This past offseason, I created a spreadsheet listing every wide receiver who entered the league since 2003 and included data like college-reception numbers, NFL games played, and fantasy points. Some of you have requested and received access to it. It's a pretty solid data set -- more than 1,000 players. So there are plenty of ways to slice it.
The proof in how effective these spreadsheets are is how clearly we can draw conclusions, how confidently we can apply those conclusions toward making predictions, and (of course) how accurate we are. Because anyone can draw conclusions and make predictions. The key is whether it's actionable.
Writing for another publication in May (if you want to see it, it's easily searchable), I wrote a piece on Lions fifth-round rookie WR Kendrick Law's fantasy outlook. He had only 86 career college catches. Based on all the retired fifth rounders who entered the league since 2003, Law's career fantasy outlook appeared dim compared to fifth rounders with considerably more receptions. It had nothing to do with his talent level or even his difficult depth-chart situation in a crowded receiver room (though that didn't help). It had *everything* to do with what my spreadsheet showed. If there's a 60% or 70% or greater chance of something happening, I'll shout it out as moderately to extremely actionable.
Law tore his ACL a few days later. Probably a fluke incident. It doesn't add much to the conclusion I've already drawn. Law could return next year and shine.
But I've returned to this spreadsheet a lot this summer, wondering if/when some of the conclusions are actionable enough. Because if fifth-round WRs with 51-100 college catches average barely half the career fantasy output of fifth-round WRs with more than 100 college catches, that's at least worth sharing. In some cases, it's worth acting on. But when do we know we have enough intel to shout it out?
I like studying the extremes (e.g., receivers with 250+ career college catches or RBs with 1,000+ career college touches), because they usually offer the most interesting results. The problem is that "extreme" data sets often are really small. So it takes only one outlier to skew the findings. That means it can take years to beef up the data set enough where it really makes sense.
High concentrations of usage fascinate me. On average, college running backs who get most of their touches in one season perform worse than college running backs with more evenly distributed touches. So last night I took that idea and tested it out on a subset of WRs, threading the needle between extreme and actionable by focusing on the "best" wide receivers coming out of college -- in this case, those drafted in the first or second round since 2003. The set consists of 118 retired players.
I divided the set into two groups. In Group A, the WRs' highest-catch college season accounted for at least 60% of their total college catches. For example, D.J. Chark (2018 second rounder) had 66 catches at LSU, including 40 in his final season, which accounted for 61% of his total college receptions.
Group B included everyone else: WRs whose highest-catch college season accounted for *less* than 60% of their total college catches. For example, Laquon Treadwell (2016 first rounder) had 202 college receptions, including 82 in his final season, accounting for 41% of his total. Group A's retirees averaged 343.7 career fantasy points. Group B averaged 804.1. That's a pretty big difference.
Digging deeper, I divided the much larger Group B into smaller parts. Group BA consisted only of first-/second-round retired WRs from 2003 onward whose largest college-reception season accounted for 50%-59% of their career college receptions. Group BB was the 40%-49% group. Group BC was everyone below 40%.
Group BA averaged 949.4 career fantasy points. BB: 875.7. BC: 585.2. Then I isolated the BA group to determine where the sweet spot was. Turns out it's 49%-59%, encompassing 31 retired receivers who averaged 1,047.4 career fantasy points, including the three highest-fantasy-scoring first-/second-round WRs since 2003: Larry Fitzgerald, Anquan Boldin, and Andre Johnson. Remarkably, only one of the 29 highest-fantasy-scoring WRs in this 118-player data set had a college-reception season that accounted for 60%+ of their total college receptions (Brandon Marshall).
Is this actionable? Not yet, because a quick glance at young, active former first/second rounders shows this is still a work in progress. Among the Group A wideouts (60%+ of career college receptions concentrated in one season), four are headed toward strong or stellar careers: Ja'Marr Chase, Justin Jefferson, Jaxon Smith-Njigba, and Jameson Williams. The other four are . . . well, they're John Metchie, Xavier Legette, Omar Cooper Jr., and John Ross.
Yes, I still have Ross on here and need to push him into the retired group. He officially retired in 2023 and then unretired in 2024. I generally wait a couple years before checking this box. But he's probably overdue.
Does this all mean there's no actual disadvantage to a 60%+ concentration of college receptions? After all, Chase and Jefferson and JSN and Williams should push the averages way up when their careers are over.
But the "sweet spot" group (49%-59%) also have plenty of young talent who might be poised for (or are well on their way toward) good/great careers, including Makai Lemon, Brian Thomas, Quentin Johnston, Ladd McConkey, Drake London, Garrett Wilson, George Pickens, DeVonta Smith, D.K. Metcalf, Michael Pittman, D.J. Moore, Mike Evans, Davante Adams, . . . and so on.
The challenge of data analytics is not knowing when we have enough information to do something with it. The good news is that with each passing year, the data set grows, and so does the likelihood that we *can* do something with it.
---