We always hear our not-so-favorite broadcasters mentioning a player’s stats against a certain pitcher. Well, I think we are all hopefully smart enough to not take the results of 15 meetings between a pitcher and hitter seriously. Instead, we split samples into larger groups like right-handed hitter vs. left-handed pitcher. I have decide to cut the difference and create a spreadsheet which takes a middle ground. I grouped pitchers by handedness, velocity and groundball tendencies and found how hitters performed against the different pitcher groups.
First off, I wanted to have this Excel-only spreadsheet available online before the season started. Well, I got it done and working in time. Since Visual Basic macros were used in the final view, it doesn’t have a online option which I wanted. So today, I am going to make it publicly available, but at some point I hope to have it working in all spreadsheet formats and/or online at a place like Google.docs.
Boring Background Bull…Data
Career pitcher stats from 2002 to 2013 were used to create the data. I used data back that far because groundball rates (GB%) were only available from 2002 and later. The pitchers were grouped by handedness, then three times by average velocity and then those groups three time by ground ball rate. Here are the rates used:
|FB Velo||GB%||FB Velo||GB%|
|> 90.3||> 45.4%||>92.3||>46.5|
|88.4 to 90.3||41.3 to 45.4||90.2 to 92.3||41.6% to 46.5%|
|< 88.4||< 41.3%||< 90.2||<41.6|
In the end, the left-handed groups had ~60 pitchers and the right-handed groups had ~140. I wanted to divide once more by walk rate, but he sample sizes were getting to small for left-handed pitchers.
Now for each hitter, the output looks at how they did historically against similar pitcher using a basic wOBA equation for each of the 18 categories. All right, it is time to start looking at the data.
Using the Spreadsheet
Go ahead and download the spreadsheet from here:
Remember, I could only get it to work in its current state in Excel. Sorry.
Open the sheet up, OK all the macros and it should look like this:
The players’ wOBA against different pitcher types from 2002 to 2013 can be viewed. Now for the big disclaimer, the non-regressed values are not even close to being the player true value. I ran out of time to figure the regression amount (a project for next off-season). I did add two places a person can regress in the league average wOBA and the player’s total wOBA.
For fantasy owners, the output can be helpful in determining which players to use if their talent levels are close. In one daily lineup league, I am using left-handed Garrett Jones and Matt Joyce against right-handed pitchers as a platoon for one outfield spot. Here is how they have done against right-handed pitching with some regression values added in.
Both of them do markedly better against flyball pitchers (<41.6% GB%) and bit better against slower throwing pitchers.
While this tool is not as good as I would like, it does function nicely in its limited environment. It can be used to find if a certain hitter does against a subset of pitchers. Until I improve it at a later date, feel free to let me know any other addition you want me to make to it then.
Thanks to Bradly Woodrum and Peter Melgren for helping with this project.
We know you play in all sorts of leagues. So to help you fine-tune the analysis you’d like to read, we’ve added three tags to the categories on the right: Roto, Head to Head, and Daily Fantasy Update. Use these to get the information that is most relevant to your leagues!
Print This Post