Super Bowl game pick
There's one more electric game to go
This weekend, both No. 1 seeds face off in a game that has been written about and analyzed a million ways. I genuinely think when you have two good teams playing this well, there’s not much that can be written about that is super useful. It’s the final game of the year, so it feels like a fun opportunity to instead talk through some of the challenges of parsing any football game.
These teams’ strengths and weaknesses are known, and more importantly they know each others’ strengths and weaknesses and have spent two weeks gameplanning. That goes both ways — it’s not just trying to exploit where you think the opponent is weak, it’s also self-scouting and covering up your own deficiencies. And what they come out with on the other end should almost be expected to be different than their past tendencies in some way, making prediction difficult. Of all the stuff I’ve seen out of Arizona this week, this quick interaction with Lane Johnson might have been my favorite:
He doesn’t really answer the question about motion, instead opting for a more generic discussion. But it’s a great question nonetheless, and it seems to unlock in Johnson a bit more honesty. Because he’s asked something so thoughtful, he seems to recognize he should meet that with a thoughtful answer, and he has some great comments about an adjacent topic.
In that response, Johnson notes, “Over the course of a season we try to change. We monitor what we do on offense. We’re trying to see if we’re being too predictable in what we’re doing. But that’s what we have analytics for. I feel like that branch of the NFL is definitely taking off over the past few years, helping teams gameplan.”
Johnson is in his 10th season with the Eagles, and has never played anywhere else, so while he feels like this is taking off in the NFL, it’s probably safe to assume he’s mostly talking about the degree to which the Eagles have utilized this stuff internally. Ultimately, he doesn’t have much knowledge about how much other teams are using analytics to identify their own tendencies. He might get a feel for it when teams they are playing are doing things differently than what they’d previously shown on tape, and if that’s happening more frequently in today’s NFL than before, but for the most part I’d guess he’s referring to his own experience in meeting rooms and in terms of how the week unfolds for him, and how that’s evolved across a decade.
And the Eagles doing this is one of the few things that’s been clear about them. I’ve written before about what I’d call “predictable inconsistency” with Philadelphia, noting a trend I first heard about from Pat Kerrane where they were — as might be expected — a more vertical passing offense against defenses that didn’t rush the passer well, and then how they could transform to a low-aDOT, chain-moving offense against the teams that did get home effectively.
What’s funny is, while I’m not trying to give Pat credit than take it away, there’s something to be said about whether that was the ultimate reason for the shifts of offensive approach in those games. I think so — it could have been the opponents’ pressure rate which triggered them intentionally approaching games differently than what they’d recently put on tape — or I could be chalking something of a coincidence up to the wrong variable. Maybe a little bit of both. Honestly, that’s the thing we never really know with football analysis. Often the second we think we’ve rooted out a trend, an adjacent variable impacts it. (In this case, maybe that could have been significant offensive line injuries for the Eagles that changed the dynamic, something that didn’t happen to their team in this season, but I’m just posing as a hypothetical.)
This is what we talk about when we say as analysts we’re always dealing with an extremely small sample. With the year-over-year turnover in the league, past season data is often difficult to rely on (though it’s not worthless in all cases). But typically, you’re looking at 20ish games at most, and that comes at this point in the season.
But even that is probably overstating the sample. Given how much rosters change through a season, and philosophies grow and evolve from September to February — something that’s a near requirement to be a successful NFL team good enough to get to the Super Bowl — some of the early-season data points are probably almost irrelevant. These two teams actually offer us a solid degree of consistency in what they are, and yet I’m still not sure the exact way I’d model anything in this matchup, because in a season with so few data points, I’m not sure we ever know the most representative sample. It may be different for a team’s offense than their defense; it may not be chronological; they may opt for such a new and unique approach in a given game that their past data couldn’t possibly have been representative, in hindsight.
Even when we look at the data and get an idea — for example — that the Eagles have an advantage running between the tackles, based on different samples and different individual player-level stuff, there’s an element where the predictiveness is dependent on the teams’ approaches. We don’t really know how the Chiefs will play it. Maybe they’ll sell out to stop the inside run. That would force Jalen Hurts to the edge and also probably open up the downfield passing game, which seems like absolute death against the Eagles, which makes this a reasonably straightforward element of the game. I’m pretty confident we’ll see the Eagles able to run efficiently inside, should they desire to, because the Chiefs can’t possibly think their best gameplan to stopping this dynamic offense is to load the box. But I’m still underlining here that we’re making a prediction that feeds on past data and also assumes something unknowable. And even when we can feel confident about that particular matchup being notable on the gameplan level, there are still other questions — do the Eagles even want to run? Is that a priority they’ll stick with? Will the scoreline allow it? How much does effective inside rushing impact a game, anyway?
Football is often talked about as a chess match, and elements of it are broken down in terms like “copycat league” when schematic stuff infects different teams across the league. Sometimes you get a truly novel gameplan against a team in the Super Bowl, like when the Patriots wrote the blueprint for how to stop Sean McVay’s first iteration of the Rams, something that forced the Rams into a bit of a lull that carried over to the next year. A line can very likely be drawn from that Patriots’ gameplan to the Matthew Stafford trade. And again, analyzing the Rams and Patriots going into that game would not have really helped what we saw, unless maybe that analysis focused on Bill Belichick’s multiyear trend of defensive gameplans. It feels unlikely a lot of people were saying, “That is what will decide this game.” (As an aside, I’ve made an assumption here that it even is what decided that game, something other analysts might reject.)
We so often wind up looking back on these games with an appreciation for greatness, whether that be the players or even the coaches. How the Giants were able to stop the undefeated Patriots in early 2008 has been a feather in Chiefs’ defensive coordinator Steve Spagnuolo’s cap for years, even if it was relatively simple — get pressure rushing four — and even if Justin Tuck and Michael Strahan and the gang on the defensive line were the biggest reason that game was winnable for the Giants. Of course, they also needed some of the variance that is present in every football game. They could have had a great defensive gameplan and still lost, if not for the helmet catch, for example. Good luck predicting that play.
I’m not trying to say there’s nothing to be learned from breaking this game down, and there have been some incredibly insightful looks at the specific matchups that will be key. Bill Barnwell does awesome work and had a great breakdown at ESPN. Friend of the newsletter Connor Allen always does a great job with his Super Bowl deep dive over at 4for4. Understanding what these teams are and aren’t, and what are most likely to be the deciding factors of the game, is massive.
But what I don’t want to lose sight of is the uncertainty. The bottom line is this game can go several different ways. Every NFL game can, but this one more so than most, because you have two teams that have been very successful in part because they have been willing to change, and now they have extra time, and I frankly think our base expectation should be that we see something unique here. There will be some different plays in the gameplan on both sides. There might be real unique stuff schematically. We might see things play out in a way that favors an unheralded name. The Chiefs’ receiving corps is probably the best place to look toward to make that point, given the uncertainty with that whole group and how we saw Marquez Valdes-Scantling step into a more well-rounded role and the most productive game he’s had all season in the AFC Championship. If the Chiefs win, you’re probably talking about someone unheralded from that group, like a Kadarius Toney or Justin Watson, being a huge talking point.
This can be true among the more known elements of these teams. There’s talk about A.J. Brown and DeVonta Smith and who the real No. 1 is, but the chess game in the NFL leading the Eagles to win in part due to a huge game from Dallas Goedert or Miles Sanders wouldn’t be some major surprise. This type of thinking is one of the reasons I love Dalton’s approach on the props side of things where he’s constantly looking for mispriced opportunities among very low probability outcomes. It’s an understanding and appreciation for uncertainty, and then trying to price that out. And his results speak for themselves that he has an incredible knack for that.
As for me, it’s at least been a profitable season on sides and totals, with us at exactly 100 official bets heading into this game. The record stands at 53-44-3 for +5.29 units, and that’s despite me really mismanaging the number of my over (11-15-1 for the season) and under (12-8) bets, given I knew going in my bias to the over was unprofitable, something that’s obviously common across sports betting.
I missed two more overs last week, sure the Championship games would feature good offenses able to dictate. At least we got the sides correct. I lean over in the Super Bowl once again, but with the total up over 50, I’m staying away. The Eagles’ ability to piece together 8+ minute drives can really limit the number of possessions, so even though I believe both offenses will likely be productive when they have the ball, I think the high total is a pretty efficient number.
I think the spread is pretty efficient as well. The Eagles have played a weaker schedule, but have a lot of strengths, and the Chiefs have Patrick Mahomes. Again, we can break this down 1,000 ways but there’s an elegance to recognizing the market for the Super Bowl is efficient, and at a certain point we’re just betting on something. So here’s how I’m playing it.