How to Spot a Fantasy Football League Winner: WRs
The phrase “league winner” gets bandied about a lot in fantasy football circles, often carelessly. Sometimes, analysts use this phrase as nothing more than an attention-getter to hype some mediocre player they love. But when used properly, there is truth to this concept – the idea that one single player can dominate for your team enough to nearly guarantee, on his own, a championship for your fantasy team.
Of course, no player can truly win a fantasy championship alone. Even Christian McCaffery’s historic 2019 season, where he scored a ridiculous 413.2 fantasy points, was only enough to get 48% of his managers into fantasy championships (still an absurd number, by the way). Nevertheless, rostering some players undoubtedly gives fantasy teams an increased chance at a fantasy championship, just like McCaffery nearly gave his managers a 50-50 shot in 2019.
From my analysis of the past five years, league-winning players typically separate themselves from the pack. That means that the top 1-3 players each year, at any given offensive position, generally score at least 30 points more than their next closest competitor, where the end-of-year rankings start to bunch up. That’s good! We want our “league winner” not just to score a few more points than RB3 or WR4; we want them to score way more points. That means these types of players really do have “league winning upside”.
In this series, I define “league winners” for each of RB, WR, QB, and TE, I analyze historical league winning seasons at all four positions, and I use history to try and spot league winners for 2021. I can’t wait to take you along for the journey. I hope you enjoy the series, and if you missed the league-winner RB article, go back and read that one too!
Defining a League Winning Wide Receiver
As I said above, I studied the fantasy finishes of the wide receiver position over the last six years. I wanted to find players who stood out, not just finished WR1 overall, but player(s) who separated themselves from the crowd. What I found was that very few wide receivers hit the 260-fantasy point threshold (in 0.5 PPR scoring), and those who did separate themselves from the rest of the top 12 wide receivers. Thus, this threshold worked quite well in establishing an objective definition of a league-winning wide receiver.
By setting that 260-fantasy point threshold, I was able to generate seventeen instances of a league-winning wide receiver in the past 6 years, which also “felt about right” (about 2-3 per year).
|League Winning WRs|
|Allen Robinson II (2015)|
|Odell Beckham (2015)|
|DeAndre Hopkins (2015)|
|Brandon Marshall (2015)|
|Julio Jones (2015)|
|Antonio Brown (2015)|
|DeAndre Hopkins (2017)|
|Antonio Brown (2017)|
|Julio Jones (2018)|
|Antonio Brown (2018)|
|Davante Adams (2018)|
|DeAndre Hopkins (2018)|
|Tyreek Hill (2018)|
|Michael Thomas (2019)|
|Stefon Diggs (2020)|
|Tyreek Hill (2020)|
|Davante Adams (2020)|
This is our set of league winners. Now let’s see if we can find out why these players ended up as league winners.
Historical Analysis of League Winning Wide Receivers
Now that we have a nice definition and a decent sample size, we can start to look at these players to decide what is important in predicting fantasy league winners.
In my research leading up to this article, I generated a list of twenty-two possible statistical or other factors that explain why these seventeen instances, represented by eleven different wide receivers, resulted in a league-winning season. Those twenty-two factors include catches, targets, yards, ADP, team wins, team passes, offensive plays, run-pass ratio, QB teammate’s fantasy finish, snap share, age, touchdowns, air yards, RACR (Receiver Air Conversion Ratio), slot snaps percentage, YAC/reception, target share, red-zone targets, team targets to WR, average separation, and Y/RR (yards per route run). In other words, I analyzed a ton of data, so much data in fact that it doesn’t fit nicely onto one chart.
|League Winning WRs||Fantasy points||Catches||Targets||Yards||ADP|
|Allen Robinson II (2015)||264||80||151||1400||79|
|Odell Beckham (2015)||271.3||96||158||1450||17|
|DeAndre Hopkins (2015)||275.6||111||192||1590||33|
|Brandon Marshall (2015)||284.7||109||173||1502||59|
|Julio Jones (2015)||307.1||136||203||1871||16|
|Antonio Brown (2015)||314.2||136||193||1834||6|
|DeAndre Hopkins (2017)||261.8||96||174||1378||32|
|Antonio Brown (2017)||259.8||101||163||1533||3|
|Julio Jones (2018)||269.3||113||170||1677||14|
|Antonio Brown (2018)||271.7||104||168||1297||5|
|Davante Adams (2018)||274.1||111||169||1386||18|
|DeAndre Hopkins (2018)||276||115||163||1572||10|
|Tyreek Hill (2018)||284.5||87||137||1479||27|
|Michael Thomas (2019)||300.1||149||185||1795||13|
|Stefon Diggs (2020)||265.1||127||168||1535||53|
|Tyreek Hill (2020)||285.4||87||134||1276||16|
|Davante Adams (2020)||300.9||115||149||1374||14|
|League Winning WRs||Wins||Plays||Run-Pass Ratio||Team WR Targets||Pass Attempts||QB Finish|
|Allen Robinson II (2015)||5||1012||59.98%||61.50%||607||4|
|Odell Beckham (2015)||6||1053||59.16%||58.00%||623||10|
|DeAndre Hopkins (2015)||9||1127||54.92%||68.80%||619||16|
|Brandon Marshall (2015)||10||1074||56.24%||75.00%||604||11|
|Julio Jones (2015)||8||1073||57.88%||61.80%||621||19|
|Antonio Brown (2015)||10||1011||58.36%||69.60%||590||13|
|DeAndre Hopkins (2017)||4||1027||51.12%||63.80%||525||13|
|Antonio Brown (2017)||13||1051||56.14%||64.20%||590||10|
|Julio Jones (2018)||7||1010||61.09%||67.60%||617||2|
|Antonio Brown (2018)||9||1058||65.12%||66.10%||689||3|
|Davante Adams (2018)||6||1026||62.38%||62.30%||640||6|
|DeAndre Hopkins (2018)||11||1040||48.65%||66.90%||506||4|
|Tyreek Hill (2018)||12||996||58.53%||51.80%||583||1|
|Michael Thomas (2019)||13||1011||57.47%||51.90%||581||6|
|Stefon Diggs (2020)||13||1034||60.93%||74.90%||630||1|
|Tyreek Hill (2020)||14||1057||56.39%||58.00%||596||4|
|Davante Adams (2020)||13||990||53.13%||54.40%||526||3|
|League Winning WRs||Snap Share||Age||Touchdowns||Total Air Yards||RACR||Slot %|
|Allen Robinson II (2015)||94.6%||22||14||1525||0.92||15.1%|
|Odell Beckham (2015)||93.3%||23||13||1912||0.76||26.1%|
|DeAndre Hopkins (2015)||94.6%||23||11||2803||0.57||15.0%|
|Brandon Marshall (2015)||96.1%||31||14||2197||0.68||22.7%|
|Julio Jones (2015)||93.5%||26||8||2132||0.88||28.3%|
|Antonio Brown (2015)||91.7%||27||10||2162||0.85||19.6%|
|DeAndre Hopkins (2017)||91.3%||25||9||2436||0.57||15.7%|
|Antonio Brown (2017)||93.8%||29||13||2347||0.65||12.4%|
|Julio Jones (2018)||94.5%||29||8||2431||0.69||20.8%|
|Antonio Brown (2018)||93.6%||30||15||1949||0.67||20.5%|
|Davante Adams (2018)||94.0%||26||13||1977||0.70||20.7%|
|DeAndre Hopkins (2018)||93.4%||26||11||1989||0.79||19.7%|
|Tyreek Hill (2018)||92.9%||24||12||2178||0.68||44.4%|
|Michael Thomas (2019)||93.9%||26||9||1517||1.18||31.8%|
|Stefon Diggs (2020)||93.5%||27||8||1932||0.79||31.5%|
|Tyreek Hill (2020)||94.2%||26||18||1729||0.74||58.6%|
|Davante Adams (2020)||94.8%||28||15||1296||1.06||33.6%|
Finally, I plotted each of the twenty-two factors against their fantasy points scored (and calculated the R2 of each plot) and also calculated the average value of each column from above. I am going to use the term R2 quite a bit below – without getting too deep into the mathematics, just understand that R2 is basically a measurement representing variance. We can use it to measure how closely two sets of data (e.g. targets and fantasy points) correlate with each other linearly.
Here is what I found:
- All team-level metrics failed to impact our league winners in any statistically measurable way. Team wins weren’t important, nor was the run-pass ratio or team pace. Equally interesting, the number of targets going to wide receivers vis-à-vis the other offensive positions barely correlated with league-winning production. In other words, NFL team tendencies and passing predispositions have no impact on whether or not an NFL team will produce a league winner. This finding completely flies in the face of conventional wisdom. For example, it has been argued on this very site that A.J. Brown cannot outperform his ADP because he plays on a team that tends to run more than they pass. The data analyzed here suggests that this narrative is wrong. Passing volume, passing volume to wide receivers, and all team-level metrics do not affect whether or not a player can become a league winner. Ignore the team-level narratives.
- QB performance had a weak R2, but the average QB fantasy finish was in the top 10 (7.41). Thus, rather obviously, QB performance is indicative of league-winning success. Notably, Julio Jones achieved league-winning status attached to Matt Ryan in 2015, when Ryan finished as the QB19. Still, I think it’s fair to say that a league-winning WR will be associated with a top-12 fantasy QB.
- Rather surprisingly, target volume and air yards did not correlate with league winners as much as some other statistics. I came into this exercise expecting both to be strong indicators of league-winning success at the wide receiver position. Still, the 100 threshold limit seems to be the bare minimum.
- Analyzing snap share produced no interesting findings, which is to be expected considering that WR1’s rarely come off the field. However, in almost every case, the league winner led his team in WR snaps. So, we are looking for a team’s WR1, but that’s to be expected. Rarely did I see a league winner having a teammate who finished the year high in end-of-year fantasy rankings – the closest was JuJu Smith-Schuster (WR8 in 2018 when Antonio Brown was WR4 )and Eric Decker (WR13 finish in 2015 when Brandon Marshall finished as WR3). Still, only two instances out of seventeen should not dissuade us. The lesson is that a league winner should be the undisputed WR1 of their own team.
- Age doesn’t seem to matter much either. We had a player as young as 22 and as old as 31, so I won’t be limiting my search for a 2021 league winner by age.
- The percentage of snaps in the slot did not reveal much, but almost all league winners spent at least 15% of their snaps in the slot for an average of 33% in the slot. Our league winner needs to do everything well, including run the occasional slot route.
- The lowest target share was 22.48% and the average was 28%. It’s safe to say we are looking for a player who expects to see 28% or more of his team’s targets. I suppose this is really just another way of saying that we are looking for a team’s undisputed WR1.
- Red zone targets and touchdowns correlated quite well with fantasy points scored, as expected. The average was 12 TDs among our league winner dataset.
- ADP is not particularly telling. Unlike RB, league-winning WRs come from all over the draft board – as high as 1.03 and as low as pick 79 (7th round in a 12 team league). The average was pick 24, so looking for league winners in the first three rounds is a fair cut-off.
- The strongest correlation I found with fantasy success was RACR and average separation (you can see the plots below), but Y/RR and aDOT were not far behind. But, it should be noted that none of these aDOTs lead the league. Generally, these are middle of the road aDOTs, which indicate that these league-winning WRs win at every level of the field, not just on deep, go-routes. Collectively, these statistics represent WR talent. For WRs, their ability to score fantasy points is less reliant upon other teammates. Usually, they need to simply win a one-on-one matchup with a cornerback or find a soft spot in the zone. The best wide receivers do this inherently, regardless of how many times the team throws it per game. Talent overcomes team tendencies at the WR position.
I know that’s a lot to take in, but I had to “show my work”. I will try to distill what I found into something more digestible.
- Team-level metrics don’t matter for the purposes of discovering a league-winning WR. For example, the Seahawks want to be a run-heavy offense, but that doesn’t hurt D.K. Metcalf‘s chances to be a league-winning WR. We want talented WRs more than WRs in good situations.
- We look for players that win at all levels of the field and utilized everywhere.
- QB talent matters, but not as much as we think. Individual WR talent still matters more.
- A league-winning WR is typically the undisputed WR1 on his team, thereby leading to a dominating target share (+28%).
- The fantasy community is good, but not great, at unearthing league-winning WRs, so our league winner likely has an ADP in the top 3 rounds.
- RACR and separation are the best measures of WR talent that we currently have. We will use these numbers to objectively measure “WR talent”.
Now, let’s use the UDK’s projections and the UDK’s ADP to see if we can find a league-winning wide receiver for 2021!
Who is the 2021 League Winner?
Yet again, the top drafted option arrived at the top for a reason. Davante Adams checks all our boxes. His RACR last year was 1.06, which is outstanding, and his average separation was 2.8, which is solid. Add on that he’s affiliated with a top-tier QB, is being drafted in the first three rounds, and has had a >30% target share each of the last two years. However, I am not going to plant my flag on him because his health is always a concern (hasn’t played a full season in the last four years). Health is very important at the WR position. So, let’s look a little deeper.
The stars have aligned, and it’s Calvin Ridley‘s time to shine. Ridley is the undisputed WR1 on his team, now that Julio has been shipped out of town, and we know that he can handle the added pressure because he has always performed better for fantasy purposes whenever Julio misses time.
Ridley had the most air yards in 2020 in the NFL and had the seventh-most targets. While his target share is a bit lacking (26%), it’s fair to assume that number will explode in 2021 without Julio demanding targets. Finally, Ridley remains tied to Matt Ryan for at least one more season. Ryan is being drafted as QB15 but has finished as high as QB2 in the past. There’s no reason to believe that Ryan cannot support a league-winning year from Ridley.
All that is nice, but the objective “talent” numbers really demonstrate Ridley’s potential to explode. Ridley was 5th in Y/RR in 2020 and was second in the league in red-zone targets. Ridley’s RACR leaves something to be desired (.66), but several of our league winners had RACRs at or about that number. Ridley had 2.8 yards of separation per reception in 2020, but historically he’s closer to 3, which is fantastic. Ridley is primed to explode.
Andy has Ridley projected for 101 receptions, 1475 yards, and eight touchdowns, but I personally think he’s low on all of them. I side more with Mike’s projections for Ridley, showing 108 receptions, 1664 yards, and nine touchdowns.
Jefferson is arguably the most talented wide receiver in the NFL. He finished second in PFF receiving grades in 2020 despite being a rookie and having a limited role in the first five games. His RACR was an incredible 0.96, which is behind only Michael Thomas‘s 2019 and Davante Adam’s 2020 in this efficiency metric among our league winners. Surprisingly, Jefferson’s separation was only 2.6, but he more than makes up for that with his ability to accumulate yards after the catch (8th in the NFL).
Jefferson’s target domination numbers were only around 24% in 2020, but again, his role expanded as the year rolled on. There’s no reason to doubt that this number won’t approach 30% in the upcoming season.
Moreover, Jefferson finished the season second in Y/RR. Said differently, Jefferson’s talent metrics pop off the page. Sure, the Vikings likely won’t be as efficient passing the ball in the red zone, but that regression affects Adam Thielen more than Jefferson. A Thielen regression is likely a Jefferson gain.
Andy has Jefferson projected for a crazy-efficient 100 receptions, 1615 yards, and eight touchdowns. If Jefferson breaches 10 TDs, which I think he will, he’s a league winner easily.
Well, that wraps it up for the Wide Receiver Position. Stay tuned this whole month as I will do a similar exercise with TE and QBs, and don’t forget to read the RB League Winner article if you haven’t! Now, go win some championships!
Amazing work. I am curious what is the percentage of landing on “league-winning” WR for those that have hit the key metrics (RACr, Average Seperation, target number, etc.)? It’s another way besides looking at correlation to see how accurate these metrics are.