The dynamic game of Daily Fantasy Sports (DFS) requires much more than simply knowing the sport for which we're entering contests to be successful. We must be adaptable, precise, and open to learning from previous endeavors, the latter of which will be the primary focus of this weekly written piece. Game Theoretic methodologies will allow us to analyze and dissect the previous week's winner of the largest and most prestigious Guaranteed Prize Pool (GPP) tournament on DraftKings – the Millionaire Maker. These same tenets of Game Theory, which can most simply be explained as the development of decision-making processes given our own skill and knowledge, assumptions of the field based on the cumulative skill and knowledge of others playing the same game, and the rules and structure of the game itself, will allow us to further train our minds to see beyond the antiquated techniques of roster building being employed by a large portion of the field. Approaching improvement through these methods will give us insight into the anatomy of successful rosters and will help us develop repeatably profitable habit patterns for the coming weeks. We'll start by looking at the previous week's winning roster, extract any pertinent lessons for future utilization, and finish with a look ahead towards the coming main slate.
I did a study of both the winning Milly Maker rosters and the corresponding optimal rosters from 2021 before the season started, and the findings were startling. From that study, I found that the winning roster scored an average of 40.29 less fantasy points than the optimal roster last season, which describes an input into our decision-making matrix regarding optimal practices. As in, we aren't playing the game of NFL DFS against a computer, we're playing against other humans, so we don't need optimal. Furthermore, the roster construction that appeared most frequently on both the list of winning rosters and the list of optimal rosters was an uncorrelated skinny, or a quarterback paired with a single pass-catcher and no bring-back from the other side of the game. That construction occurred a whopping seven of 18 times (38.9%) on the optimal and five of 18 times (27.8%) on the winning roster, yet it is utilized on less than 10% of rosters in play on a standard week.
Now, those findings would be far different in small field, single-entry, and three-max contests, where the delta in scoring between the winning roster and the optimal roster is much larger – but in a contest like the Milly Maker (and other large field mass multi-entry contests), the winning score creeps closer to optimal with so many rosters in play. And since we must adopt a payoff dominance mindset in these contests due to the top-heavy nature of their payouts, utilizing a roster construction that is proven to hit at a higher rate on the optimal builds than both the fields' usage and of the winning rosters generates excellent leverage from a macro perspective. DraftKings user amhunt2178 executed this idea through the utilization of Geno Smith and Tyler Lockett, who combined to put up almost a six times salary multiplier on $12,100 combined salary.
We've talked about this before in this space, but the Raiders have been the most concentrated offense in the league this season, with over 60% of their fantasy production flowing through just two players – Davante Adams and Josh Jacobs. As such, we can consider a rule of “at least one of Davante Adams or Josh Jacobs on every roster” until the field either adjusts to their production or their ownership begins to reflect the production we've seen to this point in the season. Consider this – either Adams or Jacobs has posted 30 or more DraftKings points in nine of 12 games this season, with one of those games being the “Adams flu game” where the Raiders got shut out by the Saints. As in, 81.2% (9 of 11) of the games in which both have been healthy this season have yielded a 30+ point fantasy score between the two. That's absurd.
The Chalk vs. Roster Construction
First off, I propose that the field is approaching Nash equilibrium in an important aspect of NFL DFS – player selection. Nash equilibrium describes an environment where no agent (or player) can generate additional edge (expected value) through the altering of their decision matrix for a given data set. As in, we are approaching a point where it isn't enough to simply play the right players because the field is too good at identifying top on-paper plays (which is amplified by recent technological advances like optimizers, projections, and range of outcomes predictors). That said, I have moved away from describing chalk as either “good” or “bad,” and instead describe it for how it interacts with the remainder of the roster in play. That's an important distinction considering two players on the winning roster this week held players at 27.7% ownership or more. But because they were solid on-paper plays, it's almost doing a disservice to write off highly owned plays entirely, meaning we must find other ways to generate leverage in the changing DFS landscape. Enter roster construction. The field has transitioned to searching for certainty at the running back position over the second half of the season, meaning typically we see ownership congregate on the players priced above $6,000. DraftKings user amhunt2178 altered their salary allocation via roster construction by utilizing two running backs priced in the lower tier of player pricing, fundamentally shrinking the amount of the field they were competing with that also rostered Garrett Wilson and Amon-Ra St. Brown.
Tyler Huntley + Mark Andrews (Uncorrelated Skinny and Hyper Concentration)
Lamar Jackson suffered a knee sprain in Week 13 and should be considered questionable-to-doubtful to play against the Steelers this coming weekend, leaving us with a potential opportunity to play an underpriced mobile quarterback with his primary pass-catcher at a suppressed price. Due to the low efficiency of the opposing offense (Pittsburgh Steelers) and the previously discussed theoretical leverage angle, we need not be concerned with forcing a correlated bring-back here.
Jerry Jeudy (Hyper Concentration)
KJ Hamler was placed on injured reserve prior to Week 12 and Courtland Sutton left early with a hamstring injury. Should Sutton miss Week 13, expect Denver's pass offense to be extremely concentrated amongst Jeudy, tight end Greg Dulcich, and journeyman Kendall Hinton. Both Jeudy and Sutton have performed best when the other has been off the field over the previous two seasons, making Jeudy and interesting bet in a plus matchup against the Chiefs at a suppressed price.
D'Onta Foreman + D'Andre Swift (The Chalk vs. Roster Construction)
Similar to last week, an early look at the slate ahead reveals that we are likelist to see rosters prioritize certainty at the running back position, meaning solid leverage can be gained by looking for players in the lower tier of player pricing that have paths to viable GPP scores through both workload, matchup, and talent. Foreman gets the 31st-ranked run defense of the Seahawks while Swift continues to perform well, regardless of workload, when not on the team's injury report. The lack of current value available at the running back position in Week 13 makes this pairing highly intriguing.
Garrett Wilson (The Chalk vs. Roster Construction)
Garrett Wilson's fantasy production without Zach Wilson as the starting quarterback has been nothing short of spectacular throughout his rookie season. Furthermore, his elite targets per route run rate and team target market share without Zach Wilson has left him with a clear path to elite volume in any game environment that we can confidently expect an increase to team pass volume, and I can think of no better opponent for that to be the case than in a game against the Bills. Wilson is priced entirely too low for his volume in games played without Zach Wilson at quarterback.