Projecting the Pros on DraftKings: Week 8

A guide to playing roster percentages like the pros.

NOTE: The pro projections below are first posted on Thursday, then updated on Saturday afternoon. On Thursday, commentary will be limited to early impressions. A more-detailed discussion of specific players will be added on Saturday afternoon.

Welcome back for Week 8 of Footballguys' new feature that uses a statistical model to project the roster percentages of "professional" tournament players on Draftkings. If you're curious about the full methodological details of the model, click here. The most important detail to keep in mind, however, is that the "professionals" in question were identified using a clustering algorithm and can be characterized by the following statement: Compared to the rest of the population, these 205 pros play more, they win more, and they do both consistently.

Quarterbacks

Below is the table for quarterbacks. In this and all subsequent tables, "Overall" is Steve Buzzard's projection of roster percentage. "Pros" is the model's projection of roster percentage for Draftkings pros only.

Quarterback Team Salary Overall Pros
Jared Goff LAR 6000 11.8% 14.5%
Patrick Mahomes II KC 7000 14.2% 13.5%
Ben Roethlisberger PIT 6100 10.9% 10.6%
Aaron Rodgers GB 6400 10.2% 10.5%
Andy Dalton CIN 6200 8.1% 7.0%
Jameis Winston TB 6000 7.0% 5.0%
Andrew Luck IND 6300 6.5% 4.7%
Matthew Stafford DET 5600 3.6% 3.9%
Derek Carr OAK 5200 2.6% 3.7%
Mitchell Trubisky CHI 6100 4.7% 3.7%
Case Keenum DEN 5100 3.7% 3.2%
Joe Flacco BAL 5300 1.8% 3.0%
Russell Wilson SEA 5900 3.0% 2.7%
Eli Manning NYG 4900 1.3% 2.6%
Cam Newton CAR 5800 2.4% 2.5%
Alex Smith WAS 5000 2.0% 2.3%
C.J. Beathard SF 4900 1.6% 2.2%
Josh Rosen ARI 4800 1.2% 2.0%
Sam Darnold NYJ 4500 1.2% 1.6%
Baker Mayfield CLE 5700 2.1% 0.8%

Early Impressions

Another week, another recommendation to roster Jared Goff heavily. Presumably because of having a slightly tougher matchup, his salary decreased from Week 7 decreased by $600; which is great because said matchup isn't that much worse. With respect to factors that go into the projection model, Goff benefits from having what is likely to be the most highly rostered wide receiver corps on the Sunday Main slate. If they do well, then he has to do well by definition.

SATURDAY UPDATE: Goff remains the pick of the pro litter. And everything that's changed statistically since Thursday has moved in Goff's favor. His 4x value probability has increased from 28.6% to 29.4%. His wide receiver corps has seen their aggregate projected roster percentage increase from 39.2% to 41.4%. And the projected roster percentage for his opponent defense has decreased from 2.3% to less than 1.0%.

Eli Manning ($4,900, 1.3% Overall Roster Percentage, 2.6% Pro Roster Percentage)

It continues to amaze how a quarterback having his best statistical season since at least 2015 -- while playing behind an awful offensive line, mind you -- continues to get guff on the feed. At $4,900, Manning's one of the lowest-priced quarterbacks on the Sunday Main slate, yet with the seventh-highest 4x value probability. Washington cornerback upstart Quinton Dunbar is on the wrong side of questionable, which should open things up for Odell Beckham. Also, referring back to that Giants offensive line, Washington ranks 21st in Adjusted Sack Rate per Football Outsiders, so there's less of a threat than usual.

Running Backs

Below is the table for running backs:

Running Back Team Salary Overall Pros
Todd Gurley LAR 9800 33.1% 34.2%
James Conner PIT 7500 29.5% 33.5%
Kareem Hunt KC 7100 24.1% 23.8%
Isaiah Crowell NYJ 3700 14.9% 17.6%
Phillip Lindsay DEN 5200 14.8% 13.5%
Kerryon Johnson DET 5300 16.1% 13.4%
Jalen Richard OAK 4200 14.2% 13.0%
Marlon Mack IND 5400 12.5% 8.9%
David Johnson ARI 6700 11.1% 8.3%
Joe Mixon CIN 7300 13.3% 7.7%
Raheem Mostert SF 3800 6.0% 7.6%
Tarik Cohen CHI 5800 9.1% 6.6%
Saquon Barkley NYG 8800 9.1% 6.5%
Adrian Peterson WAS 4900 5.2% 5.1%
Chris Carson SEA 4300 5.1% 4.7%
Duke Johnson Jr CLE 3900 3.8% 4.3%
Jordan Howard CHI 4800 4.7% 4.2%
Christian McCaffrey CAR 7000 4.8% 4.2%
Nick Chubb CLE 4700 3.5% 4.2%
Trenton Cannon NYJ 3400 1.9% 3.9%
Alex Collins BAL 4600 2.3% 3.5%
Chris Thompson WAS 5500 3.2% 3.3%
Devontae Booker DEN 3700 2.5% 3.3%
Doug Martin OAK 4400 3.3% 3.0%
Alfred Morris SF 3500 1.5% 3.0%
Nyheim Hines IND 3800 3.3% 3.0%
Kyle Juszczyk SF 3400 0.8% 2.6%
Javorius Allen BAL 4000 1.3% 2.6%
Aaron Jones GB 4100 2.1% 2.4%
LeGarrette Blount DET 3500 0.8% 2.3%
Jamaal Williams GB 3600 1.7% 2.1%
Spencer Ware KC 4000 0.8% 1.9%
Ameer Abdullah DET 3000 0.8% 1.8%
Mark Walton CIN 3500 0.8% 1.7%

Early Impressions

It looks like a significant portion of pro GPP lineups will pair high-cost Todd Gurley with one of three low-cost options, Isaiah Crowell, Jalen Richard, or Raheem Mostert, all of whom figure to see increased usage due to an injury to one of their backfield mates.

SATURDAY UPDATE: Things have changed. Instead of pairing Gurley with one of Crowell, Richard, or Mostert, the pro projection model now suggests pairing James Conner with Crowell or Mostert. The shift towards Conner is odd given that, while his 4x value probability has increased from 20.1% to 20.8%, a negative indicator, Pittsburgh's wide receiver roster percentage, has gone in the wrong direction.

Isaiah Crowell ($3,700, 14.9%, 17.6%)

First and foremost, Crowell has the No. 1 likelihood of achieving 4x value probability among running backs (39.1%). Yes, Chicago surrenders the fifth-fewest points to opposing running backs, but Khalil Mack is on the wrong side of questionable in a game where resting him makes perfect sense, as the Bears are heavy home favorites. Elsewhere on the non-model side of the spectrum, rostering Crowell more than the overall public also makes sense based on the strategic knowledge handed down to us from Ryan Zamichieli's Sharp Report: Exploit situations in which a running back will see increased usage due to injury.

Raheem Mostert ($3,800, 6.0%, 7.6%)

Obviously, this is only a pro play if Matt Breida is inactive. Per current expectations, Mostert has the second-highest 4x value probability at the position (34.4%). Albeit on the road, he's facing a Cardinals defense allowing the most Draftkings points per game to opposing running backs. And then there's the completely subjective point this author should make in his role as 49ers recap writer for the site: Reviewing the coaches film suggests Mostert's performance the past two weeks is no fluke.

Tarik Cohen ($5,800, 9.1%, 6.6%)

This one may seem a bit contrarian, but Cohen only ranks in the middle of the pack at running back in terms of 4x value probability (14.7%, 18th). The Bears' wide receiver corps also ranks in the middle of the pack in terms of aggregate roster percentage (9.7%, 14th). As a heavy home favorite, this sure seems like a Jordan Howard game, who has a similar stat profile according to the model and is $1,000 cheaper in terms of salary.

Wide Receivers

Below is the table for wide receivers:

Wide Receiver Team Salary Overall Pros
Robert Woods LAR 6800 23.6% 32.5%
A.J. Green CIN 8000 19.3% 19.2%
T.Y. Hilton IND 6300 14.6% 12.9%
Antonio Brown PIT 8500 15.0% 12.3%
Jordy Nelson OAK 4700 14.1% 12.1%
Davante Adams GB 7900 14.0% 11.6%
Brandin Cooks LAR 6900 13.7% 11.5%
Golden Tate DET 6000 11.0% 10.3%
Tyler Boyd CIN 6700 12.7% 10.0%
JuJu Smith-Schuster PIT 7400 12.6% 9.5%
Tyreek Hill KC 8600 11.9% 9.2%
Martavis Bryant OAK 3700 9.1% 8.9%
Odell Beckham Jr NYG 7500 9.7% 8.2%
Sammy Watkins KC 4600 8.9% 8.2%
Emmanuel Sanders DEN 6500 9.5% 7.4%
Michael Crabtree BAL 4800 6.5% 6.9%
Doug Baldwin SEA 5500 7.1% 6.7%
Demaryius Thomas DEN 4900 6.6% 6.3%
Kenny Golladay DET 5700 6.9% 6.1%
Christian Kirk ARI 4500 5.3% 5.4%
Larry Fitzgerald ARI 4400 4.5% 5.2%
Chris Godwin TB 4500 4.5% 4.9%
Jermaine Kearse NYJ 4300 4.9% 4.9%
Marvin Jones Jr DET 4800 4.2% 4.7%
John Brown BAL 5900 4.2% 4.7%
Taylor Gabriel CHI 5100 4.5% 4.5%
Sterling Shepard NYG 5200 3.7% 4.4%
Josh Reynolds LAR 3600 4.1% 4.4%
Marquise Goodwin SF 5000 3.3% 4.4%
Courtland Sutton DEN 3800 3.7% 4.4%
Seth Roberts OAK 3400 3.0% 4.2%
Chester Rogers IND 4200 3.4% 4.2%
Willie Snead IV BAL 4200 2.3% 4.1%
Geronimo Allison GB 5000 5.1% 4.1%
Tyler Lockett SEA 5200 3.4% 4.0%
Devin Funchess CAR 5400 2.6% 3.9%
Mike Evans TB 7800 5.7% 3.8%
Randall Cobb GB 4900 5.3% 3.8%
D.J. Moore CAR 3600 1.4% 3.7%
Adam Humphries TB 3400 1.3% 3.5%
Allen Robinson CHI 5600 3.3% 3.5%
Andre Roberts NYJ 3100 1.4% 3.4%
Josh Doctson WAS 3700 1.0% 3.3%
DeSean Jackson TB 5300 3.3% 3.3%
David Moore SEA 3600 2.0% 3.0%
Anthony Miller CHI 3400 1.9% 2.8%
Jarvis Landry CLE 7200 4.6% 2.7%
Marquez Valdes-Scantling GB 4600 2.0% 2.6%
Damion Ratley CLE 3600 1.4% 2.0%
Antonio Callaway CLE 3800 1.4% 2.0%

Early Impressions

With a third-ranked probability of achieving 4x value (22.6%) and the best matchup among Rams' wide receivers according to Pro Football Focus, Robert Woods is projected to appear in significantly more pro lineups than those of the overall public. The same could be said for Robby Anderson, but it's looking increasingly unlikely that he's going to play on Sunday. Stay tuned.

SATURDAY UPDATE: Anderson is doubtful, so scratch him off the list. Woods is still the bee's knees, though.

Robert Woods ($6,800, 23.6%, 32.5%)

Woods' 4x value probability is now up to 25.5%. And, as was mentioned earlier with respect to Jared Goff, Green Bay's defense is now projected for a roster percentage of less than one percent. It's also noteworthy that Pro Football Focus has Woods enjoying the most favorable individual matchup among Rams wide receivers in Week 8.

A.J. Green ($8,000, 19.3%, 19.2%)

With Woods taking up so much of the oxygen at the position, almost all of the other wide receivers projected for high overall rostering show up technically as "fades" according to the pro projection model -- except Green. His 4x value probability isn't anything to write home about (16.4%, 21st), but of course said correspondence is unnecessary as he's already at home against a Buccaneers defense with a 16th-ranked 2.3% projected roster percentage. And according to Pro Football Focus, he has the most favorable individual matchup on the Sunday Main slate.

Tight Ends

Below is the table for tight ends:

Tight End Team Salary Overall Pros
Travis Kelce KC 6800 15.4% 16.6%
C.J. Uzomah CIN 3500 11.1% 13.1%
Jared Cook OAK 5000 10.4% 9.6%
Vance McDonald PIT 3700 8.5% 7.3%
David Njoku CLE 4600 10.4% 6.9%
Jimmy Graham GB 4700 5.3% 5.5%
Eric Ebron IND 4900 5.7% 5.1%
George Kittle SF 5500 5.4% 4.8%
O.J. Howard TB 3900 5.1% 4.8%
Ricky Seals-Jones ARI 3000 3.7% 4.8%
Trey Burton CHI 4800 4.3% 4.1%
Jordan Reed WAS 4500 4.0% 4.1%
Chris Herndon NYJ 2800 2.3% 3.9%
Jeff Heuerman DEN 2600 3.7% 3.6%
Evan Engram NYG 4300 3.5% 2.8%
Greg Olsen CAR 4200 2.9% 2.4%
Gerald Everett LAR 2600 1.2% 2.2%
Cameron Brate TB 2700 0.9% 2.0%

Early Impressions

Travis Kelce (15.7%) and Jared Cook (16.4%) have nearly identical 4x value probabilities, so it's interesting to see the model projecting a roster percentage that diverges even more for the pros than for the overall public. The culprit here is roster percentage for opposing defense, as the Colts' (4.5%) is double that of the Broncos (2.3%).

SATURDAY UPDATE: Thursday's situation holds. Kelce's and Cook's value probabilities are still nearly identical, and the latter's opponent defense is still projected to appear on double the proportion of rosters than the former's.

C.J. Uzomah ($3,500, 11.1%, 13.1%)

A multitude of indicators, both inside and outside the pro projection model point toward Uzomah this week. Among model factors, Uzomah has the No. 1 likelihood of achieving 4x value among tight ends (27.1%), and his opponent defense, Tampa Bay, has the fourth-lowest overall roster percentage projection at the position. Outside the model, the aforementioned Buccaneers rank dead last in Draftkings points allowed to tight ends. In addition, the Bengals-Buccaneers game has the second-highest over-under on the Sunday Main slate.

Defenses

Below is the table for defenses:

Defense Team Salary Overall Pros
Pittsburgh Steelers PIT 2300 25.0% 26.8%
Chicago Bears CHI 4100 8.7% 5.2%
Kansas City Chiefs KC 2600 7.5% 5.1%
Baltimore Ravens BAL 2800 6.5% 4.8%
Cincinnati Bengals CIN 2500 7.3% 4.7%
San Francisco 49ers SF 2500 5.0% 4.6%
Arizona Cardinals ARI 3200 5.8% 4.5%
Washington Redskins WAS 2700 4.3% 4.3%
Carolina Panthers CAR 2500 2.8% 3.9%
Detroit Lions DET 2400 3.1% 3.8%
New York Jets NYJ 2200 3.4% 3.8%
Los Angeles Rams LAR 2600 4.4% 3.7%
Seattle Seahawks SEA 2300 2.7% 3.7%
Indianapolis Colts IND 2900 2.8% 3.6%
New York Giants NYG 2400 1.5% 3.5%
Oakland Raiders OAK 2200 2.0% 3.4%
Tampa Bay Buccaneers TB 2100 2.3% 3.3%
Cleveland Browns CLE 2100 2.4% 3.2%
Denver Broncos DEN 2000 1.4% 2.9%

Early Impressions

With a whopping 53.9% chance of achieving 4x value and facing one of the lowest-rostered quarterbacks of the week, it will be hard for pros to pass up heavily rostering the Steelers, who -- oh by the way -- are playing at home coming off a bye.

SATURDAY UPDATE: Pittsburgh has further cemented its position as the bona fide chalk thanks to Baker Mayfield's projected roster percentage decreasing from 3.3% to 2.1% since Thursday.

Carolina Panthers ($2,500, 2.8%, 3.9%)

The Panthers may be on the road this week, but they're playing in a game that Vegas projects to be one of the lowest-scoring of the week. Based on their points projection and salary, Carolina has the fourth-highest likelihood of achieving 4x value, at 41.7%. They may not have as much upside as desired, but they have a high floor at low cost, which allows you to increase salary allocation at other, more valuable, positions.


More articles from Danny Tuccitto

See all

More articles on: Daily FF

See all

More articles on: DraftKings

See all

More articles on: Projections

See all