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 11 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.
_--_QB.png)
Early Impressions
It's somewhat surprising to see that Carson Wentz is so disfavored by the pro projection model considering Saints-Eagles figures to be a shootout. Digging into the details reveals that Wentz ranks a mere 10th in the main positive factors (i.e., 4x value probability and aggregate wide receiver roster percentage), but ranks 5th in the main negative factor (i.e., projected roster percentage for the opposing defense). This situation is especially worth monitoring over the next couple of days.
SATURDAY UPDATE: Wentz's pro projection is even lower now than it was on Thursday, and the gap between it and the overall projection has widened. This is due to the Saints defense moving up from fifth to fourth in terms of projected rostership.
Cam Newton ($6,200, 14.8% Overall Roster Percentage, 16.1% Pro Roster Percentage)
At 45.5%, Newton has the highest probability of achieving 4x value among quarterbacks on the Sunday Main slate. The model also likes that he's facing a Lions defense projected for the fourth-lowest rostership overall (2.3%) and second-lowest among pros (2.1%). In terms of the matchup, Detroit is allowing 17% more fantasy points than league average to opposing quarterbacks over their past five games, which ranks fifth-highest.
Eli Manning ($5,200, 6.3%, 7.1%)
Manning is projected for 5.5 fewer Draftkings points than Newton (7.8 fewer in terms of upside). But with a salary $1,000 lower, his probability of achieving 4x value still ranks fourth among quarterbacks on the slate. Where Manning gets a significant boost to his pro projection is with respect to the aggregate roster percentage for his wide receivers: Odell Beckham (24.1%) and Sterling Shepard (8.1%) combine for 32.2% projected rostership, which ranks No. 1. As we'll see, stacking Giants appears to be the pro way to go in tournaments on Sunday.
Running Backs
Below is the table for running backs:
_--_RB.png)
Early Impressions
David Johnson sits atop the table due to a) having the seventh-ranked probability of achieving 4x value despite having the sixth-highest running back salary, and b) going against a Raiders defense that has the sixth-lowest projected overall rostership among defenses.
SATURDAY UPDATE: Johnson is now even more favored by the pro projection model for two reasons. First, the overall rostership projection for Oakland's defense has decreased from 3.2% to 2.0%. Second, the aggregate projection for Arizona's wide receivers has also dropped, from 12.0% to 9.6%.
Dion Lewis ($4,800, 22.6%, 24.9%)
It may be difficult to go to the well again after last week's disappointing performance, but Indianapolis has allowed the fifth-most Draftkings points to opposing running backs over the past five weeks. The Colts have been especially friendly to pass-catching backs, having given up 8 or more receptions in five of their nine games, including an 11-112-1 receiving stat line by Jacksonville's backfield just last week. In terms of model factors, Lewis' appeal is primarily due to his 27.8% probability of achieving 4x value, which ranks first among running backs in the Sunday Main slate.
Alex Collins ($4,400, 10.0%, 9.4%)
If you want to avoid Lewis' high rostership, Collins is a lower-rostered running back with a favorable matchup at a similar price point. The model does suggest having less exposure to him than the overall public, but the gap is much smaller than players around him in the table. The reason for this is because, at 26.8%, Collins ranks behind only Lewis in 4x value probability.
Although it's unfortunate that Collins isn't involved much in the passing game, the rushing matchup is about as good as it's ever going to get. This season, Cincinnati has allowed the second-most Draftkings points to opposing running backs. That ranking is the same if you adjust for opponent and focus only on the last five weeks. Baltimore is a six-point favorite at home against this defense, so Collins is likely to have every opportunity to get that 100-yard bonus, which makes up for the lack of receiving points.
Wide Receivers
Below is the table for wide receivers:
_--_WR.png)
Early Impressions
Michael Thomas is favored by the model because he's facing the only defense on the Sunday Main slate that's projected to appear in less than one percent of tournament lineups. It also helps that Philadelphia will be playing multiple backups at cornerback, which unsurprisingly translates to Thomas having one of the best individual matchups on the slate according to Pro Football Focus.
SATURDAY UPDATE: Thomas pro projection has increased slightly, but the gap between that and his overall projection has narrowed slightly. In terms of model factors, the only thing that's changed is Thomas' 4x value probability, which has increased a smidgeon, from 15.8% to 16.3%.
Odell Beckham ($8,400, 24.1%, 25.3%)
Despite similar salaries, Beckham actually ranks six spots higher than Thomas in 4x value probability (19.4%, 11th). Indeed, other than DeAndre Hopkins(24.2%, fifth), Beckham is the only high-salary wideout to rank in the Top 15. Other than that, one doesn't need to write a paragraph about the dumpster fire that is Tampa Bay's pass defense. Suffice it to say that the matchup is highly favorable, both at the team and individual level. It would not be unreasonable for around 10% of your lineups to include a Manning-Beckham stack.
Corey Davis ($5,600, 21.5%, 24.5%)
Davis has the third-highest probability of achieving 4x value among wide receivers (27.4%), and he's facing a Colts defense projected to appear in only 2.8% of lineups. Indianapolis' pass defense has fallen off a cliff recently as, although they rank 10th in Draftkings points per game allowed to wide receivers over the course of the full season, they're 26th in the past five weeks. And that's against -- shall we say -- unimpressive wide receiver corps in Jacksonville, Oakland, Buffalo, and the New York Jets. It's also worth noting that, at 39%, Davis has the seventh-highest market share of air yards in the NFL.
Amari Cooper ($5,400, 16.8%, 16.1%)
Like Alex Collins in the previous section, this is a situation where the model technically says to "fade" Cooper, but the overall/pro gap is small enough to look around for overriding factors. For instance, at 24.7% Cooper has the fourth-highest 4x value probability at the position. He also faces a Falcons defense projected to appear in only 3.3% of lineups. Speaking of his opponent, Atlanta has allowed the third-most Draftkings points per game to wide receivers, and they've allowed a fourth-most 18% more points than league average over the past five weeks. Even his individual matchup is favorable, as Pro Football Focus gives him a distinct advantage over Desmond Trufant. These indicators are more than enough to at least be as heavy on Cooper as the overall public. Oh, and one final note. Only Julio Jones (47%) has a higher market share of air yards than Cooper's (43%) since trade to the Cowboys.
Tight Ends
Below is the table for tight ends:
_--_TE.png)
Early Impressions
Zach Ertz is the overall chalk and gets a heavy boost in the pro projection model. This is almost entirely due to having a sixth-ranked probability of achieving 4x value despite being the most expensive tight end on the Sunday Main slate. One statistical detail that's interesting is how Ertz' matchup looks different depending on whether you look at season-long raw data or adjusted data over the past 5 games (i.e., Austin Lee's Normalized Strength of Schedule). According to the former, New Orleans has allowed the fifth-fewest Draftkings points to opposing tight ends this season. According to the latter, however, they've performed at exactly the league average.
SATURDAY UPDATE: Ertz is now even more chalky than he was on Thursday. This is despite the fact that the projected roster percentage for New Orleans' defense has actually increased from 7.2% to 8.2% since Thursday.
Ricky Seals-Jones ($2,900, 12.8%, 13.7%)
Seals-Jones' pro projection has increased significantly since Thursday due to his own overall roster projection, as well as the aforementioned drop in the roster projection for Oakland's defense. Couple that with Seals-Jones' top-ranked 28.9% probability of achieving 4x value, and you've found yourself a considerably cheaper tight end if you choose to pivot away from Ertz. And unlike Ertz, this is an unequivocally positive matchup, as the Raiders a) have allowed 24% more fantasy points to tight ends than the league average over the past five weeks, and b) have allowed the fourth-most points over the course of the full season.
Defenses
Below is the table for defenses:
_--_DST.png)
Early Impressions
Arizona and Pittsburgh form a clear top tier of roster percentages. The reason their projections are higher for pros than for the overall public is because they rank third-lowest and sixth-lowest, respectively, among defenses in terms of expected rostership for the other team's quarterback. If Derek Carr and Blake Bortles don't warrant high rostering, then the Cardinals and Steelers should be rostered even more than you'd think.
SATURDAY UPDATE: The Cardinals are favored even more by the model, but now the Steelers are not. This is because Bortles has gone from 2.4% projected rostership to 2.7%.
Houston Texans ($2,600, 6.2%, 6.3%)
If you're looking to pivot away from Arizona for some salary relief, the model favors Houston as they have the third-highest 4x value probability (46.1%) and are facing a quarterback projected to appear in only 1.1% of tournament entries. The matchup against a Washington offense decimated by injuries is favorable as well. Alex Smith has scored 30% below league average over the past five weeks, which is only higher than Josh Rosen and whatever guy off the street Buffalo has decided to start that game. Furthermore, with the Washington offensive line in shambles, Pro Football Focus gives Houston's front seven one of the largest matchup advantages of the week. The only issue with rostering the Texans heavily is the danger that habitually risk-averse Smith won't provide many opportunities for interceptions.