There's a lot of 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, and 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.

Breaking the Code

Here at Dynasty, in Theory, I have a code: nothing practical, nothing actionable. We have a lot of really strong dynasty articles on Footballguys dedicated to giving advice for managing your teams. My focus is more on the weird, interesting, or conceptual. Some might accuse me of naval-gazing. (I certainly wouldn't argue the point.)

But to quote a fictional pirate, the code is more what you'd call guidelines than actual rules. Thinking is for doing, as the social psychologists say. The most elegant theory in the world is useless if it doesn't match reality. To that end, there is one thing I do every year that is-- and it pains me greatly to admit this-- incredibly practical.

I have a model for evaluating rookie receivers, and every offseason, I publish the results. No, not this year's rookies-- there will be gallons of ink devoted to that cause already, there's no way I can add any value on top. My model evaluates last year's rookies.

While there is a ton of effort devoted to valuing prospects before they reach the NFL, there is much less dedicated to revising those evaluations once they're here, so it's much easier to find a comparative advantage. Which is good because I'm incredibly lazy and want to get the maximum return on the very least amount of effort possible.

This model was mostly built on dumb luck. A decade ago, I spotted what I thought was a fairly glaring market inefficiency. I watched it for years, and it persisted. Eventually, I realized there was an extraordinary edge to be had, and building the model was the path of least resistance compared to muddling along without it.

It turns out (for wide receivers, at least) that, despite conventional wisdom that players need a couple of years before we can be sure of who they are, rookie seasons are shockingly predictive of the overall course of their careers. Most importantly, they're predictive regardless of draft position. A model that tells you that a Top 10 draft pick with a monster rookie year like Ja'Marr Chase might go on to have a pretty good career isn't especially helpful. Not because it's wrong, but because you didn't need a fancy model to tell you that.

But my model has a great track record at identifying off-the-radar players before their value reaches its peak. From 2006 to 2022, the Top 10 scores for players who were drafted outside the first two rounds belonged to Marques Colston, Terry McLaurin, Keenan Allen, Mike Williams (Tampa Bay version), Stefon Diggs, Cooper Kupp, Doug Baldwin, Tyreek Hill, Hunter Renfrow, Denarius Moore, T.Y. Hilton, and Amon-Ra St. Brown.

That's not a 100% hit rate. It would be foolish to expect perfection; if a model does give a 100% hit rate, you can be confident that it's overfit. But I have startup dynasty ADP since 2014, and almost all of those guys were extraordinarily cheap to acquire after their rookie seasons. In that span, McLaurin was the 27th WR off the board, Allen was 9th, Diggs was 34th, Kupp was 30th, Hill was 34th, Renfrow was 65th, and St. Brown was 22nd. If you acquired all of those players at prevailing market rates, you probably built yourself a dynasty.

Crucially, I have found that once you have a player's score, knowing their draft position adds very little predictive power, meaning rookie performance is almost entirely new information that's not already accounted for in draft capital.

The Basics of the Model

The core of the model is yards per route run (or YPRR), which I've studied for years and have found to be very predictive of career outcomes. Yards per route run is exactly what it sounds like-- the number of yards a receiver gained divided by the number of routes he ran. In my opinion, this is the only true "efficiency" stat for receivers. (Many people like to use yards per target-- or YPT-- but YPT is a bad statistic for reasons both conceptual and practical that I'll detail in a bit.)

I'm further adjusting YPRR by adding a bonus for every touchdown. I've tested the model in the past and found that scoring at a disproportionate rate as a rookie does tend to carry predictive signal for the rest of a player's career.

There are several different ways to calculate "routes run". Some sites only count routes run on plays where a pass is attempted. Other sites count routes on any play where it's clear that the offense's intention at the snap was to pass the ball. (This means it counts routes on sacks and scrambles even though the ball was never thrown, but it doesn't count routes on draw plays or designed quarterback runs.)

There are pros and cons to each approach, but I'm using the latter definition of a "route run". Under this definition, any value over 2.0 is extremely good. Different methods produce different baselines; if you only count routes on attempted passes, a YPRR of 2.0 is less impressive.

Of course, if a receiver runs one route all year and catches a 13-yard pass on it, he'll have a YPRR of 13. We need some way to ensure small-sample guys like this don't dominate the model. I have two means of dealing with this.

The first is a qualifying threshold; receivers must run at least 250 routes to qualify for the model. On average, we see around 10 rookies a year reach that total. This year, we saw 18, which shattered the previous record (15 qualifiers from the legendary 2014 draft class).

The second way I protect against small samples is by including a "usage rate" term. Currently, I'm using (routes per game) per (team pass attempt per game). This means if a receiver averages 30 routes per game and his team throws 40 passes per game, his "usage rate" is 75%. When I've tested, I've found that penalizing players who only play in specific packages improves performance.

I normalize both terms so that the sample average results in a score of 100 and every standard deviation above or below adds or subtracts 15 points, and then average the two scores together, putting twice as much weight on the efficiency term as the usage term. This produces the final score.

(Note that because these values are normalized to the sample average and distribution, scores will change slightly over time as new data is added. These shifts are always small and rarely change the ordering of players.)

Why Yards Per Route Run?

There are two primary reasons. The first is conceptual: any "efficiency" stat should be "units of production divided by units of opportunity".

Many think that the target is the unit of opportunity for the wide receiver; you can't gain yards if you aren't targeted. But earning targets is a skill; if a bad receiver and a good receiver are both running a route on a play, the quarterback is more likely to throw to the good receiver than the bad receiver. Role players might post huge numbers on a per-target basis, but they're only earning a target when they're comparatively more wide open.

The second and more important reason to use YPRR is simple: because it works. If I rebuilt my model using YPT instead of YPRR (but kept everything else the same), the rookie receivers who would benefit the most are Kenny Stills, Mecole Hardman, J.J. Arcega-Whiteside, Gabriel Davis, Tre'Quan Smith, Dante Pettis, Hank Baskett, Jahan Dotson, Henry Ruggs III, Anthony Miller, Jalen Hyatt, DeVante Parker, Terrance Williams, Malcolm Mitchell, Tyler Lockett, Mike Wallace, Chester Rogers, Michael Wilson, Robert Foster, and George Pickens.

Despite my philosophical objections to Yards per Target, I would be glad to use it if it improved results, but Lockett and Wallace notwithstanding, that is not a list of receivers you wish you had been more invested in for dynasty. It largely fits with the conceptual case: they're mostly situational deep threats who posted a high yard per target average because YPT is biased towards deeper passes and because these players saw a disproportionate share of their targets on broken coverages.

On the other end, these are the receivers who would be downgraded the most by a move from YPRR to YPT: Tyreek Hill, Davone Bess, Puka Nacua, Chris Olave, Drake London, Odell Beckham Jr., Rondale Moore, Demario Douglas, Donnie Avery, Kelvin Benjamin, Rashee Rice, Percy Harvin, DeSean Jackson, Garrett Wilson, Doug Baldwin, Jarvis Landry, Jaylen Waddle, Cordarrelle Patterson, Kendall Wright, and Allen Robinson II. Again, it's not a perfect correlation-- I doubt managers would be upset about avoiding Kelvin Benjamin and Cordarrelle Patterson after their rookie years. But taken as a whole, that's definitely not a list of receivers you wanted less exposure to.

Results To Date

When presenting the data I often divide it into rough categories. This is merely for convenience-- scores are continuous, so a higher score is always better than a lower one. Notice that the players at the top of each group tend to have more in common with the players at the bottom of the group above than they do with the players at the bottom of their group.

With that out of the way, here are the previous qualifiers:

Superstars (Scores of 118+)

There's no such thing as a sure thing in football, but this is about as close as a receiver can get. I don't have dynasty valuation data from Colston's prime, but every other receiver on this list peaked as a Top 6 dynasty WR except for Dell (who ended his rookie year with a significant injury and didn't look the same in his follow-up campaign), Olave, and McLaurin, (who both peaked at 7th but have largely been held back by terrible quarterback play. With the best support of his career this year, McLaurin finished as the #7 receiver in fantasy.)

Strong Starters (Scores Between 108 and 117)

Here we see several misses starting to creep in, but around two thirds of this cohort became strong multi-year starters in fantasy and nearly a third became superstars, cracking the Top 6 dynasty receivers at some point.

Good Bets (103-108)

Players in this range still have elite upside, but the success rate begins to noticeably decline, especially towards the bottom. About half of the players in this group became multi-year fantasy starters.

Average Rookies (97-103)

By this point, there's not much meat left on the bone. Only about 33% of players in this group became multi-year starters, and Hopkins was the lone star to emerge. Smith-Njigba looks like he might have the potential to join him.

Bad Bets (93-97)

This group produced no stars and few starters. It's typically not worth considering any receiver in this range unless they're available quite cheap-- Chris Godwin (WR42), Michael Pittman Jr (WR45), and Tyler Boyd (WR57) all outperformed their ranking after their rookie year, but every receiver in this group who was valued within the Top 40 at their position strongly underperformed expectations.

Terrible Bets (<93)

This isn't quite "abandon all hope, ye who enter here" territory. Two of these receivers-- Davante Adams and Jordy Nelson-- became fantasy stars. A monster season from Nico Collins suggests he might join them. Jakobi Meyers has a few years as a WR3 / potential flex. Nelson Agholor had a pair of 8-touchdown campaigns.

That's pretty much all the positive production from this cohort.

How the Class of 2023 Fares

With our emptors properly caveated, it's time to get down to brass tacks. As we prepare for 2024, here's how last year's rookie class stacks up.

2023 joins 2019 (Brown and McLaurin), 2014 (Beckham and Evans), and 2011 (Green and Jones) as classes with two players who grade out at Superstar status.

There are some minor concerns with Dell. His 11 games was the fewest of any player in the Top 40-- although Odell Beckham Jr and Brandon Aiyuk played 12, while Stefon Diggs and Julio Jones played 13, and Dell ran 320 routes in his 11 games, which places him well above the minimum cutoff. Dell's touchdown rate was the 6th-highest in the sample, though the five players above him were Christian Watson, Beckham, Mike Evans, Tyreek Hill, and Ja'Marr Chase.

Meanwhile, Puka Nacua's resume is beyond reproach-- he posted the 5th-highest score in the model-- but he's just the second rookie to reach this tier despite not being drafted within the first three rounds, the first since Marques Colston back in 2006.

But the whole point of the model is finding players like these before the community is sold on them. Nacua is currently valued around WR7, but Dell is valued at WR20. At that price, the model can be wrong and he can still be a value.

Rashee Rice, Jayden Reed, and Zay Flowers all score as solid starters; remember, about 66% of receivers in this range go on to become multi-year fantasy starters. Jordan Addison and Dontayvion Wicks profile as Good Bets, while Michael Wilson, Josh Downs, Demario Douglas, and Jaxon Smith-Njigba finish with roughly average rookie seasons.

No rookies finish in the "bad bets" range, but a whopping seven receivers score as terrible bets, easily the most to date. (2018 had five, and no other class has more than four.) It seems much of the credit for this year's record number of qualifiers comes from teams giving large roles to rookies who likely weren't ready for them.

Is this a trend? We'll have to see more seasons to find out, but it's something to keep an eye on. The league is constantly shifting and I don't expect any model will work forever, so it's important to monitor any shifts for signs that the time is up.

How I Use These Results

It's always tempting to seek one-size-fits-all solutions, but unfortunately, the most successful approach is to consider all information available; I tend to use the model not as a rank-ordering of players, but as an additional datapoint. I move my original opinion of players up or down in response to their performance, but I do not ever overwrite that original opinion completely.

When the model ranked Christian Watson above Garrett Wilson in 2022, I wrote that I did not prefer Watson to Wilson as a result. I recognized that many elements of Watson's production were suspect; because of injuries, his route total was among the lowest of any receiver in the sample and while rookie touchdowns are typically meaningful, Watson's 2.59% touchdown per route run rate was the highest in the sample and was disproportionately impacting his score.

(Beckham, Evans, Hill, and Chase round out the Top 5 in touchdown rate. Again, the ability to score touchdowns as a rookie does carry positive signal. But touchdowns are still stochastic and prone to vary for reasons outside of a receiver's control.)

Meanwhile, Watson was also playing with Aaron Rodgers, who had a history of goosing his receivers' efficiency stats, while Wilson was in a much less functional environment. And Wilson was drafted higher and typically looked more impressive.

But Watson's high score did cause me to revise my opinion of his prospects upwards. And while I liked Wilson a lot before the season and his rookie year gave me no cause to downgrade him, I generally preferred London and Olave, who had similar draft capital but both scored higher in the model.

(As noted, many of these same concerns apply to Tank Dell this year.)

On the other end, while Nico Collins scored in the "terrible bets" range, I held him for years in one of my dynasty leagues simply because he was so cheap (WR74 in ADP after his rookie year). At the end of your roster, all players are terrible bets, but I liked Collins' size and draft capital and was willing to give him a bit of a pass on a very dysfunctional franchise.

Jaxon Smith-Njigba's score is much worse than we were hoping for given his draft position and college success, but his score does not completely negate his draft position and college success. I might revise my opinion of his prospects downward somewhat, but every situation is different (most receivers don't have to compete against veterans the caliber of DK Metcalf and Tyler Lockett) and every career follows a unique path.

In my opinion, the value in the model largely isn't in the highly-regarded players with big scores. You certainly didn't need my model to tell you to buy Ja'Marr Chase, Odell Beckham Jr, or Justin Jefferson after record-breaking rookie years. The true value, in my opinion, is in the players that were drafted later and who the community is still lukewarm on after their rookie campaign: A.J. Brown, Terry McLaurin, Stefon Diggs, Cooper Kupp, Jayden Reed, Doug Baldwin, Tyler Lockett, and the like.

When the model prompts me to buy or sell, I always try to index to prevailing market rates. In hindsight, Amon-Ra St. Brown would have been a bargain even if you paid WR10 prices to acquire him, but by buying closer to his WR22 price tag, you maximize your potential for profit and minimize your downside risk.

I also find that the model is valuable not just in the offseason immediately following, but for years afterward. I found myself more likely to roster Stefon Diggs, Cooper Kupp, and Hunter Renfrow for years because of their strong rookie campaigns; all three eventually rewarded my belief. 

Similarly, when Drake London opened the season with a 0-catch game, I became a committed buyer in large part because of his bulletproof rookie score. He finished his follow-up season with 39 more yards in one fewer game and his value has risen as the Falcons will look for a quarterback in free agency or the draft.

I am absolutely begging you not to trade Drake London for Zay Flowers.

Please. https://t.co/Ykh7H6o4qq

— Adam Harstad (@AdamHarstad) September 12, 2023

Anyway, that's how I use the model, but how you use it is ultimately up to you. If you want to treat it as a straight rank-ordering, I won't stop you; certainly you could do much worse. If you want to ignore whatever results don't suit your priors, that's fine, too (I did as much with Collins and was rewarded for it).

Even when this column produces something practical, it's not especially concerned with what you do with it. I want to provide more data for you to consider, but most importantly, I want you to consider how best to use that data.