In 2007, pitching analysis was changed forever with the addition of the Pitch FX system. While the technology in 2007 was not as widely available throughout all ballparks as it is today, it has helped the average fan be able to take a critical eye to pitches. Have you ever wondered how much vertical break Barry Zito’s curveball had? How about the angle Randy Johnson’s slider took as it swept across the plate? What about how our own Jonathan Papelbon throws his fastballs? Now you have the power.
I’ve used Pitch FX before when doing in-depth analysis of pitchers, which is especially helpful when investigating subtle differences in things like a player’s suddenly burgeoning strikeout rate, or even a change in a specific pitch type’s frequency, angle, and location. A piece on Jon Lester I wrote a few months ago was a collection of most of those points.
In order to begin to do your own analysis, there’s a few things you’ll need, some of which require a technical touch. For those of you like me, with a programming and database architectural background, Baseball on a Stick is a newer Python and MySQL based system you can use to create your own personal database. Mike Fast has also written a thorough step-by-step method to create your database based off the XML serialization of MLB’s gameday data and a group of perl spider scripts.
But if you are like the 90% of the baseball world that would instead choose to leave the database jumbling to us engineering nerds, BrooksBaseball.net has a web-based Pitch FX tool that lets you run your own queries on a game-by-game basis. While it’s much more difficult to serialize yearly or career data of a specific player, being able to analyze a pitcher’s performance during a specific game is very valuable.
Let’s run through a quick example. Here are Jonathan Papelbon’s 2008 and 2009 fastballs.
Papelbon's 2008 (blue) and 2009 fastballs
Here’s the easiest way to picture how a baseball acts when reading this graph. The 0,0 point below and to the right of the center can be pictured as a pitch that is thrown with no horizontal or vertical movement supplied by the pitcher. This means the pitch is affected solely by natural forces (gravity, air resistance). Imagine a pitcher throwing an 8-Ball: a smooth surfaced projectile* that cannot have its direction influenced by spin because its surface consists of no irregularities needed to generate an opposed force to gravity (like a fastball thrown with backspin) or harmonious to gravity (a curveball).
* Please note that, yes, a billard-ball is not a frictionless surface, but the point I’m making is purely for academic comparison to how a pitch with 0,0 movement should act.
Horizontal movement is measured in relation to the catcher’s perspective. Negative horizontal movement signifies pitches that move away from the catcher’s right hand. This is why you’ll see a lefthander’s curveball have negative horizontal movement, while a righthander’s two seam fastballs also have negative horizontal movement.
Vertical movement is the expected “drop” of a pitch. A pitch with 0 vertical movement is only affected by gravity and will fall to the earth as such. Pitches with positive vertical movement are thrown with top spin and fall to earth slower than expected. Remember when analysts would rave about Pedro’s magical “rising” fastball? He could throw his fastball with a Herculean amount of spin such that the batters would expect the ball to drop sooner, creating an illusion of “rise”. All rising fastballs are are fastballs with extreme back spin that very gradually lose their altitude in comparison to a normal pitch.
So what can we glean from this simple pitch graph? We all know Paps throws lots of fastballs, right? We also know that his fastballs have some movement into right handers, so we should expect to see two large collections* of fastballs: Ones that look like 4 seamers and ones that look like 2 seamers. However, all of Jon’s fastballs seem to be an amorphous blob, almost like the average area of the two.
*For reference, SOSH has a nice example of expected pitch locations for a right handed pitcher (flip the graph to imagine a left hander’s release). This is where I make my assumption of where pitches should end up. Eventually, you’ll be able to identify pitches solely by their horizontal and vertical movements!
I decided to dub this blob Jon’s “3-seamers” as it’s basically a combination of 4 seamer and 2 seamer action. He throws it for speed and some inward movement towards a right handed batter, but unlike most power pitchers, he doesn’t really throw two very distinct fastballs. If you look at Josh Beckett’s Pitch FX, you’ll see two distinct groupings of fastballs like the example linked above. This “3-seamer” Paps throws almost seems like an “average” of what a 4 seamer and 2 seamer would look like.
While Jon does dial up a few fastballs to 99 and throw more classic 4-seamers on those 0-2 counts to left-handed sluggers, he’s become adept at throwing a hard fastball while still adding some late and deceptive tail. This is why we see one big general fastball area, and a few outliers of more horizontal movement but more drop (like a classic 2-seamer) or less horizontal movement and less drop (a classic 4-seamer).
I came to this “3-seamer” conclusion by looking at these Pitch FX graphs. It’s almost as if Jon is slowly refining his fastball to be more reliable and act in a repeated manner; you can see this as his 2008 fastballs movement is more varied and spread, and his 2009 fastballs are more tightly clustered. When looking at Pitch FX graphs, tightly clustered pitches give you an indication of consistency.
After 930 words, we’ve discussed one pitch for one player. This is the magic of Pitch FX. There is a ton of data and a seemingly endless amount of situations you can analyze. As the season starts, we’ll hopefully be showing you more game-by-game Pitch FX collections. You’ll be amazed as you see that one flat curveball, and then realize that was the one that went over the fence.
Have any analysis about Sox pitchers you’d want to see put under the Pitch FX microscope? Want to attach some reasons to what you witness a pitcher doing? Let us know, and we can gladly oblige.