As was discussed in the introduction to this series of articles about creating your own MLB DFS Process, the first criteria must always be what type of contest you are entering. I won’t spend time going into the differences between Cash games and GPP games again here except to say that you MUST know what kind of game you entering to figure out what kind of lineup you want to build. The remainder of this article will be devoted to the factors you should consider when building a Cash game lineup.
You will remember that in a Cash game lineup you are attempting to build a lineup with the highest floor possible. For example, if you have a choice between two players, Player A and Player B, where Player A has gone 2 for 5 every game for the last 10 games with 20 singles, and Player B who went 5 for 5 in four games and 0-5 in six games with a total of 20 singles over the same 10 game span. Which player do you want to select for your cash game? The answer is Player A because he has provided 2 hits every game for the last 10 games and is less likely to put up zero points in the next game while Player B has had four amazing games going 5-5 out of the last 10 games, but has also had six games in which he scored zero points. So while Player B does have the higher upside or ceiling, Player A has the superior floor and thus should be rostered over Player B in your Cash game lineup. Now that we have identified what kind of player we are looking for, let’s talk about how we find them.
The first place I like to start in my research for hitters is the Vegas Over/Under lines for each game. If you have followed the Process flow chart you should already have some awareness as to what pitchers Vegas likes and should make an attempt to rule out hitters playing against the pitchers you considered rostering for your own lineup. The opposite is also true. The pitchers that you definitely did not want to roster in your lineup now become targets. This step should take no longer than 5-10 minutes to quickly review the Vegas Over/Under lines since you probably already spent time looking at them while selecting your pitcher. At this point we are just generally getting a sense of what pitchers are going to be soft targets for hitters. Do not roster any hitters facing your pitcher.
Splits are an important part of DFS MLB. There are several different types of splits for DFSers to consider when building lineups. These splits help us gain further insight into a given player’s tendencies and also help us understand how to better interpret the player statistics. The split that is most commonly used for DFS is the Righty/Lefty splits. In general, most hitters will hit better against the opposite handed pitcher. For example, Josh Donaldson is a great player and one that gives both right-handed and left-handed pitchers difficulties. That said, he famously CRUSHES left-handers to the extent that if you were going to roster him and pay his typically exorbitant salary, you would want to do so on a night when he is facing a lefty because as a rule he hits better against lefties than he does against righties. Most of the stats that will follow should be viewed through the lens of R/L Splits because doing so helps us get a much clearer picture of what the hitter is most likely to do when you roster him today. Another split you may want to consider is Home/Away splits that show how well a player hits in his home stadium compared to how they hit on the road. Some players, either because of comfort or familiarity or maybe sleeping in their own bed, hit better at home than they do on the road. Others, as you might imagine, are just the opposite and hit better on the road than they do at home, particularly when their home stadium is a pitchers park. Batter vs. Pitcher Splits are the last split that will be covered in this article. BvP is just that, the actual history of a batter versus the pitcher he is facing today. There is much controversy and consternation in the DFS community surrounding the usefulness of BvP, or lack thereof. As such, it will be covered on its own towards the end of the article.
Once you know the pitchers you would like to target and how to use splits to gain insight into hitters’ statistics, now it’s time to actually dig into the statistics. The first step when building a cash game lineup should be wOBA (weighted On-Base Average). wOBA is the statistical attempt to account for all of the different types of offensive statistics and weight them according to their actual worth. Although no one statistic can tell you everything you need to know, at this point in time, wOBA is widely regarded as the best single statistical indicator for selecting hitters, especially in cash games where you want to maximize the floor in your lineups. When looking at this statistic it is important to have some benchmarks to help you understand what you are looking at when reviewing wOBA. I often times hear DFSers say that a player with a .320 wOBA is a good play on a given night. It is true that .320 wOBA is generally a solid number but the real value is in understanding wOBA on a positional basis. In the example above a Shortstop with a wOBA of .320 would be .16 points above the positional average and would be a good option for your cash lineup, but a First Baseman with the same .320 wOBA would be .13 points below the positional average and would likely not be a good option for your DFS cash lineup. Without knowing the context of the positional benchmarks for wOBA the value of the statistic is impaired. The table below was created by Fangraphs.com and provides positional bench marks (as of 2014) to help you better understand how to interpret wOBA when you are building your lineups.
ISO measures the amount of extra bases per at bat. It can be quickly calculated as follows: OPS-AVG=ISO. While ISO is not typically thought of as a cash game tool it can be used in concert with wOBA to get a better idea of what kind of player you are looking at. While cash games are typically about building the highest floor possible it is still important to know what a given player’s upside is as well. All else being equal, if you are choosing between two players with equal floors, you want the player with the biggest upside on your team so that if he does have a big night, he can help carry your team. ISO should be used more as a tie breaker between two players that you like similarly.
Batting Average on Balls in Play or BABIP is very useful in figuring out how lucky or unlucky a given hitter has been. Players usually will play to their career average for BABIP over the course of a season and most players will play to around a .300 BABIP. If you can compare their career BABIP against their current batting average you can quickly tell if they have been unlucky (significantly below .300) or lucky (significantly above .300). If a player has a BABIP significantly below .300 you may want to roster him as he is due for a correction or regression to the mean of .300. Early in the season this is not as useful as it will be later in the season because of the small sample size but it is another factor to consider if you like two players equally and are having a difficult time deciding which on to roster.
Batted Ball Profiles are the percentages of ground balls, line drives and fly balls that a given player has put into play vs. the total amount of balls he has put into play. Historically the benchmarks are as follows: 45% ground balls, 35% fly balls, and 20% line drives. As you might have guessed, line drive % has by far and away the highest correlation to runs scored followed by fly ball % and then ground ball %. If you are considering two like players with same or similar wOBA and Player A has a GB% that is 55% while Player B is right on the league average, it is an indication that Player A has been getting luckier in that he has been achieving the same results despite a high and unfavorable GB%. In this case Player B would be the player you should roster because his results are less reliant on luck. LD% is a very clear indicator of just how well a given player is hitting the ball. If a player has a less than optimal wOBA for a given position but has a very high LD% he is due to have a run of good luck given how well he is hitting the ball (high LD%) and would be an ideal candidate for a DFS lineup.
We have already discussed the value of rostering consistent hitters while playing cash games, but the search for the optimal player does not end there. Much like ISO, HR% helps us determine what kind of a hitter we are looking at. While we are not searching for players with the highest upside when building cash lineups, we certainly are not averse to upside! If, for example, after considering the factors above you determine that two players are similarly suited for your cash lineup but you can only roster one, HR% might provide some clarity as to which one you should select. The player with the higher career HR%, all else being equal, should be the player you are more likely to roster because he has greater upside and thus improves your chances of winning. HR% is one of the most consistent statistics year over year and thus can be relied on more heavily than other statistics such as batting average year over year or on-base percentage. An over simplified explanation goes as follows, the players who hit a lot of homeruns last year a likely going to be the same players who hit a lot of homeruns this year and thus can be relied on to produce as such. Again, this is not a standalone statistic but when taken into account with the rest of your research can provide good insight into your roster building lineup.
Batter vs. Pitcher data, or BvP, is perhaps one of the most hotly contested statistics in the community of MLB DFS. Supporters of BvP, and often times newer players, argue that the best way to figure how a batter is going to perform against a given pitcher is to look at how he has ACTUALLY performed against this pitcher in the past. Seems like a reasonable statement, right? Well, the problem with BvP is that the sample size is simply not sufficient for us to draw any really meaningful conclusions from BvP alone. Just because a player is 5 for 10 with 2 HRs against a given pitcher is not an indicator in and of itself that he will produce the next time he faces that pitcher. In fact, it could be an indicator of just the opposite given that the league average for BABIP is .300 and this player is currently hitting at a clip of .500 against this pitcher.
That said, I believe that it is foolish to completely ignore BvP. Some batters simply perform better against specific pitchers. Let me illustrate this with an example. Felix Hernandez has pitched over 2,200 innings and has a career batting average against of .238. Mike Trout has a BvP against Hernandez of .354 (23 for 65) with 4 HRs. Albert Pujols has a BvP against Hernandez of .208 (11 for 53) with 1 HR. To simply dismiss the differences between the Pujols and Trout BvP results against Hernandez because of a small sample set to me seems almost as far-sighted as the argument about the batter who is 5 for 10 owning a pitcher is near-sighted. The truth simply lies somewhere in between the two extremes. I recommend using BvP as a tie breaker on players that you have evaluated to be similarly suited for your cash lineup on a given day. In doing so give extra emphasis to BvP stats that have a sample of at least 50 ABs which, while still a very small sample size, might give us a better idea of how a player will produce the next time he faces a given pitcher.
All in all, the process of creating any DFS lineups is just that, a process. It is a process of gathering pieces of information that when all viewed together present a picture of a hitter that gives you a good understanding of how that hitter will perform on a given day. No one statistic can provide you with the understanding or insight needed to build consistent building lineups, there is no magic bullet. I hope that this article about some of the factors I consider has been helpful to you in building your Cash lineup MLB DFS Process.