Fastball, slider, changeup, curveball—an analysis
16
|
The Hardball Times found this 12/20/2007 on www.hardballtimes.com [flag] |
Tags:
MLB
Comments (7)
-
studes +1Great stuff again, John. Enjoyable reading. Thanks.-
walshj58 Thanks, studes.
-
-
mikefast John, great article!
What software are you using for the k-means clustering? Is that R? I'm trying to figure out how to do automated clustering. I gave it go in R but was a bit daunted by trying to figure it out from scratch. Do you have any pointers?
Also, I'm a bit curious about the large light-gray spreads around each of the pitch types. Some of the light gray areas I can conjure reasonable explanations for, like cutters for the fastballs that have a lot of positive horizontal movement. But I can't figure out how a pitch with +10 or +15 inches of pfx_z would get classified as a curveball under any scenario. Or how the pitches on the right side of the changeup distribution aren't being classified as sliders.
-
John Beamer Mike
Not sure what John uses but when I do clustering I use the kmeans function in R with start speed, pfx_x and pfx_z as variables. The only thing you have to specify is the number of pitches a pitcher has as that defines the number of clusters you are looking for.
I then run the algorithm that throws up a plot a couple of time because Kmeans has random start points, which can cause some odd clusterings.
Here is the code I use for a cluster of 3 with 1398 data points
datamatrix<-matrix(c(startspeed, pfx_x,pfx_z),nrow=1398,ncol=3)
clust<-kmeans(datamatrix,3,iter.max=30,nstart=1)
Hope that helps
-
mikefast Thanks a lot, John. Yes, that helps.
-
-
-
iblemetrics Has anyone looked at the effect on pitch selection of men on base? How many more fastballs are thrown with a runner on? Does it matter how fast he is? -
rpankovich i'm curious as to the effect of "hanging" pitches, i.e. pitches that don't move as anticipated. What effect does this have on the analysis? Perhaps this is how some of those curveballs have positive vertical movement, although if this were the case i would imagine that the grip of the ball was off given that a properly gripped curveball should have no component of backspin, thus no apparent upward movement.
Links (16)
Roundup - Schilling calls out Clemens
Published 12/20/2007 by Dave Rouleau at Baseball Digest Daily
... candidate right here . - Wonder who had the best eye at the plate in 2007? Who swung at the most pitches? Here is BDD's first offering by the Diva - MLB Trade Rumors just reported that Geoff Jenkins had signed a two-year deal with the Phillies worth 13 M$ with an option for 2010. - Boston starter Curt Schilling 's take on the Mitchell Report . Even some hometown events are wondering whether he should be present or not . - Report on the Fastball, Slider, Changeup and Curveball by John Walsh over at the excellent Hardball Times. - ...
Fastball, slider, changeup, curveball
Published 12/20/2007 by Tangotiger (tangotiger@yahoo.com) at THE BOOK--Playing The Percentages In Baseball
... Me hero of 2007 lays out his landscape for pitch types, counts, batting results. This is the kinda article that you need to print, and refer back to often. ...
White Sox Holiday shopping list
Published 12/21/2007 by thewizardsofoz <info@southsidesox.com> at South Side Sox
... Anderson but the odds of an Ozzie-Brian reunion don't look good. How does your Sox Holiday shopping list look like? ***** In other news, David Eckstein's ZiPS projection: .277/.334/.344 (80 OPS+). Hot Stove: The Flubs are considering signing So Taguchi. That could lead to Matt Murton being traded for less than he's worth. BTW, here are the 2009 MLB Free Agents from MLBTR. UZRs from 2003 to mid-2007. THT: Fastball, slider, changeup, curveball--an analysis. Tango gives it two thumbs up! ...
Breaking Down Adam Eaton's Breaking Ball, Part II
Published 12/21/2007 by Doc at Balls, Sticks, & Stuff | Phillies, Eagles, golf and other matters of great importance...
... and vertical breaks of Eaton's curveball versus the MLB averages. Vert. Horiz Eaton -1.6 4.9 MLB Avg -3.3 5.2 Diff(in) 1.7 0.3 It's pretty clear that Eaton's curveball breaks less than the league average, both horizontally and vertically. Now, whether this makes a difference or not is hard to know, but, when you consider that his curveball breaks only half as much as the league average's, well, it doesn't bode well. ----------------- Thanks go out to Josh Kalk and The Hardball Times for the use of the Pitch f/x data.
White Sox Pitch Type Data
Published 12/21/2007 by The Cheat <info@southsidesox.com> at South Side Sox
Prompted by a recent article at The Hardball Times, and the easy to use Pitch f/x tool created by Josh Kalk, I decided to see what we could learn about some White Sox pitchers. Before we go any further, I have to point out two rather large caveats. The linked tool and the linked THT article use different algorithms to determine pitch type. Do to the size of the sample (270K pitches), however, I feel we can still draw some useful info from comparing the two. This pitch data is incomplete for each player. MLBAM's pitch f/x system was ...
Can we classify every pitch?
Published 12/22/2007 by Mike Fast at Statistically Speaking
... having a universal system for classifying >95% of major league pitches accurately. A published system with that level of accuracy would serve as the spark to ignite a great deal of other research that could transform the game of baseball. Perhaps Sportvision and MLBAM are developing such a system in private. Perhaps Joe Sheehan’s unpublished system is closer to that goal than I realize. Perhaps Josh Kalk is on the way to making refinements to his system that will get us closer. Maybe John Walsh’s approach , which went to press as I was writing this article, of using K-means ...
Winter Wonderland
Published 12/22/2007 by Joe Sheehan at Baseball Analysts
... John Walsh wrote a fantastic piece on Thursday about the differences between fastballs, sliders, changeups and curveballs, and what happens when those pitches are put in play. I've done some research into this area myself and wanted to graphically present some of my findings. ...
Yep, Josh Fields can't hit fastballs
Published 12/28/2007 by The Cheat <info@southsidesox.com> at South Side Sox
After looking at the White Sox Pitch Type Data for the returning starting pitchers, I decided to see if we could learn anything from the available pitch type data for our hitters. Well, just Josh Fields for now, but I plan on expanding that in the near future. Again, I'll point you towards The Hardball Times article with league averages and let you draw your own conclusions from Fields' data. I've included Contact%, which isn't included in the THT article, but is available on all player pages at Baseball-Reference.com. I don't have the pitch-type ...
Tales of the curve: an analysis of Erik Bedard
Published 1/3/2008 by Mike Fast at Statistically Speaking
... The league average information comes from John Walsh’s article, and I’ve adapted his format in presenting this information. His pitch types probably don’t correspond exactly to mine since he lumps sinkers and cutters in with four-seam fastballs and splitters in with changeups. I believe it’s still helpful to use his league-wide information for comparison since I haven’t established a league-wide baseline on my own yet. ...
Special f/x
Published 1/3/2008 by Jim McLennan <info@azsnakepit.com> at AZ Snakepit
... That aside, it is still possible to draw some overall conclusions abour pitch selection. We'll hold over beginning the analysis of specific D-back pitchers and hitters until next time, but we again turn to the invaluable work of John Walsh, for an overview of what pitches get thrown in specific counts. Cnt | FB% | SL% | CB% | CU% -----+------+------+------+----- 0-0 | 0.63 | 0.15 | 0.12 | 0.09 0-1 | 0.52 | 0.20 | 0.15 | 0.12 0-2 | 0.51 | 0.21 | 0.18 | 0.09 1-0 | 0.63 | 0.15 | 0.08 | 0.13 1-1 | 0.53 | 0.19 | 0.13 | 0.14 1-2 | 0.48 | 0.22 | 0.19 | ...
Tales of the changeup: an analysis of Johan Santana
Published 1/9/2008 by Mike Fast at Statistically Speaking
... The league average information comes from John Walsh’s article, and once again I’m using an adaptation of his format to present this information. ...
A PITCHf/x analysis of Kelvim Escobar
Published 1/29/2008 by Mike Fast at Statistically Speaking
... Finally, his curveball runs 79-84 mph, and its average spin deflection is a 3-inch drop and a 1-inch cut in toward 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.) ...
Winning with an 89-mph fastball: an analysis of Brian Bannister (Part 1)
Published 2/24/2008 by Mike Fast at Statistically Speaking
... at the Baseball Hall of Fame. It is well-known that a pitch knee-high on the outside corner will not have the same batting average or OBP/SLG/OPS as one waist-high right down the middle. Here is a comparison of the batting averages and slugging percentage on my fastball vs. my curveball: Fastball: .246/.404 Curveball: .184/.265 We do know from John Walsh’s work something about batting average and slugging percentage against the typical major-league fastball (.330/.521) and curveball (.310/.471). If Bannister is correct in his numbers, he’s doing quite a bit better than the ...
Winning with an 89-mph fastball: an analysis of Brian Bannister (Part 1)
Published 2/24/2008 by Mike Fast at Statistically Speaking
... at the Baseball Hall of Fame. It is well-known that a pitch knee-high on the outside corner will not have the same batting average or OBP/SLG/OPS as one waist-high right down the middle. Here is a comparison of the batting averages and slugging percentage on my fastball vs. my curveball: Fastball: .246/.404 Curveball: .184/.265 We do know from John Walsh’s work something about batting average and slugging percentage against the typical major-league fastball (.330/.521) and curveball (.310/.471). If Bannister is correct in his numbers, he’s doing quite a bit better than the ...
Winning with an 89-mph fastball: an analysis of Brian Bannister (Part 2)
Published 2/26/2008 by Mike Fast at Statistically Speaking
... I can see evidence of two clusters when pitchers throw both types of fastballs, even if the exact delineation is hard to make. In Bannister’s case, I didn’t even see two separate clusters. In addition to the fastball, he threw a changeup, slider, and two varieties of curveball. Let’s examine these pitches in more detail. Bannister’s fastball runs 87-91 mph, and the average spin deflection he gets on the fastball is a 9-inch hop and a 1-inch tail in toward right-handers. Compared to a league-average fastball, that’s 2 mph slower and with about 5 inches less lateral movement. ...
Winning with an 89-mph fastball: an analysis of Brian Bannister (Part 3)
Published 2/28/2008 by Mike Fast at Statistically Speaking
... on fly balls 0.251 on ground balls 0.729 on line drives 0.025 on popups (As an aside, why do people prefer ground ball pitchers if the BABIP on ground balls is higher than on that on fly balls? It’s because 78% of fly ball hits go for extra bases, whereas only 9% of ground ball hits go for extra bases.) Second, what is the expected distribution of balls in play to the different types, more to the point, what is the expected distribution against the fastball? Here I’ll draw on o ne of John Walsh’s many excellent pieces on PITCHf/x. According to his numbers, the typical ...
