I wrote yesterday a little bit about how I think the 2022 season will be viewed in a variety of ways, and some of those might not be particularly useful. You guys know I’m not super into best ball, but I’ve been seeing some really interesting commentary from that community. The first note comments on how random the Underdog playoff structure is.
Jakob has a lot of other great thoughts on this, as well as on tons of topics, over at his Twitter. But his point is a great one. What Akers has done over the past couple weeks has been more impactful, because to advance in Underdog you have to have ceiling weeks every week this time of year. As Jakob points out elsewhere, everyone knew the deal going in, so the task was building a strategy around that. That said, my opinion of the structure is that it is very bad, and all it took was a high-variance season to really shine light on that.
That’s not a thing I’m going to go too deep into today, but it does transition into the discussions I’m seeing about different draft strategies. And when we talk structural drafting, Zero RB is very often first to be discussed. As it turned out, hyperfragile teams with early RBs seem to have advanced to various Underdog finals at high rates. Some have taken to the same old criticisms of Zero RB as a result, despite the fact that the regular season winner of a cool $1 million in Underdog’s biggest tournament was a Zero RB team. There are three total $1 million winners this year in BBM, and the only one that has been decided was a Zero RB team, but this season was a failure for Zero RB, or so some tweets I’ve read this morning argue.
That’s another thing I’m not going to go too deep into today, other than to say that it definitely tilts me to no end, which I know you guys can pick up on. I don’t really care about being wrong if I’m actually proven wrong, but I have this obsession with facts, truth, what have you — it stems from being a middle child, and I was always a stickler for rules in board games and that kind of thing — and I mean if you want to get under my skin, all you have to do is make some claim that isn’t actually supported by anything. Most humans will just let that slide; I will straight up argue with you for hours.
I am not proud of this. This is not an enviable trait I possess. But it does lead me to write a lot of the stuff I do here. What I do want to emphasize today is that I’m already preparing to reject the types of conclusions I expect from the best ball community. Advance rate data has become very popular, and shows what types of structures were successful or unsuccessful. But as Jakob noted in the tweet at top, so much of what advances is entirely dependent on a few weeks. Part of the Cam Akers/Josh Jacobs comparison is that Jacobs hasn’t done a ton the past couple weeks. For drafters who had him, that came at the worst possible time; for advance rate data, the sequencing of Jacobs’ production immediately shifts the types of rosters that move on, because he was clearly the biggest late-round RB hit.
Stepping back a minute, what we know about early-round RBs is they are basically the only RBs we feel good about having roles. I often comment about how there is no such thing as guaranteed touches for RBs, but the reality is the first 15 or so RBs in ADP every season actually more or less do have guarantees. And that’s exactly why they go there. In best ball, it’s a pretty big asset to just have guys who will continue to get work at a position where there is so much turnover. Even when they don’t produce. Joe Mixon has been terrible in the green zone all year, and not much better anywhere else on the field, and his job has never been in jeopardy. That’s exactly why people were content taking him as high as they did, and I don’t dispute anyone who argues that touch floors are more or less worth paying up for, particularly in best ball, and what all of that means.
Hyperfragile rosters are obviously viable. Basically anyone who appreciates the potential of Zero RB rosters will also acknowledge the virtues of other roster builds. I’m constantly trying to communicate that Zero RB is not a draft strategy you plan for in advance and make sure to follow through on.
Those are the facts about two key draft structures. And within that context, and within the context of the past two weeks of NFL production, it’s not at all surprising that advance rates would tilt toward earlier RB rosters. There weren’t a lot of big weeks from later RBs, there were some big weeks from later WRs like Zay Jones in Week 15 and mid-round WRs like DeVonta Smith in Week 16, and we’ve gotten big weeks from guys like Dawson Knox and T.J. Hockenson and even George Kittle who didn’t go super early that centered the all-important TE scoring on mid-round options as opposed to a scenario where Travis Kelce was the TE1 in these weeks, or Stefon Diggs and Davante Adams didn’t go quiet.
What I’m saying is if you understand the various possible successful roster builds, and you understand the tournament structures that are being discussed, it’s not some huge piece of evidence to comment on what types of rosters advanced. To advance, you have to finish first, not second or any other position, in your pod. You have to have ceiling weeks. The advancing teams are going to be concentrated around the teams that have players who posted ceiling scores. You can’t win a playoff pod made up entirely of good rosters through a team that didn’t have any of the 30-point players that week. So what I’m saying is there’s not even anything all that interesting about advance rate data that we shouldn’t already be able to figure out just by looking at where the points came in. It becomes a sort of confirmation bias, but really it’s just telling the story of those particular weeks and what it meant for 2022, which was one possible outcome of a number of ways future seasons could play out.
That’s the analysis that helps you make actionable decisions going forward. Because you still need to adjust to wherever ADP will go, and the ways the player pool will be impacted by offseason moves and uncertainty, and all those other factors. The whole Zero RB discussion that has raged on for a decade now is between one side (Zero RBers) who are saying “This is a viable strategy among a host of strategies” and another that seeks to argue it is a strategy for zealots and should be laughed at. One of those sides is correctly aligned with the broader lessons that there are a ton of ways things can go in fantasy football, while the other seeks to exclude and set rigid rules around what has to happen (you can’t win by drafted no RBs through Round X). The openness/closedness of the two sides already tells you who is correct, because we are dealing with subject matter that humbles everyone who makes predictions about it.
Anyway, I just saw some commentary on best ball strategies, and started to get that feeling regarding what I wrote yesterday. The takes about 2022 are coming. And many of them are going to be bad. People will tell you the thing that is happening is exactly how it is and can’t be any other way, and they will relish in being right if it continues down that path. And then when situations arise like Tony Pollard out-snapping Ezekiel Elliott for three straight weeks while both are healthy, they’ll change the discussion and pretend like we didn’t have to listen to them say that was literally impossible and we were idiots for believing otherwise for like two years.
Let’s finish off the final Stealing Signals of the year. I didn’t have PROE data yesterday, but I have to note that the Saints and Browns were -30.5% and -23.4% in that extreme weather game, the Titans were -24.4% with Malik Willis, and both Baltimore and Atlanta were just beyond -15% in their windy matchup. Meanwhile, Cincinnati led the league again at +15.1%, and both Minnesota (+13.3%) and the Giants (+7.8%) were pass-heavy in their matchup.
Data for Stealing Signals is typically courtesy of NFL fastR via the awesome Sam Hoppen, but I also pull from RotoViz apps, Pro Football Reference, PFF, RotoGrinders, Add More Funds, 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.
49ers 37, Commanders 20
RB Snap Notes: Christian McCaffrey: 75% (-14 vs. W15 high), Tyrion Davis-Price: 25% (return), Brian Robinson: 44% (+6 vs. W15), Antonio Gibson: 29% (-31 vs. W15), Jonathan Williams: 27% (+19 vs. high)
Key Stat: Commanders — (-17.5%) PROE (fourth lowest in Week 16)
San Francisco got out to a lead, Taylor Heinicke finally got benched, and the Commanders went extremely run-heavy despite their deficit en route to speeding up their loss. As a result, San Francisco only ran 51 plays in their win.
Christian McCaffrey (15-46-1, 2-2-12) had an inefficient day, but luckily squeaked out a late touchdown. He had just 4 HVTs, tied for his low since his first game with San Francisco when he played sparingly, but Brock Purdy also only threw 22 times. Tyrion Davis-Price (9-30) took over the Jordan Mason role and got a decent amount of work, but four of those carries came on SF’s final drive of the game, and CMC got both green zone touches, his fifth straight game with multiple touches inside the 10-yard line, including one to cap that final drive. Over his first seven games of the season, CMC had just three total green zone touches.
George Kittle (8-6-120-2) got there despite the low-volume game, hitting for two more long touchdowns. Brandon Aiyuk (7-5-81) also had a pretty strong game all things considered, as that duo combined for 15 targets and 11 catches in a 15-completion, 22-attempt game. Part of the low volume was also do to Ray-Ray McCloud hitting for a 71-yard TD run in the second quarter that helped no one in the fantasy world and more or less both canceled out a potential full drive of plays for San Francisco while also pushing them towards positive script. Kittle’s TDs could be viewed similarly, as SF’s volume was victimized by their own efficiency in this one, to a degree.
Brian Robinson (22-58, 1-0-0) got a big workload but did little with it, while Antonio Gibson (5-10, 3-2-21) was held back a bit, likely due to being on the injury report all week and not being 100%. But rather than using Gibson’s reduced role to get Robinson involved in the passing game, the team used Jonathan Williams (3-13, 3-2-28) as a pseudo-Gibson to fill his other reps. Gibson ran routes on 35% of dropbacks, down from 70% last week, while Williams was at 30% and Robinson just 16%. Robinson did get four green zone touches, but he was unable to score.
All three of the WRs found the end zone in this one, with Jahan Dotson (9-6-76-1) continuing his impressive late-season run, Terry McLaurin (5-4-77-1) getting there on lower-than-usual volume, and Curtis Samuel (5-5-52-1) enjoying Carson Wentz taking back over, as he caught a late touchdown from the QB who helped him to a fast start earlier this year. Logan Thomas (8-6-35) also set a season high with 8 targets as the passing game did have some production despite a massively negative PROE, largely due to Wentz throwing 16 fourth-quarter passes.
Signal: Christian McCaffrey — multiple green zone touches in five straight games (had three total in first seven games this year while with Carolina); Brian Robinson — despite Antonio Gibson being limited, did not see any increased passing game usage, as Jonathan Williams stepped into 30% routes and Robinson sat at 16%
Noise: 49ers — 22 passes (Washington went heavy negative PROE in trail script, SF scored on splash plays to limit own play volume)
Cowboys 40, Eagles 34
RB Snap Notes: Tony Pollard: 53% (-2 vs. W15), Ezekiel Elliott: 47% (-4 vs. W15), Miles Sanders: 54% (-5 vs. avg), Kenneth Gainwell: 42% (+2 vs. W15 high)
TE Snap Notes: Dallas Goedert: 94% (return)
Key Stat: Tony Pollard, Kenneth Gainwell — 7 HVTs (tied for fourth most in Week 16)
Gardner Minshew came to play, Dak Prescott rallied from a bad pick-six early, and we got a pretty epic shootout in a week where scoring was down due to weather almost everywhere else.
CeeDee Lamb (11-10-120-2) had a monster game, dominating the passing volume for Dallas. Tony Pollard (9-19, 8-6-61) was easily the second top passing-game option in a game Dallas clearly wanted to win, and they showed something we’ve talked about with them dating back to the early part of the season, where they focused on getting the ball to the players they wanted the ball to go to. It’s obviously not that simple — defenses are trying to take away your top weapons — but Miami is probably the best example of it this season, and Dallas does a good job of it as well. In contrast, some teams seem to forget the talent they have. All of this is a little difficult to parse relative to the idea that volume in the passing game is earned by that talent, so it’s kind of a nuanced point about situation that I doubt I can properly explain in Week 16, but I guess I’d argue it’s more of a point on the weekly level, while the target-earning stuff is better understood over longer time periods. Anyway, when we get this kind of deliberate concentration to the top options, it’s obviously awesome for fantasy.
Pollard didn’t have a massive day, but he did total 7 HVTs, including a green zone carry from the 9-yard line early, which he took down to the 1. Unfortunately for Pollard managers, Ezekiel Elliott (16-55-1, 1-1-6) predictably came in to punch that in. Everything I just wrote is in some ways contradicted by Elliott’s usage, but most of his runs feels like means to an end, and at least Dallas is at the point of using Pollard on more snaps overall and for more high-value touches. Elliott would get both of the other two green zone touches for three total, and remains the preferred green zone guy, but not in a way that completely prevents Pollard from touchdown equity in close, and Pollard also adds big-play TD potential. Pollard ran routes on 57% of dropbacks to just 36% for Elliott as the other big note on their split.
Dalton Schultz (4-3-43) and Michael Gallup (7-4-36-1) were involved as well, as expected, while T.Y. Hilton (1-1-52) hit for a big play in his 2022 debut, converting a third-and-30 on one of his 12 snaps. Hilton adds a nice dynamic to this offense.
With Minshew under center, the Eagles totaled 10 HVTs, tying their previous season high. Unfortunately for Miles Sanders (21-65, 1-1-6) managers, he didn’t score on either of his green zone touches, and Kenneth Gainwell (4-17, 4-4-41) dominated the receiving work, while also getting three green zone touches of his own (that he also did not score on). Gainwell ran routes on a season-high 56% of dropbacks.
Speaking of concentrated offenses, the Eagles are a great example in the passing game. DeVonta Smith (12-8-113-2) had the bigger day than A.J. Brown (8-6-103), as Trevon Diggs shadowed Brown for most of the game and the Eagles picked on other matchups. Smith had a shot at a third touchdown late that could have put the Eagles in front, but it went out of the back of the end zone. Dallas Goedert (3-3-67) returned to a big snaps and routes role but with limited production.
Signal: Kenneth Gainwell — season-high 56% routes; Dallas Goedert — 94% snaps, 90% snaps (definitely healthy in return)
Noise: Eagles, Cowboys — 861 combined yards, 74 points