I wanted to follow up this article by checking out the difference in movement in Johan’s change-up between the two seasons, focusing on platoon splits. The fact that Johan threw far fewer change-ups to lefties in 2009 made me suspect a couple of things.
- That he might throw his change-up quite differently to lefties when compared to righties
- That the injury to his arm might have mainly affected his ability to throw the change-up to lefties.
Thankfully pitchf/x is a wonderful thing which allows us to investigate this further. This is my first time analyzing movement data using pitchf/x so I’m going to explain my understanding of what the numbers mean. Hopefully this is all correct.
- A positive score in horizontal movement represents the ball moving in to a left handed batter and away from a right handed batter relative to how much you would expect if the ball had no spin on it and is measured in inches.
- A positive score in vertical movement represents the ball sinking less than one would expect if the ball had no spin on it. This is also measured in inches.
So let’s take a quick look at the movement on Johan’s change-up from year to year.
There are a few rather interesting things to notice here.
- There was a huge difference between Johan’s change-up to lefties and righties in 2008. About 3 inches of horizontal movement and and inch and a half of vertical movement. This means that the ball moved away from righties more than it moved towards lefties and it sank more against lefties than it did against righties.
- This split goes completely away in the 2009 data. In fact the change-up moved in towards lefties way, way more in 2009 than it did in 2008. It also sank less against both lefties and righties in 2009 than it did in 2008.
This seems to confirm my initial suspicions quite nicely. The main difference between 2009 and 2008 in Johan’s pitches was his inability to throw his change-up to lefties the way he prefers to throw it. Since he was unable to throw this pitch effectively he cut his usage of it in half and replaced those missing change-ups with sliders instead. Hopefully now that his arm is healthy he will be able to throw his 2008 change-up to lefties once again and left-handed batters will not have the same success against Johan which they had in 2009.
Very interesting article. A few things I’d like to point out … with the PITCHf/x data, you have to be aware that there are park variances that affect movement and velocity game-to-game, and this is particularly evident in the 2007 and 2008 data. Also, if there are a few stray sliders accidentally classified as changeups, the average horizontal movement will have a lower number.
All good points. I was looking more after posting this and saw that a bunch of sliders were probably classified as change-ups (http://sabometrics.com/?p=370) which means I have to do a LOT more looking before I believe this is actually true. I’m too busy right now but later in the week I’m hoping to do more analysis.
Yeah … after I posted the comment, I saw the follow-up article. The site is great, by the way; keep up the good work.