A PITCHf/x analysis of Kelvim Escobar

Despite winning the American League West with a 94-68 record last year, the LA of Anaheim Angels have gotten short shrift from the PITCHf/x analysts thus far. The only writeup that the pitching staff has gotten was one by Joe Sheehan on John Lackey three weeks into the season. I’d like to remedy that a little bit today. The Angels had three outstanding starters: Lackey, Kelvim Escobar, and Jered Weaver. Let’s take a detailed look into the pitching performance of Kelvim Escobar.
Escobar is a 31-year-old right hander from LaGuaira, Venezuela. He was a former starter turned reliever (and closer) and back to starter again for the Toronto Blue Jays before joining the Anaheim Angels in 2004. He’s struggled to stay completely healthy, but overall he has turned in some fine numbers for the Angels in four years: a 43-35 record and 3.60 ERA in 109 starts, allowing 611 hits and 213 walks against 561 strikeouts in 653 innings.
Since the Big A was one of the original nine stadiums to have a camera system installed from the beginning of the 2007 season, the large majority of Escobar’s season was recorded by the PITCHf/x system, 2469 of his total 3141 pitches. This gives us a good data set to identify his pitches and examine his pitching tendencies.
Escobar throws quite an array of pitches: a four-seam and two-seam fastball, a changeup and split-finger, a slider and a curveball. According to scouting reports, he is capable with all six pitches.
Escobar Pitch Speed vs. Spin Force Angle
Here Iíve shown two graphs that I use for pitch classification. The first graph shows the speed of his pitches versus the direction they break, in polar graph format. The second graph shows the movement due to the forces of spin deflection and gravity on his pitches in the last quarter-second before they cross the plate.
Escobar Late Break
There are a couple other ways to look at the vertical vs. horizontal deflection over the whole pitch trajectory:
Escobar Vertical vs. Horizontal Pitch Deflection (spin + gravity)
Escobar Vertical vs. Horizontal Spin Deflection (spin only)
Escobar’s four-seam fastball runs 92-96 mph, and the average spin deflection he gets on the four-seamer is a 10-inch hop and a 4-inch tail in toward right-handers. Compared to a league-average fastball, that’s 3 mph faster but with a couple inches less lateral movement, probably due to the fact that his motion is more over-the-top than many right-handed pitchers. The four-seamer is one of Escobar’s main pitches to both lefties (26% of the time) and righties (27%).
Escobar’s two-seam fastball also runs 92-96 mph, but its average spin deflection is an 8-inch hop and a 7-inch tail in toward right-handers. The two-seamer is his primary pitch to lefties (28% of the time) and also a main pitch to right-handers (24%). I made the division between the four-seamer and the two-seamer by looking at the spin direction of each pitch on a game-by-game basis, but the dividing line between the two is still a bit fuzzy to me.
His split-finger fastball runs 85-89 mph, and its average spin deflection is a 6-inch hop and a 6-inch tail in toward right-handers. Escobar uses the splitter fairly often to left-handers (15% of the time) but only infrequently to right-handers (6%).
His changeup runs 83-87 mph, and its average spin deflection is a 10-inch hop and a 3-inch tail in toward righties. The 9-mph separation between his fastball and changeup is about average for major league pitchers. He uses the changeup more often to lefties (16% of the time) but also some against righties (11%).
Escobar’s slider runs 85-89 mph, and its average spin deflection is a 3-inch hop and a 2-inch break away from righties. That’s about 3 mph harder than the average major-league slider, with typical movement. The slider is one of his favorite pitches to right-handed hitters (25% of the time) and is rarely used against lefties (2%).
Finally, his curveball runs 79-84 mph, and its average spin deflection is a 3-inch drop and a 1-inch break away from right-handers. That’s about 4 mph harder than the average major-league curveball, with 12-to-6 movement that is somewhat rare. (The spin deflection on the average major-league curveball is a 2-inch drop and a 5-inch cut. John Walsh’s article is my source for league average numbers.)
Next, letís look at how Escobar mixes his pitches in different ball-strike counts, which Iíve split out by batter handedness. The picture gets a bit messy when a man throws six different pitches, but let’s dive in and see what we see.
Escobar Pitch Mix to LHH by CountEscobar Pitch Mix to RHH by Count
To lefties, Escobar uses the four-seamer on any count and relies on it a little more if he falls behind. He throws the curveball early in the count, 22% of the time with no balls, 9% of the time with 1 ball, and only 3% of the time with 2 or 3 balls in the count. He favors the two-seamer with 0 or 1 strike, 33% of the time, but uses it only 16% of the time with 2 strikes. Instead, with 2 strikes he relies on the splitter 32% of the time. He’ll throw the changeup at almost any count except 0-2 and 3-0, but he likes to throw it more when he’s behind in the count, in which case he throws it 25% of the time.
Early in the count with Escobar, lefties should expect to see the two-seamer, the four-seamer, the curveball, and the changeup, in that order. If Escobar gets the hitter down 0-2 or 1-2, he should expect the splitter (41% of the time) or perhaps a fastball (41%), but if the count goes 2-2 or 3-2, he should start to watch for the changeup, too (33%).
To righties, early in the count, Escobar throws hard stuff, 31% two-seamers, 28% sliders, 23% four-seamers, and only 18% of his other three pitches combined. When he gets 2 strikes, the two-seamer disappears (only 3%), but he’s willing to show the splitter (14%). The changeup gets used a little with 1 strike (11%), but at 2-1 or 2-2 it’s a favored pitch (26%), and at 3-2, it’s his favorite pitch (34%), like it was to lefties. Righties can expect the curveball mainly at a single count: 0-2, where Escobar uses it 28% of the time; it’s little used (6%) in other counts.
What kind of results does Escobar get with each of his pitches? His four-seam fastball is a pretty good pitch, but his two-seamer grades out poorer. All four of his off-speed pitches are above average. I should mention that the PITCHf/x games for Escobar are missing his two worst starts of the year, which skews all the following numbers a little bit in his favor.

LHH _Ball_ _CS_ _Foul_ _SS_ InPlay _Avg_ _BABIP_ _SLG_ __HR__
4-seamer 0.33 0.26 0.17 0.06 0.18 0.315 0.315 0.444 0.000
2-seamer 0.40 0.17 0.18 0.05 0.20 0.338 0.317 0.523 0.031
Splitter 0.35 0.09 0.19 0.17 0.19 0.294 0.273 0.441 0.029
Changeup 0.39 0.14 0.15 0.12 0.20 0.216 0.216 0.297 0.000
Slider 0.22 0.11 0.39 0.06 0.22 0.500 0.500 0.750 0.000
Curveball 0.35 0.32 0.09 0.13 0.11 0.353 0.353 0.412 0.000
RHH _Ball_ _CS_ _Foul_ _SS_ InPlay _Avg_ _BABIP_ _SLG_ __HR__
4-seamer 0.41 0.17 0.20 0.08 0.14 0.176 0.176 0.216 0.000
2-seamer 0.39 0.26 0.14 0.04 0.17 0.415 0.392 0.585 0.038
Splitter 0.36 0.04 0.15 0.19 0.26 0.158 0.158 0.211 0.000
Changeup 0.27 0.06 0.12 0.28 0.26 0.237 0.216 0.316 0.026
Slider 0.35 0.17 0.11 0.16 0.22 0.254 0.243 0.352 0.014
Curveball 0.39 0.23 0.11 0.14 0.14 0.154 0.154 0.154 0.000
Lg. Avg. _Ball_ _CS_ _Foul_ _SS_ InPlay _Avg_ _BABIP_ _SLG_ __HR__
Fastball 0.36 0.19 0.19 0.06 0.19 0.330 0.304 0.521 0.037
Changeup 0.40 0.11 0.14 0.13 0.21 0.319 0.295 0.502 0.035
Slider 0.36 0.14 0.17 0.13 0.20 0.310 0.286 0.481 0.033
Curveball 0.40 0.19 0.13 0.11 0.21 0.310 0.290 0.471 0.029

The league average information comes from John Walshís article. In the following pitch location charts, I’ve changed my color-coding a bit to try to improve readability for those with color blindness. Hopefully the new system is an improvement.
Bedard Four-Seam Fastball Location
Escobar works with the four-seamer on the outer half of the plate to both lefties and righties, although with lefties he works down more and avoids coming inside, and with righties he works up more and works inside just off the plate. He has some trouble throwing the four-seamer for strikes to righties (only 59%, compared to 64% league average), but when he does, and they put in play, he gets very good results: .176/.216 (avg/slg), compared to .330/.521 major-league average off the fastball.
To lefties, he’s much better at throwing the four-seamer for strikes (67%), and he gets a lot of called strikes (26% compared to 19% league average), but his results on balls in play are only fair: .315/.444 avg/slg. He didn’t allow a single home run in 31 fly balls hit off the four-seamer in PITCHf/x games. That is unusual–fastballs are the most homered-upon pitch for most pitchers.
Escobar Two-Seam Fastball Location
Escobar has trouble throwing the two-seamer for strikes, getting it over only 60% of the time. As with the four-seamer, he works mainly on the outer part of the plate to both lefties and righties. However, both lefties and righties have good success when they put the two-seamer into play. Lefties hit .338/.523, and righties hit .415/.585.
Escobar Split-Finger Fastball Location
The splitter is Escobar’s strikeout pitch to lefties, and you can see why. They swing and miss at it down and away more often than not. He doesn’t necessarily throw it in the strike zone that much, but he gets strikes because the hitters chase it. When he does get it in the zone, hitters do much better with it, making at least decent contact and racking up a .294/.441 line, including a home run.
He doesn’t throw the splitter nearly as much to righties, although I wonder if maybe he should. He still gets a lot of swings and misses (19%, compared to 13% league average), but righties are able to put the ball in play almost every time he gets the splitter in the zone. However, the right-handed hitters don’t fare nearly as well as lefties on balls in play, hitting only a meager .158/.211. Perhaps it’s the small sample size (19 balls in play), or maybe righties really do have trouble getting good wood on the splitter.
Escobar Changeup Location
The changeup is the first pitch where we see a marked contrast in Escobar’s location to lefties and righties. To lefties, he pitches away, away, away. He gets some swings and misses in the zone, but lefties don’t chase the changeup out of the strike zone much. On balls hit into play by lefties, Escobar does well, a .216/.297 line, compared to .319/.502 against an average major-league changeup.
To righties, he throws the changeup mostly in the zone or on the corner low and away. He gets a lot of swings and misses, especially on the outside corner. The changeup is a very effective pitch against righties. No wonder he likes to throw it as a strikeout pitch to righties. Moreover, even though he pounds the heart of the zone, righties have little luck on balls in play, hitting only .237/.316. Most right-handed pitchers avoid throwing the changeup to right-handed hitters, but for Escobar in that situation, it’s a great pitch and one he could perhaps use even more often.
Escobar Slider Location
As you can see, his slider is rarely used to lefties, mostly thrown up and in and fouled off. To righties, he uses the slider a lot, and to good effect. He gets a good number of called strikes (17%, versus 14% league average) and swinging strikes (16%, versus 13% average), and when the ball is put in play, Escobar also fares well, allowing a .254/.352 avg/slg, compared to .310/.481 against an average major-league slider. Those numbers include allowing only 1 home run on 27 fly balls hit by righties off the slider–luck or skill?
Escobar Curveball Location
Finally, we come to the curveball, Escobar’s least-used pitch. He throws it mostly down and away to both righties and lefties, although he also throws it in the zone quite a bit. He gets a lot of called strikes, especially to lefties (32%), but also to righties (23%), compared to league average of 19% with the curve. Most pitchers rarely throw the curveball as the first pitch to a batter. Escobar, on the other hand, often throws a lefty a curveball right across the plate for strike one looking. Lefties don’t often make contact with the curveball, but when they do, the results are decent: .353/.412, compared to league average against the curve of .310/.471.
Right-handers see the curveball more often with two strikes, and it’s a good strikeout pitch for Escobar, both swinging (at balls in the dirt) and looking. Righties don’t make contact with the curve very often, either, and when they do, their results are particularly poor: in 13 curveballs in play, righties hit 10 groundballs (including two double plays), 2 fly balls, and one line drive. The line drive and one groundball landed as singles, for a .154 average.
In summary, Escobar has a solid four-seam fastball which he complements with a weaker two-seamer, and his array of off-speed pitches is impressive. His changeup, splitter, curveball, and slider are all well above average pitches, and some of them, particularly his changeup, are among the best in baseball. He struggles with control on his fastball, and this, along with the recurrent health problems, is probably all that keeps him from being one of the very best pitchers in baseball.
As a final note, I thought this was a great photo from MLB.com of Kelvim Escobar in full stride.

If you enjoyed this article, you might be interested in my similar previous analysis of Erik Bedard, Johan Santana, James Shields, Mariano Rivera, Joakim Soria, Josh Beckett, Joba Chamberlain, or Eric Gagne.


10 Responses to A PITCHf/x analysis of Kelvim Escobar

  1. jefeboy says:

    Nice work.

  2. Pizza Cutter says:

    Mike, when you gonna start looking at the Pitch F/X data for batters?

  3. Mike Fast says:

    Pizza, when I learn all there is to learn about pitchers. That’s my half-serious answer.
    I find the pitcher’s perspective much more fascinating. It’s easier for me to see an imaginary storyline in one pitcher’s data against hundreds of different hitters across a season than it is for me to see a connected story in one hitter’s data against hundreds of different pitchers.
    Understand the ins and outs of a given pitcher’s approach and you can understand half the ballgame on any given day. Understand a given hitter’s approach and you’re still left with a lot of missing puzzle pieces.
    This is probably why the majority of my favorite players have been pitchers…Charlie Leibrandt, Bret Saberhagen, Dan Quisenberry, Mark Gubicza, Jeff Montgomery, Kevin Appier, Tom Gordon, Mike Magnante, Jose Rosado, Mike MacDougal, Zack Greinke, Joakim Soria.
    That’s not to say I can’t or won’t look at the hitter’s side of things, too. I understand the analytical value there. Did you have something in mind?

  4. Mike, Mark Gubicza’s sister was my pre-school teacher.

  5. Mike Fast says:

    Gubicza was a incredible workhorse for the span of about three years. Saberhagen, Gubicza, and Montgomery were just phenomenal in 1989. It’s a shame the Royals didn’t have anything else to put around them. Well, Bo Jackson was a lot of fun, too, but 32 home runs doesn’t make an offense on its own, and The Throw was only one run…no, what I am saying, The Throw will live on as the best defensive play ever!

  6. Pizza Cutter says:

    Mike, judging by that list, I wonder what team you were watching all those years! I guess I’ve always been more impressed by hitting.

  7. dan says:

    Just curious Mike, about how long do these take you from start to finish?

  8. Mike Fast says:

    Dan, I do the analysis in bits and pieces, so I’m not sure exactly. Maybe 10-15 hours on average, would be my guess. Escobar took a bit longer because he had so many pitches recorded by the system, and someone like Bedard or Soria with less pitches takes less time.

  9. Mike Fast says:

    SirKodiak, the articles get too long if I try to include all the data. I intend to make my spreadsheet with raw data available for download, but I haven’t done that yet.
    Here’s the data broken down by pitch type (4-seam, 2-seam, split, change, slider, curve): to LHH (296, 330, 175, 185, 18, 156) and to RHH (358, 309, 74, 144, 330, 93).
    Broken down by count (0-0, 0-1, 0-2, 1-0, 1-1, 1-2, 2-0, 2-1, 2-2, 3-0, 3-1, 3-2): to LHH (307, 152, 69, 120, 109, 94, 48, 61, 94, 14, 33, 59) and to RHH (348, 167, 79, 143, 137, 97, 51, 76, 107, 14, 30, 59).

  10. SirKodiak says:

    It would be nice if when you chart the percentages you included the number of pitches in each category so we had some idea of usage to put with the %.
    Nice article on an interesting subject. Many pitchers are criticized for having too many pitch types and not focusing on less. His pitches seem to complement each other, especially the slider/change/splitter that are in same speed band, but different breaks (away/none/in).

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: