Stealing Signals, Week 1, introduction

Key macro insights into the 2025 meta

I am so fully activated right now, it’s absurd. As I was working through things this morning, my second introduction (Introduction No. 2: The Macro Boogaloo) got a little lengthy. So here it is as a standalone, both to give you something to read now and also so the actual Part 1 isn’t so cluttered before the game-by-game stuff.

---

It’s Week 1, and there’s just so much I have to get off my chest about the NFL, a lot of which builds off my Stealing Signals, Part 0.5 writeup Friday, after the opener.

We know scoring was back up last year, but offensive efficiency wasn’t necessarily way better. Teams scored at a higher clip per drive, and seemed to understand defenses now just want to limit explosives, and built high success rate offenses. But because defenses now do less taking away the opposing team’s best player and that kind of stuff and do more intentional stuff on their side of the ball where they do what they want instead of reacting to the offense, we’ve seen the multiyear trend of offenses being the same way.

The current fantasy meta is one where you can have very low play and drive volume in a game, because the high success rate stuff all keeps the clock moving — runs and short passes with high completion rates and YAC potential — but you also do have the offenses being able to get the ball where they want it to go. For a bit when passing blew up and offenses were amazing around like 2015-2020 No. 1 WR target shares dipped, as offenses were taking what the defense gave them out of increasingly high spread and three-WR sets where they had several viable eligible WRs. That kind of sucked for fantasy football, honestly.

The current meta kind of sucks in different ways, but one way it’s good is we have a pretty strong idea where the ball is intended to go. NFL teams have been adding more and more motion and misdirection as “the analytics” (always in scare quotes) continue to harp on the benefit of those things, and with QB mobility such a key in this modern, success-rate oriented sport — in no small part because of how it offers an escape hatch to any play design where you can still gain 3 yards, or limit the big losses teams try to avoid on sacks, etc. — we get all the read options and RPOs and the different combo and choice routes.

These things have always existed, as people love to discuss, but the NFL is not cyclical in the way it sometimes gets interpreted. Yes, we do wind up with certain frameworks going in and out of popularity, but the sport is constantly evolving. I constantly hear about how the Cover-2 wasn’t just invented recently and Tampa was running it with Tony Dungy like that’s some interesting point, and it’s just not. Just because things work back around doesn’t mean the same objectives are being attempted and accomplished; it’s a massively, massively different sport right now, but the Cover-2 became popular again because of how it allows for post-snap fluidity.

A quick example of NFL defenses being fluid and not sticking to individual rules. The Packers baited Jared Goff into an interception right before half in their win by showing they were dropping into a specific zone that cleared Amon-Ra St. Brown’s curl route, with basically all relevant defenders accounted for, but it was an elaborate bait job where the safety starts barreling down to undercut that specific throw from the moment the ball is snapped, and Goff is just not accounting for a deep safety being able to read and react and get that kind of a break on the ball, because for basically the history of football a safety wasn’t in this play. But in the modern NFL, the defense sometimes just goes, “Hey, we’re going to sell out for this turnover here and hope we’re right” like it’s Tecmo Super Bowl and they got the playcall correct because they know how massive the turnover is.

Look at these images. In the first, the ball has just been snapped, and the safety has gotten the biggest jump on the whole field (except for maybe Micah Parsons). Check out how hard he is already coming downhill.

That’s why this stuff works, where rather than defenders having space or men to cover and responding to route coverages, the play design is so that one (or a series of) defender’s whole approach on a given play is to sprint to a different spot than where he was pre-snap, specifically to mess with what the QB thinks he is seeing from the defense.

Here’s the ball in the air, right before the interception, and you can understand why Goff thought he had that throw based on where the closest defenders all were at this point. But you also see how dead he is. This probably should’ve been a pick six based on how easily the defender beat the throw and undercut the route, but the throw was a bit low and the defender needed to dive.

This is a “post-snap rotation” as it is discussed, but it’s maybe better just called chaos, because what do you do if you’re a QB? Everything you’ve done and do is based on what you see pre-snap, how you account for the 11 defenders on the field, and then you try to figure out where the ball should go based on that. It’s almost like NFL QBs need to start reading the place the ball shouldn’t go based on the pre-snap look, and then just assume that’s what’s going to wind up being open.

Because on any given play, one of the DBs might sprint to an area of the field apart from where his alignment suggests his responsibilities might be, and because this is actually controlled chaos and not just some DB going rogue, there’s rotation behind it and the Packers actually do have everything else covered up. If you go back to those images, look at how 29 winds up in the middle of the field in the second one, after being the other safety at the top of the screen in the first image. They show the 2-high shell, and roll into a single-high look, which means Goff’s throw was to the outside where he had single coverage. But they don’t do this every snap, and you have like 2.5 seconds to figure out if they did it or left the safeties where they were.

This is why all these people who write articles like, “The rate of Cover-2 isn’t even increasing” are silly, which we’ve talked about before. We don’t even have the right stats right now to quantify this stuff. I’m sure some sources chart post-snap rotation, but again it’s a fluidity that I feel like is just very difficult to bucket into a “yes” or “no” column.

This is the defense calling a designed play, like the offense does, and executing it. We see that sometimes in the front seven with blitzes, but this is a coverage blitz. And when you do have Micah Parsons and can speed up Goff’s process while only rushing four, that’s massive. It unlocks the whole thing. Parsons worked Penei Sewell on an inside move and is right there in Goff’s lap in that second image above.

So we see the low passing aDOTs in Week 1, and quarterbacks not trusting their eyes and those things. Scoring was down. But the offensive counterpunch has been that misdirection and disguise and using motion to move the defense around pre-snap and force communication on the back end, and all of that. I did the research on how RPOs changed the No. 1 WR target shares a few years ago, and one of the big offenses we saw that from was Shane Steichen’s Colts, and Michael Pittman had 156 targets that season. And then last year, with Pittman banged up, Josh Downs seemed to be the primary target on the RPOs.

As a refresher, the reason RPOs consolidate target share is the offensive linemen get called for ineligible downfield if the QB throws the ball but not immediately, because of how they are actually run blocking on those plays because of the run option on the RPO (run-pass option), so there’s no room for the QB to get through a progression and the only results are a run play or if the QB does read pass and pull the ball, it’s a pass to the first read and no chance to go through a progression or any of that.

Anyway, I’m talking about the Colts because one of the most interesting Week 1 takeaways is it seems like Tyler Warren is the RPO guy this year. Pittman still earned a bunch of looks, but on the pull-and-quick-throw plays, we got a ton of Warren, and that’s extremely bullish for his outlook.

But it’s not just RPOs consolidating target share, because one of the byproducts of defenses playing this way is how offenses get into different formations, and do different things, but they do seem to get the ball to their specific playmakers better in this era than that 2015-2020 period I was talking about. And that’s the part of the current meta that’s fun for fantasy. Some of these teams only have like three guys who touch the ball. And that’s awesome for our purposes!

I don’t have enough of a grasp of the specifics to be able to discuss the “why” behind this, but I talked back in the first full year (2022) teams really started playing this way through the full season (the passing aDOT shift downward happened mid-2021), how it sort of left offenses to their own devices, and in 2022 we got multiple teams running the ball at historic levels (Arthur Smith’s Falcons, Justin Fields’ Bears) and multiple teams passing at historic levels (Tom Brady’s Bucs, Kellen Moore’s Chargers, and that passing was a lot of underneath stuff as their related-but-different answer to the heavy running in Atlanta and Chicago).

Things have leveled out since then as offenses just understood what defenses were doing better and naturally tended toward more balance, but there’s an element here where I think on the subset of plays like the one I highlighted above where the defenses aren’t as reactionary and are instead proactively doing what they want, that the offenses aren’t doing as much of the schematic chess match stuff (where it’s all the stuff you’ve heard ex-coaches and ex-players talk about where it’s like, “When I see that look, I have my man beater,” or how they have specific answers for specific defensive formational stuff that is common), but rather when the uncommon stuff happens, the offense is just trying to read that as quickly as possible and then it’s like, “We have numbers, go to the best player.” Or something like that, because it winds up that the No. 1 options have seen their target shares rebound, and splits data shows it’s overwhelming that when the defense does rotate into single-high that the No. 1s are crushing, which is in part because they are often outside receivers and you don’t have the same safety presence over the top, but also I think is a bit of an indication of the QB just reading these safety rotations in a way that is the answer in itself. Those things are maybe the same thing. This is all very involved and I’m working through it, not the person who has all the answers.

But the fascinating thing I’m trying to get to is I saw multiple teams run the same plays repeatedly this week. DeMario Douglas caught a TD on the second play in a row where he was running along the back of the end zone toward the back left pylon, and he was targeted on both plays. Drake London got back-to-back targets in the end zone on the same inside whip route in a key spot down the stretch. Shawn Syed had a great, quick explainer about how the Bucs hit Egbuka for the game-winner on a concept they’d run twice before, and both of the two previous ones Shawn highlights are also in the fourth quarter, so again you see that rapid iteration of like, “We know the defense is not adjusting to what we’re doing, because they are proactively doing what they want instead of reacting to us, and we had this play but we just didn’t necessarily execute, so let’s just call the exact same thing.” As with everything, the sport evolves toward where Madden gamers were 20 years ago, where they were spamming the same plays until you stopped them, and then just switching gears and finding another couple plays to call until you stopped those ones.

It’s not that simple in real life, obviously. But man, this sport is so fascinating to watch evolve. And then the fantasy football takeaways are going to be very relevant to all this stuff. I’m so bullish on Tyler Warren because of everything I just wrote, not because some database tells me he got a certain number of first-read targets. I feel really good about that for Warren, but I may not in another situation. Every situation is pretty unique, and you have to understand the underlying offenses to analyze the players for fantasy football.