Moving Away From Spray Charts
Spray charts are something that have been around for a long time. They were and still are a really nice way to gather a lot of information on hitters. For those who are not familiar a Spray Chart is a graph that shows where batted balls are hit during a given period of time.
For Example here is Juan Soto from 2020.
They often will fail to give coaches and hitters concrete ideas of what is happening. We can see he has a lot of batted baseballs in different directions. Even if we start to sort out by different statistics I think it gets really hard to determine what makes the most sense and what the total outcome is from this type of plot.
The pros to a spray plot are extensive such as showing overall hitting tendencies and general production when balls are hit into play by a batter or team. When given filter options we can start to see where the power of a spray chart comes in.
Again this plot does start to give us some information to work from. The biggest downfall for application to player development or even helping coaches understand players production more is the fact that these types of charts are ambiguous for action. We don’t get clear takeaways that people can go off of.
Without the proper manipulation of data though spray charts are often vague and can be more misleading than anything else. We can take the above chart in a lot of different directions. Some might interpret the information in all different ways.
The goal for application of data isn’t to just take as much information as we can and throw it at people, but rather to help them to understand a clearer picture of reality.
Slicing Up The Field
Taking Spray Charts in a new direction by slicing the field into different pie pieces (Thanksgiving is around the corner) to me makes more sense. We have all seen the visual while watching a game:
The general idea that I had was to take this type of plot, and expand it over the entire field. This creates a type of binning effect to the information. This should allow us to have a more actionable visual to work off of. For example:
Here is Juan Soto with the Pie Slice filled with his BIP per Field:
Lets compare the two plots now to just see the differences:
It becomes striking how once can provide us with more information quickly rather than working to find key points.
Being able to provide players with this type of information and having it visualized in an easily understood way is the key to making the data integration work. If we want players to be able to take the information and run with it we need to make it easy for them to understand. This is where spray charts fall short. Spray charts are easy to see but they don’t really provide enough context and information for the player to take action from them.
Taking action with the information is the next step. I think that this is what can start to help us. Deciding what we value as a coaching staff or organization should be the first step in the process. Do we want hard hit baseball, do we care about spray percentages for all balls? Do we want players to be able to drive the baseball in all directions? I think that starting with a key metric and diving into it that way should be the first priority for a staff. Once you have that goal in mind you can start to piece together what you want to accomplish.
Once you decide what you value you can start to feed in different metric such as Launch Angle or Max Exit Velocity to different fields. Here is a smattering of different plots from Juan Soto.
Giving players and staff access to easily digest information is what we should be aiming for, rather than blanket sets of information or just reams of data that might overwhelm rather than add to a system.
Overall I think that baseball has an overall positive view of spray charts and in the past this reputation is well deserved. I just wonder if now that we are gathering more information and it can be presented in different ways that maybe we should update how we view Batted Ball information.
You can find more information and plots here:
https://joshua-rodrigues.shinyapps.io/HitterDashBoard/
“The ability to drive the baseball to all fields”
Wanted to also take the space to look the old adage of players being able to hit to all fields. I wanted to take a peek at how hitting the baseball hard plays in terms of how it correlates with overall production.
Here we are going to take the field and cut it into 5 pie slices similar to above. Then we are going to get a count for how many different fields a hitter is able to hit the baseball greater than 100.
(We only find that we have 25 players in 2020 who were able to hit to all 5 fields harder than 105. Here are a few with the complete list linked below. (https://rpubs.com/jrodjrod/field5)
Here is how the actual boxplot plays out.
We can see that players who can hit to 4 or 5 fields do better than those who have hard hit baseballs to only a few field locations.
What if we lower the threshold to 95 MPH?
We see again that players who can hit baseballs hard to all field come out on top again. So it would seem that it is something to keep track of, and to potentially reward. It could be that the best players are just able to hit baseballs hard to all fields. This would seem to be an area that we can dig deeper into.
