DraftKings Cash Game Strategy

Excerpt from the Footballguys' Cracking DraftKings Book.

Failing to appreciate the differences between a cash-game roster and a tournament roster can be an expensive lesson for the beginner. To be clear, it is folly to construct DFS lineups for cash games in the same manner as tournament lineups.

To help you circumvent that lengthy (and costly) learning curve, the next few paragraphs will describe the factors and strategies that should guide a daily fantasy player’s thinking process when building each type of lineup. There is an art to constructing DFS lineups.

In daily fantasy circles, the idea of value is perhaps the most frequently discussed topic. A beginner will put together a roster that he thinks will score a lot of points; a veteran knows approximately how many points he needs to win and uses value as a means to determine exactly whom to select for his lineup. But what exactly is value and why is it important?

To underscore the importance of value, let’s first discuss its presence in our everyday lives. With our incomes, we seek out value in the material things and services that we purchase. If there are two gas stations next to one another, do you compare prices and give your business to the cheaper of the two? When considering cable options, do you price compare and try to negotiate a better deal between competitors? When purchasing a car, do you try to haggle the asking price to a lower level before agreeing to a sale?

The common denominator in all of these situations is that you, the buyer, are attempting to maximize the value of your dollar…much like an investor tries to do on Wall Street by buying low and selling high. The value that one achieves in these examples is known as return on investment, or ROI. The ROI in daily fantasy is measured in points per dollar and is the primary basis for decision-making during roster construction.

The term “value” is, therefore, a qualitative measure used to describe the number of expected points per dollar that a player can potentially return in any given week. There are two components to value: first, site-specific salary, and second, projected points in the site-specific scoring system.

The first component of that formula (salary) will become available on DraftKings on Sunday evening. Deriving the second component (projected points) will be discussed in detail in a later section (Section IV.D – Using Projections) of this book. If you are not so inclined to generate your own projections, Footballguys provides all subscribers detailed projections by the Wednesday of every week during the season, which dramatically lessens the workload for an otherwise busy fantasy player.

With these two pieces of information, one simply has to divide the player’s salary by his projected points to determine that player’s points per dollar. Many successful DFS managers build an Excel spreadsheet to sort all players by their values each week.

Once value is determined on a point-per-dollar basis, it becomes fairly simple to sort this column from lowest to highest. Viewing this list, even the most inexperienced DFS players can identify the best values for the upcoming week. If you have never opened an Excel spreadsheet, if your personal and professional life cannot permit such activities, or if you just cannot be bothered to spend the time to put together such a tool…remember that the Interactive Value Charts on Footballguys.com accomplish this very task by Tuesday of every week, thereby eliminating the need to perform this weekly activity on your own!

With value determined on a point-per-dollar basis, it is now possible to begin working backwards to construct a solid cash-game roster. The first rule on DraftKings is to aim for 150 points to win your cash games. A retrospective analysis of DraftKings cash games in 2014 demonstrated that a score of 150 or above will win cash games approximately 90% of the time. With a $50K salary cap on DraftKings, it becomes a simple math exercise to recognize that a winning DFS player must achieve 3 points for every $1,000 spent on player salary.

Thus, when selecting potential players for a cash game roster, the process should include dividing that player’s salary by 1,000 and subsequently multiplying by 3 to determine the number of fantasy points he needs to score to reach value. For example, if Drew Brees’ salary is $9,800, he would need to score 29.4 fantasy points in order to justify his salary for cash games ($9,800/1,000 = 9.8 x 3 = 29.4).

Now that the scoring goal has been set, strategic player selection becomes the next objective. There are two main considerations for optimal cash-game player selection: (1) a history of low variance, (2) a high floor. A history of low variance can be more succinctly defined as consistency. For cash games, we require a consistent, reliable performance from week to week. Starting a player who is consistent will ensure that your roster will not suffer from a one-catch, six-yard performance from your wide receiver or a 9-carry, 18-yard performance from your running back. Cash games are all-or-nothing propositions with tightly clustered scores in the middle of the final standings. Suffering a poor game at any position is often the difference between winning and losing.

It is therefore imperative to roster players who are heavily involved in their offensive schemes to ensure consistency of scoring. Logically, for wide receivers and tight ends, attention should be given to highly targeted receivers (for example, Antonio Brown in 2014). For running backs, consider players who routinely touch the ball 20 or more times out of the backfield (bonus points for pass-catching backs like David Johnson). For quarterbacks, use passers who will throw the ball often, regardless of the game plan (for example, Drew Brees and Andrew Luck in 2014).

The second component of selecting solid cash game players is to choose those individuals with a high floor. A player’s floor is defined as the minimum number of fantasy points realistically projected to be scored in the upcoming game, barring unforeseen events (namely an injury). Not surprisingly, this parameter is closely related to the first rule, which stated that a cash game player should exhibit low variance from week to week. A player who is heavily involved in his team’s offense should score fantasy points through sheer volume of opportunity.

When touchdowns are so heavily weighted in the scoring system, yet also unpredictable, it is essential to roster players whose receptions and yardage can be loosely predicted based on their usage and importance to their team’s overall offensive scheme. As a general rule, it is advisable to assume that a player will not score a touchdown. This will yield an absolute basement number for that player’s floor that week. Once the floor has been determined for each player at a given position, particularly in the context of value, the number of viable options for cash games becomes quite limited.

Before moving onto other considerations for cash games, it is important to stress that you must exercise simple common sense when employing the aforementioned strategies. For example, if a WR2 has been heavily targeted for several consecutive games while the WR1 for that same team has been injured, do not expect that trend to continue when the WR1 returns from injury. Similarly, if a top-tier quarterback goes down to injury and is replaced by a rookie quarterback, everybody in the offense should be significantly downgraded to reflect the loss of leadership, experience, and overall ability on the field.

On the opposite end of the spectrum, do not be afraid to roster an inexpensive backup player at a given position if the primary player at that position is announced as inactive; the caveat is that you must have confidence that the replacement player will still see significant action in the new role. There are many potential mistakes that novice DFS players can make, but consulting with a reliable set of projections can often help avoid poor decisions.

One of the intriguing aspects of playing NFL DFS is that there are always a significant number of injuries from week to week that affect value at each position. For this reason, it is essential that a player first capitalizes on mispriced players due to injury or site-specific pricing mistakes (it happens from time to time). After those extreme value players find their way onto your roster, you should be looking to incorporate further value with consistent players who project to have high floors for that weekend.

All things being equal, the position that consistently demonstrates the least amount of variance is the quarterback position. Why? Because they are the only players who throw the ball in the passing game, whereas running backs tend to share rushing duties with other running backs, and receivers are subject to similar restrictions (matchups, sharing targets, etc.). Thus, it often makes a lot of sense to spend a substantial amount of your salary (after first earmarking extreme value as described above) on the quarterback position because it is the position where the expected ROI is most likely to occur on a consistent basis. That said, it is completely inadvisable to spend so much at the quarterback position that you leave too little salary to strategically fill the remaining roster slots; one should therefore exercise caution to ensure that the selection of any single player is not too expensive such that it prohibits rostering other likely productive players.

Of all the required DraftKings positions, defense remains the least predictable from week to week. In Section IV, we describe a retrospective-based system to help reasonably project fantasy output at those positions, but those projections are still limited by the highly variant scoring nature of those roster positions. In other words, it is advisable to construct your cash game rosters with skill positions first and add the defense as an ancillary component thereafter.

An often-overlooked strategic parameter of cash game roster construction is the avoidance of negatively correlated players. A negatively correlated player is one whose on-field production reduces the fantasy output of another player on the same team. For example, if Melvin Gordon has an excellent day rushing the ball for San Diego, it is highly unlikely that Philip Rivers will also have a plus day in the passing game (or at the very least, achieve value based on his hefty salary). In cash games, where consistency and guaranteed points (floor) are valued, it rarely is logical to roster a quarterback and a running back from the same team because those positions are often negatively correlated.

Positively correlated plays (e.g., a QB-WR combination) are generally attributed to tournament rosters (see the following section), but there is a positive correlation play or two that can be employed for cash games, the first of which is pairing a running back and team defense. It is common practice in the NFL for teams to build an early lead and lean on their workhorse running back in the second half to “run the clock” so as to force the trailing team into desperation mode late in the game. When losing teams are forced into becoming one-dimensional, they often make mistakes that lead to sacks, turnovers, and defensive touchdowns, all of which benefit the fantasy prospects of the opposing defense. For these reasons, if you feel strongly about a running back playing for a heavy favorite, stacking a running back and team defense is a solid positive-correlation play for cash games and tournaments alike.