Cleveland @ Yankees: Musings on Pitch f/x

Technology is pretty darn cool.


I got to see a few moments of the Yankees-Indians game yesterday, and while I'm certainly glad that the Yankees won, well, heck, they're supposed to beat Paul Byrd, so that's not too terribly exciting, is it?


However, in the process I got to see Joba Chamberlain and later Mariano Rivera pitch. With the exception of a double to right field that Bobby "Gold Glove" Abreu mis-played, Rivera was his usual, Automatic self.


During the at-bat of Franklin Guttierez, however, I got a little annoyed whren I saw that Mo's 1-2 fastball, which appeared to be right down Broadway, as they say in New Yawk, was called Ball 2. Disagreeing with the umpires, for good or bad, is nothing new, but then I remembered that with Pitch f/x, we can actually see whether or not they were right. So I wnet online and looked. Here is the image I downloaded from the Guttierez at-bat:




It shows you the trajectory and eventual location of each pitch thrown in the at-bat, and if you're in the module online, you can mouse over each pitch and it will tell you both the release speed and the eventual speed of the pitch as it crossed the plate, plus how much the pitch broke in each direction while en route, and what the result was (i.e. called strike, foul ball, etc.).


Green means "Ball", red means "Strike" and blue means the ball was hit in play somewhere. The numbers in the circles represent the order of the pitches. Here, for example, you can see that Pitch #1 was a ball, knee-high over the middle of the plate, which therefore should have been called a strike. Pitch #3 was a strike (in this case a foul). Pitch #4 was another ball, and this was the one I thought was a strike, though it appears to be right on the inside corner (Guttierez, at 6'2" is tall enough that it the ball was right at his belt, despite what the graphic to the left seems to show). Pitch #5 was in play, a pop-up to Robinson Cano.


And Pitch #2? You can't see #2 because it is literally underneath Pitch #4, in the EXACT SAME LOCATION, only that one was called a strike. Evidently the strikezone gets smaller for home plate umpire Scott Barry as the at-bat goes on. Gonna make you work for it, this guy.


To his credit, Mariano does not let this get to him. Ever the consummate professional, he just smirks, lets it roll off his back, and then throws a third pitch in almost the same spot, inducing Guttierez to pop up on a pitch that was actually off the plate inside, if only by a hair. Rivera's velocity and control appear as good as they've ever been, despite his advancing years.


Contrast this with Mike Mussina, who stared the game for the Yankees. Moose pitched well enough to win, but he did give up three runs in only five innings of work. At the beginning of the game he was throwing in the mid- to low-80's, and the Pitch f/x technology couldn't decide how to categorize his pitches:

As you can see, he's got three pitches at almost exactly the same speed (85-86 mph) with a 7" break and a 14-15" PFX (whatever that means), but two of them are called change-ups and one is a "fast"ball. And it's not that the benchmark for calling something a fastball is 86 mph, either. They called an 84-mph pitch a fastball in the at-bat before this one. Furthermore, pitch #2, at 81-mph looks to me like it's almost exactly the same pitch, with just a little off it, and that one gets called a "slider". Mike Mussina doesn't throw a slider. He has a big overhand curve (what I think used to be known as his knuckle curve, though I don't know if anyone calls it that anymore) and another, little side-arm curveball, with sharper but less pronounced action on it. But not a slider.


Looking at the rest of his outing, I noticed that Pitch f/x frequently calls pitches that seem to have very similar characteristics either sliders or change-ups, and often calls pitches as fast as 85 mph change-ups, and sometimes calls pitches as slow as 83 or 84 mph fastballs, which of course is impossible. At least one of those labels has to be wrong.


While Mussina did manage to dial it up as high as 88 mph once or twice yesterday, most of the time his fastball sits around 84-85 these days, an dof course, that's just the speed at the release point. Wind resistance and spin can slow a pitch down 5-6 mph on its way to the plate, which means that when Mussina throws a fastball at 84 mph, it's really only about 79 mph when it gets to the plate. At the ballpark or on TV, for the sake of the fans, they usually have the radar gun up a few ticks, usually about 2-3 mph, which you can see from almost any game, comparing the Pitch f/x numbers (assuming even those are accurate) and the numbers on the screen. When Joba Chamberlain wa sin to pitch the 8th inning, they were adding as much ad 4-5 mph to his velocity, nearly always calling it 99 mph, while it was really more like 94-97, which is plenty, I think.

Bob Feller

By comparison, here's some video from an old newsreel of Bob Feller pitching, measured at 98.6 mph. But the machine they're using is measuring the speed of the pitch as it crosses the plate, whereas the radar gun readings you typically see on TV and at the ballpark ar at the pitcher's release. Feller must have let it go at around 104 mph!

That's neither here nor there. I just found it interesting.

More important than speed, of course, is location, and as you can see from the Mussina/Carroll at-bat above, all of Moose's pitches were about belt-high, regardless of their speed or type, which is not good. Several of his at-bats look like this, though they're not always this consistent. With that kind of predictability, it's amazing he only gave up four hits.