Loading scores...
Fantasy Usage Model

Winks' 2021 QB, RB, WR, TE Draft Models

by Hayden Winks
Updated On: April 7, 2021, 11:09 am ET

This year’s model is probably going to be worse than in years prior because of sample size and athletic testing issues. Most colleges played fewer games in 2020 and the pro day numbers have been all over the place. It’s gotten to the point where I was thinking about not even posting these, especially this early when athletic testing numbers are often unofficial. But I’ll share what I have anyways. Just know that some prospects are NOT included because of data collection issues and that these percentiles will change in the three weeks leading up to the actual draft. Are we all on the same page? Good.

 

I also have an announcement to make. This is my last column for NBC Sports. I’d like to thank Ed Williams and Evan Silva for getting me hired three years ago, and RotoPat, Norris, Daigle, and Mensio for the mentorship since then. This has been a very fun environment to work in, and I’ve learned a lot in a short amount of time because of these fine gentlemen. I take pride in working for a company with such a deep history in sports media, and I look forward to representing Rotoworld’s stacked alumni as I move on. For my readers and listeners, I don’t plan on being a stranger. If you ever want to reach out, visit me on Twitter (@HaydenWinks). You have all made this a memorable experience. Much love!

 


Quarterback Model

All numbers are percentiles among drafted players since 2005, so 0.50 would mean average for drafted FCS quarterbacks. The model incorporates draft capital (essentially a film grade), plus the production and schedule grades. The “Production” metric is EPA-based for each quarterback’s final collegiate season, and it’s age-adjusted because younger quarterbacks are better. It’s the foundation of the entire model.

Prospect

Model

Production

Schedule

Justin Fields

0.94

0.89

0.91

Mac Jones

0.93

0.97

0.92

Trevor Lawrence

0.90

0.85

0.43

Zach Wilson

0.88

0.85

0.13

Kyle Trask

0.76

0.87

0.88

Kellen Mond

0.65

0.53

0.75

Davis Mills

0.47

0.72

0.17

Jamie Newman

0.37

0.40

0.27

Sam Ehlinger

0.17

0.50

0.54

Ian Book

0.13

0.67

0.32

Feleipe Franks

0.08

0.08

0.98

Shane Buechele

0.01

0.60

0.05

Trey Lance

NA

NA

NA

 

All five consensus top-ranked quarterbacks are worthy of a high selection per my model, just not in the order that most assume. The model has Fields ranked above Wilson by a decent margin mostly because Wilson’s strength of schedule is down at the 13th percentile while Fields’ is at 91st. It’s not a total dealbreaker, but BYU's Mickey Mouse schedule makes Wilson’s projection volatile. The same can obviously be said for Lance, who destroyed the FBS in his lone season.

The other big takeaway is Jones’ favorable rating (93rd percentile), which stems from his 97th percentile age-adjusted EPA and his 92nd percentile strength of schedule. Both of those percentiles were better than Tua Tagovailoa’s (86th and 80th) in 2019, yet they are treated very differently by most draft analysts and fans. I’m on team “QBs need to be mobile in today’s NFL”, but Jones’ outlier season is very hard to ignore. The only prospects in my database with a higher age-adjusted EPA than Jones are Kyler Murray, Joe Burrow, Marcus Mariota, Patrick Mahomes, and Jalen Hurts (with Baker Mayfield, Cam Newton, Colin Kaepernick, and Lamar Jackson rounding out the top 10). I wouldn’t draft Jones over Fields, but I view Jones as a reasonable top-five selection for a quarterback-needy franchise. If that gets me cancelled, oh well.

The next four quarterbacks (Trask, Mond, Mills, Newman) are the last of the quarterbacks my model would suggest drafting, but all are total long shots for various reasons. Trask (76th percentile) has the best odds of earning a long-term backup job after a very good senior season. His ceiling is just lower than the three others as an in-pocket only player with a forgettable film grade. Round 3, 4, or 5 feels more appropriate than Round 2 for this tier.

 


Running Back Model

All numbers are percentiles among drafted players since 2005, so 0.50 would mean average for drafted FCS running backs. “Production” in this case is team- and age-adjusted, favoring younger prospects from better schools. It’s the baseline of the model with “Athleticism” (aka Adjusted SPARQ) playing just a minor role in the final model’s projection. As always, my model incorporates projected draft capital (essentially a film grade). 

Prospect

Model

Production

Athleticism

Travis Etienne

0.95

0.90

0.93

Javonte Williams

0.93

0.95

0.54

Najee Harris

0.91

0.96

NA

Kenneth Gainwell

0.70

0.87

0.04

Michael Carter

0.66

0.70

0.35

Trey Sermon

0.64

0.63

0.73

Rhamondre Stevenson

0.57

0.77

0.16

Chuba Hubbard

0.50

0.60

0.46

Kylin Hill

0.45

0.35

0.39

Jermar Jefferson

0.42

0.67

0.10

Jaret Patterson

0.39

1.00

0.11

Demetric Felton

0.28

0.61

0.00

Elijah Mitchell

0.27

0.24

0.53

Khalil Herbert

0.26

0.27

0.41

Larry Rountree III

0.19

0.22

0.06

Javian Hawkins

0.16

0.64

0.05

 

Like everyone already believes, my model says there’s a clear top-three at running back this year. Any of the three could be ranked RB1, although all three backs bring different skill sets to the table. As long as the team fit makes sense, Etienne, Harris, and Williams project as fantasy starters in dynasty and possibly in redraft.

There’s a big drop off after them, however, leaving a debate for this year’s RB4. Gainwell (70th percentile), Carter (66th), Sermon (64th), Stevenson (57th), and Chubbard (50th) are all in the mix, but only Sermon has a potential three-down workload in my opinion. If I were to bet on any running back straying away from my model’s projection, it’d be Sermon because injuries and a transfer have suppressed his college production. If I’m betting on outliers, I like to do it with efficient, big school prospects who pass athletic thresholds.

Athleticism tends to be overrated at the position, but more than a handful of prospects had tough pro days in my Adjusted SPARQ metric. Gainwell had a very concerning three-cone, 10-yard split, and speed score for an undersized running back (201 pounds), and it’s something I noticed while watching his tape before seeing his workout metrics. Stevenson, Jefferson, Patterson, Felton, Rountree, Hawkins, and others unfortunately joined Gainwell as sub-25th percentile size-adjusted athletes.

Editor’s Note: Get an edge with our premium Betting Tools that are packed with live odds, betting trends, predictions, player prop projections, our extensive Edge Finder and much more. And don't forget to use promo code WELCOME10 to get 10% off. Click here to learn more!


Receiver Model

All numbers are percentiles among drafted players since 2005, so 0.50 would mean average for drafted FCS receivers. For whatever reason, modeling receivers has had the best results of any position compared to draft capital. Most of my model is “Production” based, which is a combination of age, team strength, team share of receiving, and other receiving stats. That paired with “Athleticism” (aka Adjusted SPARQ) and projected draft capital (essentially a film grade) does the trick.

Prospect

Model

Production

Athleticism

Ja'Marr Chase

0.99

0.99

0.89

Devonta Smith

0.98

0.99

NA

Elijah Moore

0.94

0.99

0.90

Jaylen Waddle

0.92

0.36

NA

Rashod Bateman

0.91

0.97

0.48

Terrace Marshall

0.88

0.93

0.68

Rondale Moore

0.88

0.89

0.97

Kadarius Toney

0.81

0.67

0.72

Dyami Brown

0.80

0.92

0.65

Amon-Ra St. Brown

0.64

0.74

0.31

Tutu Atwell

0.64

0.77

0.47

Amari Rodgers

0.63

0.86

0.56

Nico Collins

0.59

0.49

0.67

Tylan Wallace

0.58

0.72

0.04

Ihmir Smith-Marsette

0.57

0.82

0.51

D'Wayne Eskridge

0.57

0.64

0.82

Anthony Schwartz

0.49

0.75

0.62

Jaelon Darden

0.48

0.79

0.47

Seth Williams

0.45

0.47

0.44

Marlon Williams

0.43

0.96

0.02

Sage Surratt

0.37

0.58

0.10

Shi Smith

0.35

0.42

0.54

Tamorrion Terry

0.28

0.36

0.67

Trevon Grimes

0.22

0.37

0.24

Cornell Powell

0.20

0.31

0.65

Dazz Newsome

0.11

0.32

0.03

Tyler Vaughns

0.05

0.23

0.11

 

This is not every receiver in the class because I'm still waiting on more numbers. Apologies!

Overall, this class is good, but it’s not as deep as last year’s according to my model. Only nine receivers sit above the 65th percentile, and many of the top prospects are slot receivers. The surprises are with Elijah Moore (93rd percentile) and Dyami Brown (80th). If I were to slap “My Guy” labels on players in this class, it’d be them because of their production numbers, above-average athleticism, and (wait for it) ... their tape. I think Moore will be a high-end slot receiver and Brown will be a quality No. 2 at the next level.

Compared to how high they’ll be drafted, Waddle and Toney have below-average production. They might be overrated by some, but it doesn’t mean they are must avoids in fantasy circles either. Some in-depth dynasty analysts tend to overlook team strength, recruiting, and consensus film grades in my opinion. I'm so done ranking Andy Isabella types highly.

Unfortunately, some of the depth of this class was hurt by athletic testing. Amon-Ra St. Brown (31st percentile Adjusted SPARQ), Tutu Atwell (47th), Tylan Wallace (4th), and Sage Surrat (10th) all dropped a tier and now should be treated as role players rather than locked-in future starters. I’d also be lying if Rashod Bateman’s 48th percentile Adjusted SPARQ athleticism wasn’t disappointing, but he’s still a 91st percentile prospect overall. That athleticism just might be the difference between Round 1 and Round 2 draft capital for what it’s worth.

 


Tight End Model

All numbers are percentiles among drafted players since 2005, so 0.50 would mean average for drafted FCS receivers. The model is very similar to the receiver model, except that size and athleticism matter more at tight end. 

 

Model

Production

Athleticism

Kyle Pitts

1.00

1.00

0.95

Pat Freiermuth

0.85

0.77

NA

Brevin Jordan

0.75

0.96

0.13

Jacob Harris 0.70 0.48 1.00
Tommy Tremble 0.67 0.65 0.57

Hunter Long

0.64

0.68

0.32

Nick Eubanks

0.24

0.18

0.70

Noah Gray

0.17

0.05

0.80

Tony Poljan

0.09

0.49

NA

Kenny Yeboah

0.05

0.50

NA

Tre McKitty

0.04

0.24

NA

 

Breaking news: Kyle Pitts is a boss. After showcasing 95th percentile athleticism, Pitts is now the top-ranked tight end in my model’s history. He’s a total outlier at the position and deserves consideration as the first non-quarterback off the board. If past tight end contracts are any indication, Pitts’ second contract will be cheaper than Ja’Marr Chase’s, so I’m firmly on team Pitts over Chase. 

The rest of the class is just okay. Freiermuth is in a tier of his own per my model, but he didn’t do athletic testing, so this is a leap of faith that he's a second-round pick still. Jordan did test, and he basically failed (13th percentile Adjusted SPARQ). He looked more athletic on tape, but he’s now in outlier territory if he were to have a successful NFL career. I’m anticipating Jordan dropping into Round 3 or 4 now. That leaves the door open for Tremble (67th percentile prospect) or Long (64th percentile prospect) as the TE3 of this class. After watching Tremble block, I won’t be surprised if he’s drafted as the No. 3 tight end somewhere near the Round 2/3 turn. That’s where I’d draft him at least.

Easily the best sleeper of the class is UCF's Jacob Harris, who is 24 years old but was productive as a redshirt senior and set the record for Adjusted SPARQ athleticism at the position in my database. He's a massive, fluid mover that received a Round 5 grade from The Athletic's Dane Brugler. Teams are now beginning to view him as a tight end rather than a receiver. Pay attention dynasty guys.

 

Once again, thank you for the support!!! See you guys on the other side!