Dynasty content is part of the ELITE package for Footballguys Premium Subscriptions. We're making this preview available so you can see the edge these subscribers are getting. Sign up here.
There's a lot of really strong dynasty analysis out there, especially when compared to five or ten years ago. But most of it is so dang practical-- Player X is undervalued, Player Y's workload is troubling, the market at this position is irrational, take this specific action to win your league. Dynasty, in Theory is meant as a corrective, offering insights and takeaways into the strategic and structural nature of the game that might not lead to an immediate benefit but which should help us become better players over time.
Most Comparisons Are Wrong
Fantasy football has exploded in popularity over the last 20 years, which is good for me (a person who writes about fantasy football) and for you (presumably a person who likes reading about fantasy football). With that rise in popularity, there's more content than ever before. Some of it is weird and unlike anything else (it probably comes as no surprise that I'm a big fan of the weird stuff). But a lot of it takes on a predictable pattern. One major genre of analysis is what I like to call "Finding This Year's X".
Here's how it works: last year, several players dramatically outperformed expectations. If you rostered one or more of those players, you had a much higher chance of winning your league. This year, players will (probably) once again dramatically outperform expectations, and since we (presumably) would like to have a much higher chance of winning our league, these pieces compare current players to last year's league winners to try to find this year's league winners.
Winning your league is a fine goal, but the problem is that this genre of analysis will rarely help you accomplish it. That's because most of these player comparisons are wrong. I don't mean that in a general sense-- as in "all player comparisons are wrong" (but some are useful). For example: if you compared Justin Jefferson or JaMarr Chase to all receivers who had massive rookie years, your pool of comparable players would include Randy Moss. If Randy Moss had ruptured his Achilles tendon in Year 2, that would make the list of comparable players to Jefferson or Chase look worse. But Randy Moss rupturing his Achilles (or not) in 1999 doesn't actually make it any more likely that Ja'Marr Chase would rupture his Achilles (or not) in 2023.
Any time you compare a player to a list of different players, that comparison is, by definition, wrong (since those other players are not actually the player in question). But it can be useful if, for instance, players who have monster rookie years tend to have broadly similar career shapes (they do).
Anyway, that's not the piece I'm writing today-- mostly because I already wrote it six years ago. Instead, today I want to argue that these player comparisons usually make a fundamental error that ensures that not only are they wrong, they're unlikely to even be useful.
How Most Player Comparisons Work
The typical first step in a player comparison is finding a past league winner you want to compare. For our purposes, let's say 2021 Cooper Kupp. In 2021, Kupp was a middling receiver who had a massive, out-of-nowhere breakout at age 28 to finish with the second-most receptions and receiving yards in history (along with 16 receiving touchdowns). Obviously, if we can find the next middling receiver who will rewrite the record books, that is a huge advantage. So comparisons will look for relevant attributes that would have predicted Kupp's breakout and then search for those attributes in other receivers.
For instance: Kupp was written off in part because he was primarily a slot receiver and slot receivers are typically assumed to have lower upside. So maybe in searching for this year's Cooper Kupp we pay special attention to slot receivers. Kupp was playing with a new quarterback in 2021, the well-regarded Matthew Stafford, so maybe let's look at receivers with a new passer. Despite the low regard in which he was held, Kupp had a Top 10 season two years prior, finishing with 94/1161/10 receiving.
So we're looking for a 28-year-old slot receiver playing with a new quarterback for the first time in his career, coming off of a disappointing season after a Top 10 age-26 campaign in 2021. Therefore, this year's Cooper Kupp is... Hunter Renfrow (whose 103/1038/9 performance in 2020 looks eerily similar to Kupp's 2019).
Now, having said all that, I want to make it clear: despite the superficial similarities, Hunter Renfrow is not, in fact, a secret league winner for the 2023 season. (Well, he's probably not; he's about as likely as any other player being drafted in a similar range, I suppose).
Why does that process not work? I want to focus on three key problems.
Problem #1: Most Years Don't Have a "This Year's X"
Who was last year's Cooper Kupp? Tyreek Hill, maybe, or A.J. Brown? But both players had a stronger history of production, were being drafted in the first three rounds, and were considered the top receivers on their own team (none of which was true for Kupp in 2021). Christian Kirk? He had the most similarities as a player, perhaps, and his 84/1108/8 was certainly a nice surprise, but he wasn't exactly a league winner (he scored just 56% as much as Kupp had the year prior).
For that matter, who was 2020's Cooper Kupp? if Cooper Kupps are so common, then why were we surprised by him in the first place?
Problem #2: It Gives No Sense of Base Rates
So Kupp was a middling slot receiver with a new quarterback who hit it big; does this mean we should be drafting middling slot receivers with new quarterbacks? It doesn't, because we're missing a lot of crucial information; how many other middling slot receivers with new quarterbacks have failed to hit it big? Maybe those really are the features that led to Kupp being devalued... but maybe that devaluation was fair. If there have been 100 receivers who fit that general mold, and Kupp was the only one of the bunch to make it big, you'd probably have been a lot better off avoiding the archetype than seeking it out.
Problem #3: There's No Half Credit
You may think you identified the attributes that led to Kupp being undervalued. But how certain are you? Because I could tell a completely different story about Kupp than the one I've been telling so far.
Kupp entered the league without much in the way of expectations but was one of the best and most efficient rookies since 2006 (per my linked rookie production model-- another point in favor of the Hunter Renfrow comparison, actually), then followed a superstar trajectory in years two (injured but on pace for a Top 10 finish) and three (finishing fourth at the position). Only Michael Thomas, Davante Adams, DeAndre Hopkins, Julio Jones, Tyreek Hill, and Mike Evans averaged more points per game than Kupp from 2018 to 2019.
Maybe the real story of Cooper Kupp is one of an ascendant star who had a weird down year but then returned to his previous trajectory. And here's the problem: if that's the "real" story, then focusing on Kupp's other attributes does us no good whatsoever. They're incidental to the breakout, mere coincidence, we shouldn't expect them to help us in the slightest in our quest for This Year's Cooper Kupp.
There's a Better Way to Make Player Comparisons
Friends, hobbyists, Footballguys, lend me your ears; I come to praise player comparisons, not to bury them. I think a well-crafted historical comparison is one of the best and most useful tools in our toolkit. So let me close with what is (in my opinion) the correct way to make comparisons, as well as an examination of why this doesn't fail where the other method does.
It all boils down to a simple shift in mindset: instead of looking for "This Year's X", look for "Last Year's Y".
Here's an example: instead of looking for the next Cooper Kupp, let's look at the current Cooper Kupp. He's a 30-year-old former All-Pro coming off of a very efficient but injury-shortened season. How have All-Pros historically done in their Age 30 season? How have 30-year-olds historically bounced back from injury?
I brought up Kupp because I was recently in a discussion about whether Cooper Kupp or Jaxon Smith-Njigba was more likely to be a league-winning receiver in three seasons (2025). Here's a thread on it (though I'll recap the highlights here in case you don't want to click through):
Historically, about half of WR prospects the caliber of JSN have been complete duds for fantasy.
— Adam Harstad (@AdamHarstad) May 16, 2023
Among the Top 30 fantasy WRs at age 29 in my dataset (1985-2019), 33% were Top 12 caliber again at age 32. (Harrison, Owens, Wayne, Carter, Rice, TBrown, SSmith, AJ, Horn, Drew Hill) https://t.co/SQof5kMVit
Looking at the Top 30 fantasy receivers at age 29 from 1985-2019, 10 were Top 12 fantasy receivers again at age 32 (Marvin Harrison, Terrell Owens, Reggie Wayne, Cris Carter, Jerry Rice, Tim Brown, Steve Smith, Andre Johnson, Joe Horn, and Drew Hill), a 33% hit rate. Meanwhile, here are the 19 receivers drafted in the top half of rookie drafts from 2014-2020: Mike Evans, Sammy Watkins, Jordan Matthews, Brandin Cooks, Amari Cooper, Kevin White, DeVante Parker, Dorial Green-Beckham, Laquon Treadwell, Josh Doctson, Corey Coleman, Sterling Shepard, Michael Thomas, Corey Davis, Mike Williams, N’Keal Harry, DK Metcalf, A.J. Brown, and CeeDee Lamb. Of those 19 players, only five were Top 12 in Year 3 (Evans, Cooks, Thomas, Metcalf, and Lamb), a 26% success rate.
This doesn't mean you should draft Kupp ahead of Smith-Njigba in dynasty leagues (though it doesn't mean you shouldn't, either). The rookies who didn't hit in Year 3 still retained plenty of value, while the vets that didn't hit in Year 3 became virtually untradeable. But by coming up with a list of players who are similar to the player in question, we can get a better sense of how likely certain outcomes are.
Why is this approach better?
Advantage #1: Most Years Don't Have a "This Year's X"
The weakness of the other method is the strength of this one; if most years don't have a good comparison, looking at thirty years of history at a time dramatically increases the chances we'll find a usable sample.
Advantage #2: It Tells You Everything About Base Rates
If most players like Cooper Kupp didn't break out, that will show up in the list of comparable players to Cooper Kupp and we can consider the downside of the acquisition as well as the upside.
Advantage #3: You Get Half Credit
Maybe the single most relevant attribute for Cooper Kupp is not that he's a 30-year-old former All-Pro. Maybe it's that he's a slot receiver or that he was injured last year. The thing is... as long as the attribute we use to generate comparable players is even a little bit relevant, the list of comparable players will still improve our understanding of the range of outcomes along that dimension. You don't need to get everything exactly right. You can just pick a few attributes that you're relatively sure matter, and the resulting list will still likely be useful.
So if this is the superior way to make player comparisons, why doesn't everyone do it this way? Because unfortunately, this method is a lot more work-- you need to generate comparisons for every single player in the league (or at least every player you're considering) instead of quickly surfacing a couple of names to focus on. Also, this way is a lot less sexy; it doesn't promise to quickly unearth next year's league winners.
But it has the advantage of being useful, which is an edge in a hobby where everyone else is making comparisons wrong.