What is value in DFS? Ever wonder what the experts are referring to when they say "look for value at this position" or "this play has tremendous value"? This article is the first in the series titled "Daily Fantasy Sports 103" and will help define what value is and give you some simple tools to use when constructing your lineups.
We have started to post player scoring projections in the Statistics/Research section here on DraftStars. While this can be very valuable when constructing your DFS lineups, it is not the be all, end all. What sets the DFS "rookie" apart from the DFS "expert" is the ability to construct a lineup using projections with value. The easiest way to determine what value a player has is to divide his salary on a specific site with his projected score, thus the reason we provide the simple score projections. So if your projection model(or better yet ours)has Peyton Manning scoring 23 pts and costs $10,400 on FanDuel, his $/fantasy pt value is $452.17. The important thing to remember is that you need to establish some kind of watermark in order to determine what is an average value. You will have some players outperform this and some will under-perform. The key here is to minimize the standard deviation, or variance. Variance is the "kryptonite" of succeeding in DFS. You will hear over and over on this site that you need to eliminate variance, not only within one single lineup, but also among all of your lineups as a whole. If you have the bankroll to stomach 15 different lineups in 100 different contests, then have at it, but the truth is most of us don't. The topic of variance is discussed in the DFS 102 series. In summary, it isn't always about finding the cheapest guy when seeking value, although lower priced players tend to have more value potential. In the same way that you cannot simply fill your roster with all of the highest projected scorers because you will not have enough cap space, you also cannot stuff your roster full of value players as you will have far too much salary left over. The key is finding the balance between the two and devising a research tool to mesh them together.