New PitchFx Location Charts
We’ve added new PitchFx charts to our game charts. They’re pitch location charts vs right and left handed batters from the catcher’s point of view. These are modeled after much of Dave Allen’s work:
David Appelman is the creator of FanGraphs.
Deeply awesome, David. Very clear graphs. Question: why do the circle sizes change?
I had the same question as studes. More than one point on top of each other?
I’d also love to have this accessible for an entire season, rather than just game by game.
Great addition!
-j
I just made the hit circles a little bit bigger because I didn’t think they were all that visible. In terms of more than one point being on top of each other, not sure what I can do about that. I think occasionally there will be pitches with very similar coordinates.
Entire season graphs are coming soon!
Any way to tweak the categories? For example:
Ball
Called Strike
Swinging Strike
Foul
Ball in Play
I think the circle outlines make it a little confusing and I agree with Ben above that there probably should be more categories. My suggestion would be to do this:
Colors = pitch type
Shapes = pitch outcome
For pitch outcomes I would use swinging strikes, called strikes, called balls, fouls, in play hits, in play outs. So a slider that was taken for a ball would be a yellow square, and a changeup that was hit for a double would be a green circle (or whatever).
I echo Dave above though. These are excellent, much better than the Brooks graphs which use tiny and indiscriminate dots for the pitches.
I’ll admit I’m a little concerned about information overload, but I’ll mock some up and see how they come out.
I’m with you on info overload, David. Keeping these simple is a good way to go. There is already a lot of information packed in there.
Well I think there is a balance. When I did individual game Pitch f/x analysis at Driveline Mechanics I used to have two charts. 1 was location by pitch type, the other was location by pitch outcome. It was pretty easy to mix and match the charts, but I figure it might be better to include them in one.
The captions need work on spelling “distance” correctly though! They look great otherwise.
The captions on the actual charts are fixed, just need to clear the cache on my end. Though, I probably won’t bother fixing this particular graph in the post.
Look great David!
Fantastic stuff. Keep up the great work guys!
These are great. One question though: Could the x- and y-axis scaling be changed so that they are the same? At the moment, the horizontal axis is expanded relative to the vertical axis, which (IMO) makes the graphs a bit harder to interpret in terms of what a batter “really” saw.
Maybe I’m missing something, but you seem to have “inside” and “outside” set at 1 foot away from the center of the plate. The plate is 17 inches wide, so wouldn’t 8.5 inches from center be the relevant distance?
Yeah, you’d think so. But it’s more complicated than that. A strike is called if any part of the ball crosses over any part of the plate, which effectively makes the plate wider by (almost) the width of a baseball on either side, or just under 20 inches (1.66 feet). But beyond that, the eye of the umpire and convention comes into play. So regardless of what the theoretical dimensions are supposed to be, the effective width of the strikezone is the region where a thrown pitch is at least 50% likely to be called a strike, right? And when researchers have looked at a large sample of pitches, they have discovered that this “real” called strike zone is almost exactly two feet wide. (For right handed-batters, it’s 12 inches to each side of the plate; for left -handed batters, it extends even further outside but not quite as far on the inside part of the plate). Here’s one study done by John Walsh in 2007 that shows this; he followed it up with another. There are others, but they all show the same thing: the “real” strike zone is about 12 inches on either side of the plate, so that’s how Brooks and TexasLeaguers and pretty much everybody that uses PitchFX data draws the zone. If you want to take issue with the width of the zone, you need to take it up with the umpires, not with Fangraphs
Thanks. I thought there might be some explanation like that.