Projecting the Pros on DraftKings: Week 9

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 9 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.

Early Impressions

It's odd to see Cam Newton and Ryan Fitzpatrick, who are facing each other in what figures to be a shootout, having such divergent pro projections. In addition to being in the same game, both have seen their salaries increase by $800 or more this week, and yet the still rank No. 1 and No. 2 among quarterbacks this week with respect to likelihood of achieving 4x value.

SATURDAY UPDATE: They've diverged even further despite the salary and value probability situation being the same. What's changed is the massive shift in the aggregate projected roster percentage for their wide receivers. Newton's has increased from 17.2% on Thursday to 30.2% on Saturday thanks largely to D.J. Moore, who's risen from 6.4% to 16.3%. In contrast, Fitzpatrick's wide receiver corps has dropped from 25.5% to 20.4%. As always, when the public is going gaga over a team's wide receivers, they tend not to increase their rostering of that team's quarterback enough (and vice versa).

Here are a couple of other quarterbacks worth discussing. The one thing they have in common is that a mid-week trade consolidated/clarified wide receiver target distribution, and therefore roster percentage.

Matthew Stafford ($5,600, 1.6% oVERALL rOSTER pERCENTAGE, 3.1% pRO rOSTER pERCENTAGE)

At 27.3%, Stafford's wide receiver corps -- now minus Golden Tate -- has the fifth-highest aggregate roster percentage among quarterbacks on the Sunday Main slate. According to the other two metrics that go into the pro projection model (i.e., 4x value probability and roster percentage for the opposing defense), he ranks in the middle of the pack. But still, whether it's Ryan Zamichieli's The Sharp Report or Scott Bischoff's Starting Stacks, when the stats side of our DFS content comports with the non-stats side of our DFS content, it's usually a sign that we're onto something.

Case Keenum ($4,900, 3.0%, 4.4%)

Like Stafford, the only thing that stands out in terms of model factors for Keenum is the aggregate projected roster percentage for his wide receivers, mostly thanks to Courtland Sutton. The bottom line of this thin-value call is that, if Denver traded Demaryius Thomas because of Sutton, and if Sutton achieves value, then Keenum will almost certainly achieve value as well, which unlocks all kinds of possibilities in the rest of your roster. Attaching Keenum to Sutton means you're able to offset the latter's chalkiness, while also having $41,200 remaining towards the salary cap.

Running Backs

Below is the table for running backs:

Early Impressions

This is the largest discrepancy of the season between overall and pro projections regarding Todd Gurley. Isaiah Crowell looks like an intriguing option at a cheaper salary. Of course, "fool me twice..."

SATURDAY UPDATE: The discrepancy for Gurley has decreased from 6.7% to 4.0%, which is more in line with previous weeks. Crowell, however, remains intriguing.

Austin Ekeler ($4,300, 3.2%, 4.1%)

Melvin Gordon III appears likely to play, but him coming off an injury absence means Ekeler's likely to get more than his usual 35% snap share. And if Gordon happens to aggravate said injury, which isn't unheard of in the context of hamstrings, Ekeler will get even more than that.

Over their past three games, Seattle's pass defense has allowed an average of 305 passing yards to competent (i.e., non-Raiders) offenses, including 10.7 Draftkings points allowed per game to running backs via receiving.

In terms of the pro projection model, the main factor elevating Ekeler is his seventh-ranked likelihood of achieving 4x value (21.6%).

Kapri Bibbs ($3,500, 2.1%, 3.7%)

Like Ekeler, Bibbs is a complementary running back with an unusually high-ranking value probability: 20.9%, ranked 11th.

Unlike Ekeler, there's no doubt Bibbs will have free reign over his offense's backfield receiver role, what with Chris Thompson already declared out for the game. And of course, it's at this point common knowledge -- mainly via Ryan Hester's years-long crusade on the Power Grid -- that a receiving back against Atlanta is the set-it-and-forget-it matchup to end all set-it-and-forget-it matchups.

Wide Receivers

Below is the table for wide receivers:

Early Impressions

With Demaryius Thomas traded to Houston after Draftkings salaries were set in stone for Week 9, Courtland Sutton is the no-brainer chalk of the week. This kind of value situation doesn't come up often at wide receiver, so it's imperative to take advantage -- even if it doesn't work out. That said, one imagines that pro projection will come down somewhat once the model is updated on Saturday.

Tight Ends

Below is the table for tight ends:

Early Impressions

Since he has never been featured in this space and that's likely to continue through Saturday's update, it's noteworthy that the tight end who ranks No. 1 in 4x value probability on the Sunday Main slate is -- wait for it -- Kyle Rudolph (22.4%)?

SATURDAY UPDATE: Whaddyaknow! Rudolph is still No. 1 at 22.4%!

Chris Herndon ($3,000, 2.3%, 3.7%)

At 19.3%, Herndon ranks second in 4x value probability this week.

The Dolphins defense may rank a middling ninth on the Sunday Main slate in terms of Draftkings points allowed to opposing tight ends, but they've allowed five touchdowns in their past three games. Furthermore, according to Pro Football Focus, even Neal Sterling (aka not the more-talented, more-effective Herndon) has a matchup advantage against safety Reshad Jones.

Granted, with both of their starting wide receivers almost certainly inactive, the Jets will likely rely on the running game. But there will be passes. Someone has to catch those passes, and Jermaine Kearse hasn't shown an ability to step into that void the past two weeks.

Defenses

Below is the table for defenses:

Early Impressions

Here's a interesting model stat: At 43.3%, the Bills likelihood of achieving 4x value is actually higher than that of the Bears (36.8%). #Discount

SATURDAY UPDATE: Although both have dipped slightly, the comparison still holds: The Bills are now at 42.4%, while the Bears are at 34.6%.

Washington Redskins ($2,600, 2.6%, 3.2%)

Or Atlanta Falcons ($2,200, 1.0%, 2.9%)

This game involves

And yet, the over-under is currently 48? If there's any game on the Sunday Main slate that's likely to go contrary to the Vegas total, it's this one.

In terms of model factors, being that they're slightly cheaper, Atlanta (35.1%) has a slightly higher 4x value probability than Washington (33.1%). Where it becomes distinctly advantage-Falcons is with respect to projected overall roster percentage for opposing quarterbacks, as Matt Ryan's 3.4% is significantly higher than Alex Smith's 2.0%.