sam Hoppen, who I sincerely hope you’re following and supporting — but I will also pull from RotoViz apps, Pro Football Reference, PFF, Next Gen Stats, Fantasy Life, the Fantasy Points Data Suite, and I get my PROE numbers from the great Michael Leone of Establish The Run. Part 1 of Week 1 included a glossary of important statistics to know for Stealing Signals.
You can always find an audio version of the posts in the Substack app, as well as easier-to-see versions of the visuals at the main site, bengretch.substack.com.
Giants 31, Cardinals 28
Key Stat: Saquon Barkley — 97% snap share, 8 HVT (tied third most in Week 2)
The Cardinals were a big surprise in this one, getting out to a fast start and leading the Giants 20-0 at the half and 28-7 late into the third quarter. But the Giants scored 24 unanswered points in the game’s final 20 minutes to steal the victory, which may have been the preference of Cardinals fans who are dreaming on the potential No. 1 pick next April, and ultimately Caleb Williams. It’s never a bad thing in these spots to see your team play hard under the new coach, and feel good about some elements, while still not winning a game you didn’t really need to win. On the Giants’ side, there are real concerns. I’ve been a fan of Brian Daboll’s, but comeback or no comeback, I don’t know how you get beat 40-0 in Week 1 and then come out this flat against arguably the worst roster in the NFL.
Saquon Barkley (17-63-1, 7-6-29-1) played a huge role in the comeback, and his 97% snap share amounted to him missing only two offensive snaps all game. One of those was the Giants’ final offensive snap, as Barkley was hurt on a second down while setting up the go-ahead field goal, and Matt Breida (1-5) came on for a run on third down, which wound up being the only Giants’ RB touch for anyone other than Barkley. In terms of reading the tea leaves about who might get the work if Barkley misses, both Breida and Gary Brightwell played two offensive snaps, while rookie Eric Gray and even Taiwan Jones (as a fifth RB) were also both active but only played on special teams. That followed what we saw in Week 1, where Breida played 16 snaps and Brightwell 11, as Barkley exited the Giants’ blowout loss early. I’d expect both backs to get work, though it’s a really tough matchup with the 49ers, and the Giants’ offense without Barkley might just be a dumpster fire.
Darren Waller (8-6-76) again led the passing game, while there was again a rotation behind him. Darius Slayton (6-3-62) dominated the air yards with 132, as expected, although Jalin Hyatt (2-2-89) received and caught two deep shots, totaling 76 air yards of his own (for a humorous 38.0 aDOT). Hyatt looked good, but only ran routes on 27% of dropbacks. Parris Campbell (6-4-21) and Isaiah Hodgins (5-4-40-1) were the other main WRs, but at 73% and 71% route shares, weren’t exactly full-timers, either. I’m not sure I’d spend a ton of time trying to figure out this WR room — there wasn’t even anything that exciting in this game, and it was a cushy matchup with the Cardinals where Daniel Jones threw for 321 yards and 2 TDs in comeback mode — but Hyatt and Wan’Dale Robinson still strike me as the best upside bets to potentially consolidate work later in the year.
The Cardinals did enough on offense to deserve me actually trying to write something about them, but I mean, really? The had another heavy negative PROE at -5.3%, operated at a slow pace, and continued to feature Zach Ertz (8-6-56) and James Conner (23-106-1, 1-0-0), so you can’t actually expect me to take them seriously. Conner’s 73% snap share and Ertz’s 75% routes mostly back up the raw volume, but much like I said on the other side, matchup is relevant here. I just don’t expect the Cardinals to move the ball like this most weeks, starting with Dallas and San Francisco as their next two opponents.
Marquise Brown (10-6-54-1) saw a bunch of targets and found the end zone, but his 5.4 yards per target is obviously not great. Michael Wilson (3-3-56) by contrast was efficient, but he lost a ton of routes relative to a full-time Week 1 role, falling all the way to a 56% share. Rondale Moore (1-1-14) also fell to a 63% route rate.
Signal: Matt Breida, Gary Brightwell — only two Giants’ RBs to see offensive snaps other than Barkley, and they have more or less split work both weeks, while rookie Eric Gray has been active but only played on special teams (I expect Brightwell and Breida to both get work this week, but it’s a tough matchup)
Noise: Giants WRs — Jalin Hyatt’s role (27% routes) should be expected to grow, but there are a lot of mouths to feed, and they are all behind Darren Waller, and there wasn’t even that much production in a trail script against a bad Cardinals defense; Cardinals — 379 yards, 28 points (tough matchups with Dallas and San Francisco will likely crater their offensive efficiency)
49ers 30, Rams 23
Key Stat: Puka Nacua — 20 targets, 15 receptions, 38% TPRR, 0.85 wTPRR, 0.81 WOPR
This game ruled, and confirmed that the Kyle Shanahan tree is at the forefront of the offensive bounce back. In my notes, I just wrote “the Key Stat options for this game are endless.” I wound up choosing Puka Nacua (20-15-147, 2-4 rushing) and I think it’s a testament to how good that dude has been that it ultimately had to be him. But you have Kyren Williams (14-52-1, 10-6-48-1) looking like an HVT god with an elite role (his 8 HVTs tied for third most in Week 2). You have Christian McCaffrey (20-116-1, 3-3-19) playing 100% of the snaps, running 100% of the routes, and posting another +42 Rush Yards Over Expectation to lead the NFL with +98 RYOE through two weeks, which is he’s elite as a rusher in this offense, he’s gone from prime Marshall Faulk to more like prime LaDainian Tomlinson now, where the HVTs are still very good, just not otherworldly, and instead of Faulk’s 80 catches and 1,300 rush yards we might be more around LT’s 60 catches and 1,500, which there’s no point to those semantical comparisons other than to say CMC suddenly finding a new way to profile as elite at this stage in his career is amazing. I also had the Rams’ +12.5% PROE — second highest in Week 2 — as a major stat here, as they aired it out and attacked a very strong 49ers’ defense, with success. The Rams are back, baby.