More fun with PITCHf/x

by dixieflatline on Sep 11, 2007 11:07 PM CDT

So I came up with an algorithm to try to identify what pitches pitchers are throwing. You can read more about it at my blog if you are interested. Anyway, I thought I would highlight a few of the Brewer pitchers here. It is a bit of a work in progress but it is getting there. All plots show the movement of the pitches not the location.

Gallardo

So the algorithm isn't perfect and you can see it with Gallardo. The pitches listed as a splitter is a curve but the relatively large horizontal break compared to the vertical break the algorithm is messing up and calling it a splitter. The real question is where is the change up in all of that? It appears that a few of those could sliders could be change ups but with such a low percentage it is getting grouped in with the sliders. I wonder if the relatively small horizontal break of his fastball is why he has such good command?

Cordero

Nothing too surprising here. Almost all fastballs and sliders. The algorithm thinks he has thrown a couple of changes in there. Not sure if that is the case or not.

Cappy

Again, the clustering algorithm is grouping Cappy's slider and change together. The real change ups are just below the fastballs. You can see why that pitch should be a good pitch for him as it resembles the fastballs except for the speed.

Shouse

Shouse's fastball is definitely a two seem variety and has some huge horizontal break. You can see how tough he is on lefties as the ball breaks from their thighs to just over the plate. His offspeed pitch is listed as a splitter (can anyone confirm that?) that rises slightly and moves the other direction than the fastball.

Anyway, it is time for bed here. Maybe I will post some more tomorrow.

Tweet Comment 10 comments | 0 recs |

Do you like this story?

Comments

sweet work!

very awesome stuff... just need to mix in velocity in the graphs. keep it up dixie, you make us proud. :)

Bring Back The Old Logo!

by jacob on Sep 11, 2007 11:23 PM CDT reply actions

I could look at stuff like that

all day. Kick ass job!

Capuano needs to develop a 3rd pitch. His fastball/change seems to do about the same thing. I think the other two pitchers may be more successful with the "gap" in their pitches.

I would guess that if you did this for Wise it'd look similar to Crappy's.

If A-Rod hits A-Bombs, does Fielder hit F-Bombs? It's dumb I know....

by Adam P on Sep 12, 2007 7:43 AM CDT reply actions

You want your change

to almost look exactly like your fastball, especially Cappy cause he throws slower.

You are right though, he definitely needs to use that other pitch more.

by Zel123 on Sep 12, 2007 9:32 AM CDT up reply actions

I like Cappy's "Unknown" pitch

It's unhittable! Batters end up standing there with a dumbfounded look on their face. "What the heck was that? It wasn't a fastball or a changeup."

"I will agree that the attitude [at BCB] is ridiculous and they have done so much to instigate animosity and then block us from responding. Real mature!"

by roguejim on Sep 12, 2007 7:56 AM CDT reply actions

Could it be....

The "Funky McNasty" as Yost called it?

If A-Rod hits A-Bombs, does Fielder hit F-Bombs? It's dumb I know....

by Adam P on Sep 12, 2007 8:57 AM CDT up reply actions

But I thought

Yo's pitch was a curveball, not a slider? No?

Hey Ned! Ever heard of small ball? Or are your balls too small?

by CATALYST on Sep 12, 2007 9:11 AM CDT reply actions

Well that grouping in the middle

Clearly seems like a slider to me (and the algorithm). It doesn't look like he throws it that much but it certainly looks like it is there. The curve definitely appears to be a much better pitch for him though.

Visit my baseball blog ...or else!

by dixieflatline on Sep 12, 2007 9:17 AM CDT up reply actions

so i did some research

and I have a friend who watches pitchers for a living and he said that the real flaw in looking at pitches like this is that:

"Those coordinates are not a real effective way to measure the movement of pitches because they just take the spatial difference in where the ball left a pitcher's hand and where the ball hits the catcher's glove. Break doesn't work that way. Break is about how the ball travels to home plate. A fastball and slider could be released at the same point and end up at the same point but not have close the same amount of break."

All Hail the Hebrew Hammer!!!

by grobbins on Sep 14, 2007 8:19 AM CDT reply actions

i'll let dixie be the authority on the data here

but from the looks i've take at the data my understanding is that the gameday data for "break" is not just release point coordinates minus crossing the plate coordinates.

It is instead two points, the release and a point about 60 feet from home plate that together determine where a ball with "no break" would end up. it's the difference between that imaginary pitch and the actual pitch that shows the break.

therefore the difference between the slider and fastball that are released from the same point and end up in the same point is actually captured here (to a degree).

Bring Back The Old Logo!

by jacob on Sep 14, 2007 9:02 AM CDT up reply actions

Absolutely correct Jacob

The break is broken up into x, y, and z components and then PITCHf/x fits each component to a parabola. This isn't perfect but pretty close. When I say vertical break I am talking about the actual break of the ball and that has nothing to do with where the pitch started or ended. Just the curvature. I can and will make plots that add in the starting or ending location but the plots above are just the break of the ball.

Visit my baseball blog ...or else!

by dixieflatline on Sep 14, 2007 9:07 AM CDT up reply actions

Comments For This Post Are Closed

FanPosts

Community blog posts and discussion.

Recent FanPosts

+ New FanPost All FanPosts >

NL Central Standings

	PCT	STRK
Chicago Bleed Cubbie Blue	.000	Lost 0
Cincinnati Red Reporter	.000	Lost 0
Houston The Crawfish Boxes	.000	Lost 0
Milwaukee Brew Crew Ball	.000	Lost 0
Pittsburgh Bucs Dugout	.000	Lost 0
St. Louis Viva El Birdos	.000	Lost 0