Scouting Jake Arrieta with PITCHf/x

August 13, 2009

Baltimore Orioles right-handed pitcher, Jake Arrieta and his Triple-A Norfolk Tides battled the Pawtucket Red Sox in the Futures at Fenway game held on August 8th, 2009. Arrieta got the win as the Tides downed the PawSox 7-3, but he was far from dominant. His start, by the numbers, looked as such:

6 IP, 6 H, 2 ER, 0 BB, 4 K, 0 HR, 82 pitches (49 strikes), 6 ground outs and 8 fly outs.

Arrieta's had a solid season but experienced a large dip in strikeout from Double-A Bowie to Norfolk (28.7% to 20.7%) -- his walk and ground ball rates have remained relatively consistent. Sporting a 4.74 ERA in Triple-A, the 6-foot-4, 225-pounder has been undone by a .346 batting average on balls in play, and has pitched nearly a run better than that -- evidenced by his 3.91 FIP.

To disect Arrieta's start in Fenway Park, I'll use PITCHf/x data and start by looking at his release points:


Image and video hosting by TinyPic


What this chart showing is not an actual "release point," just where the PITCHf/x camera system picks up the baseball, which is around 50 feet from home plate. Once they've actually released the ball, pitchers are generally around 55 feet from home plate, depending on the pitcher's height and stride length. Both the x and y axis are in feet. And the view is from the catcher's perspective. It seems that Arrieta was releasing the baseball, on average, around the five-and-a-half foot mark. That seems low given his 6-foot-4 frame and a delivery that doesn't seem low-3/4 to me.

To show how erratic he seemed to be with his "release point," here's all of his release points on his 82 pitches:


Image and video hosting by TinyPic


He ranged from right about the 6-foot mark down to the 5-foot mark, but was mostly somewhere inbetween. This inconsistency may be what caused his inability to pound the strike zone, as seen below:


Image and video hosting by TinyPic


This is also from the catcher's view. Arrieta didn't throw much at the knees, leaving a lot of pitches thigh-high and above -- though, most of those were fastballs). He did get squeezed on a handful of pitches that were strikes, but borderline calls are tough to get when you're missing the zone regularly.

Arrieta threw a four-seam fastball, change-up and a slider. There were a handful of pitches that Gameday coded as curveballs, but I think they were still sliders, so I've re-coded them as such. Here's a couple pie graphs on how he attacked both left-handed and right-handed hitters:


Image and video hosting by TinyPic


The bold number inside of each slice is the total number of those types of pitches thrown. Like nearly all right-handed pitchers, his change-up is dropped against right-handed hitters in favor of his breaking ball. Conversely, he used his change-up more often against left-handers, while decreasing his use of the breaking ball.

Finally, the PITCHf/x data can be transformed into a pitch flight generator, and this comes courtesy of PITCHf/x guru Harry Pavlidis' template. But it's been improved upon (with Harry's help) by Zach Sanders, and now includes a home plate for the bird's eye view and a "zone" for the first base view to give an impression on where his average pitch was crossing home plate.


Image and video hosting by TinyPic


As you can see, his average fastball (FF; blue line) was up in the zone (darkened black line on the 1st base view) and over the heart of the plate (home plate replica on the bird's eye view), and that kind of pitching will get hit around. He did seem to locate his change-up (CH; orange line) well to the outside corner against left-handed hitters and down at the knees. He controlled his slider (SL; white line) to the outside corner (mainly against right-handed hitters) and at the knees, as well.

Some negatives can be drawn, but may be directly related to his lack of control: His fastball looked pretty flat in this particular outing. He didn't generate a lot of arm-side run, either. His change-up mirrored his fastball in arm-side run, but had some solid diving action to it. His slider has some sweeping action to it and good diving action. It seems to be closer to an 11-to-5 slider than a 12-to-6 one.

There are some major caveats, though. 

1) I'm not a PITCHf/x guru in the least.

2) I have re-coded pitches at my discretion and very well could be wrong on that. I feel that there are probably some two-seam fastballs that I left as change-up's, but I'm not entirely sure in either direction.

3) PITCHf/x should always be supplemented with first-hand, in-person scouting reports.

4) This is not a perfect system and there are constant tweaks that are needed. The new Yankee Stadium's cameras are off by a couple of inches in their judgments of the "release points."

5) This is just one single start for Jake Arrieta. Much in the way that you can get a bare-bones outline on what a pitcher throws, you need to see multiple outings -- the more the better -- to get a solid hold on his arsenal of pitches, and to know what kind of an injury risk his delivery comes with, as well as how well he hides the baseball from hitters in his wind-up.

All that said, this is a fun way to look at some top pitching prospects making their major league debut or, in this case, an early look at an interesting pitching prospect. 

(Please join in the discussion on Arrieta in our Minor League forum.)


Mike Rogers runs his own blog, Fire Jim Leyland (no, he doesn't want him fired anymore). He also contributes to a San Diego Padres blog, Friar Forecast, and at the great sabermetric blog Beyond the Boxscore. He can be reached at