Okay, so previously we’ve established that batter versus pitcher and streaks are non-predictive and thus a faulty means to construct a daily fantasy lineup. That begs the question, “what should be done?”

Approaching things from the obvious perspective that in salary cap games the objective is to pay for as many potential points as possible, it follows the task at hand is to identify players most likely to realize a positive return on investment, relative to their price point. In a nutshell, there are two general approaches.

SALARY SET BELOW TALENT LEVEL

The various daily sites have proprietary algorithms they use to generate the player’s salaries. The inputs include their current as well as yearly performance. It stands to reason that if they are currently slumping and/or have underperformed to date, their salary would be depressed.

Circling back to the data that shows slumps are non-predictive, assuming the skills of the player have not changed (more on that in a minute) then by extension the downward adjustment puts the salary below what the player is actually worth. This makes him a value play since all he has to do is what he’s supposed to do to earn a profit.

The catch here is whether the skills have indeed sustained. How do we know if a player is in a slump or if he’s not as skilled as originally assumed coming into the season? I’ll save the details for another column more apropos, but there are studies that determine the number of plate appearances and innings (actually batters faced) necessary for skills to stabilize in-season. For example, say a batter is fanning at an elevated rate. After a specific number of at bats (I’m guessing not nearly as many as you may believe) the contact rate has a 50 percent chance of being real. Each skill has its own level. Have no fear, the newly introduced Fantasy Alarm projection tool will adjust baseline expectations each week using this stabilization data affording you the best chance to identify players with improving or declining skills.

I know, this doesn’t really answer the question – how can you tell when a skill change is real. The best course of action is to assume players with three years of experience have an established skill level or at least won’t vary too much from what was expected coming into the season. At minimum, the number of veteran players with a true skills decline is minimal so it’s safe to assume slumping players have a positive return-on-investment built into their salary and make for viable daily players, so long as they meet some of the other criteria soon to be discussed.

As an aside, I’m currently working on a means to generate a true value salary for each player per site. Think of it as if I run my own daily site and want to figure out how much each player is worth so I know where to begin setting salaries. Truth be told, I wouldn’t adjust based on current performance but that’s neither here nor there. The point is, down the road you’ll have a baseline salary for each player based on their season-long projection. All you’ll need to do is compare that day’s price with what it should be to find candidates to supply a positive return on investment.

SALARY AT OR ABOVE TALENT LEVEL

The implication here is the player is producing at his expected rate or maybe a little better. Keep in mind that price is based on numbers to date which lump everything together. There are manners those numbers car be parsed into production better and worse than the total production. The key to identifying players with a possible positive return on investment is to pinpoint players in a situation where their expected performance exceeds that of their average. Yeah, this is just a long-winded way to say to use splits.

Veteran DFS players know where this is going but the four areas which can be gamed the most are

  • Handedness
  • Home/Away
  • Park
  • Quality of opposition

We’ll handle each of these in depth down the line, but here’s the Cliff Note’s version. Hmm, do Cliff Note’s still exist?

Handedness – LHB v RHP enjoy the biggest advantage followed by RHB v LHP, RHB v RHP and LHB v LHP.

Home/Away – The baseline skills of a player are on average 10percent better at home

Park – No real explanation necessary other than to say while park effects are important, they are secondary to

Quality of Opposition – While there are some players that rise to the occasion against top competition, you want hitters squaring off against lesser hurlers and pitchers facing weak lineups.

Next time we’ll take a closer look at handedness and how it alters the baseline expectation.