I was playing around with some data here at lunch and I stumbled over to Manny Parra and I decided to do a little hand editing to his player card to tease out his splitter from his change up.  Normally this isn't a problem that my algorithm has to worry about because only a few pitchers throw both a splitter and a change up.  Generally, the pitches tend to move similarly and are thrown at about the same speed so I would presume that pitchers pick whatever one they throw better and just use that pitch.  Because Manny uses both I was curious to see when and how he was using them.  Here is what I found.

Starting with his movement chart you can now see the difference between the splitter and the change.  Let me first note that Parra's change is a straight change which means it mimics the movement of the fastball.  The more I look the more this seems to be the change up of choice for lefties as other lefties such as Johan Santana and Tom Glavine throw the straight change.  Cole Hamels is an exception and you can see the extra horizontal and vertical movement he gets with his change.  This is important because if Parra threw a circle change that change up would also have less vertical movement than his fastball and tend to merge with his splitter.

While both Parra's change and splitter are thrown at 84 MPH and he throws both about 13 percent of the time, the change up moves more horizontally and vertically.  This really creates two separate pitches and Parra uses them completely differently.  First, Parra almost never throws his change up to a left handed batter which is something that is a very common trend especially with left handed pitchers.  Parra will gladly throw his splitter to lefties however as that pitch doesn't move in so he can keep that pitch from ending in the happy zone of down and in to lefties.

The splitter is his strikeout pitch and he throws it a ton on 0-2, 1-2, and 2-2 (37, 40, 31 percent respectively) but rarely early in the count and never when he is behind.  The change up is most used when he falls behind 1-0 or 2-1 when he throws it 24 and 29 percent of the time.  He also will throw it early in the count but rarely if the count gets to three balls.  So if you are watching a game and are wondering if that pitch was a splitter or a change up check the count (and the handedness of the hitter).

I haven't talked about Manny's fastball or curveball but both are plus pitches.  Manny is very Ben Sheets like with these two pitches throwing his fastball at 93 MPH and a very over the top curve that he throws harder than many lefties who tend to feature lollipop curves (Think Barry Zito, Ted Lilly, etc).

What I really want to spend so time on is Parra's release point.  Actually, not his release point but where he is standing on the rubber.  Generally pitchers pick one spot that they are comfortable with and stick with it throughout the year but Parra has been fiddling with this the entire year.  Here is a look.

When I saw this I expected that there would be some cut off date where a switch was made from one of the clusters to the other but actually he has been going back and forth all year and even during some games.  In fact, if you look really closely you can see three clusters; one around 1.75 feet, one around 0.75 feet and one basically at zero and he has moved back and forth as the year has gone.  Here is a link to his game log so you can follow along.  Starting in April Parra's first three starts he was way over to the first base side exclusively.  On April 22^nd against the Cardinals, Parra moved to the middle release point.

Then on April 27^th against the Marlins, Parra moved to the third base side of the rubber for right handed batters but in the middle for left handed batters.  I haven't seen this before though I assume that other pitchers might be doing this.  I might have to go back and watch this game to see this as the difference is more than a foot and should be easy to see on TV.

Even though the Brewers lost that game Parra pitched pretty well so he continued this in his next game against the Astros in Minute Maid Park.  You might remember the results as Parra served up two home runs and six runs, five earned, in just four innings.  Undeterred, Parra again did this routine against the Cardinals in his next start on May 9^th before finally scrapping it and moving to the middle for his next start on May 14^th against the Dodgers.

Parra stayed in the middle throughout the middle part of the season until July 3^rd against the Diamondbacks where he went back to the first base side.  That continued only one more start however as he against faced the Astros at their place.  Despite getting the win, Parra didn't pitch particularly well and he went back to the middle for the rest of July.

August started in Cincinnati and here Parra moved over to the third base side the entire game.  Dunn and Griffey had been traded so the Reds were mostly right handed that game and I guess Parra wanted to pitch inside on them but after a poor performance that day Parra moved back to the middle and has stayed there up until now (actually I don't have numbers for his last start against St. Louis yet).

This is really interesting behavior from a pitcher and should be something interesting to watch as the season goes on.  This moving back and forth hasn't appeared to have changed his stuff in any way, all his balls still move the same and the vertical part of his release has generally stayed the same which is why I know this is a just him adjusting where he stands on the rubber and not an alteration of his arm angle or something like that.