More on Sonnanstine and "hittable"
by rglass44 on Jun 5, 2009 12:02 PM EDT
Email
Print
Inspired by FreeZorilla's great post, I decided to delve into the data just a bit deeper. So I collected all pitcher seasosn from 2002-2008, and compiled every pitchers batted ball data who pitched at least 300 innings over that time frame. From there I took a look at what their BABIPs were, and how their xBABIP (for this purpose it is figured the same way he did it but I used a rough mean of his 2008/2009 figures) compared. This, in a sense, is a measure of how "hittable" a pitcher is.
My sample contained 322 players, of which there are 8 current Rays and Edwin Jackson. Here is how they fared compared to the grand total of all pitchers who pitched from 2002-2008.
| Rank | Player | IP | LD% | GB% | FB% | BABIP | xBABIP | Diff |
| 2 | Andy Sonnanstine | 323.3 | 17% | 41% | 42% | .319 | .283 | 0.03576 |
| 4 | Scott Kazmir | 721.6 | 18% | 40% | 42% | .317 | .288 | 0.02963 |
| 9 | Edwin Jackson | 454.6 | 20% | 42% | 38% | .321 | .299 | 0.02139 |
| 16 | Matt Garza | 317.2 | 18% | 42% | 39% | .308 | .291 | 0.01724 |
| 40 | James Shields | 554.2 | 18% | 44% | 38% | .301 | .290 | 0.01174 |
| 113 | Lance Cormier | 313.8 | 22% | 51% | 27% | .321 | .318 | 0.00283 |
| 194 | Dan Wheeler | 400.5 | 19% | 37% | 44% | .284 | .289 | -0.00567 |
| 317 | Brian Shouse | 308.8 | 22% | 56% | 22% | .293 | .324 | -0.03084 |
| 320 | Jason Isringhausen | 406.6 | 19% | 47% | 34% | .266 | .299 | -0.03295 |
| Grand Total | 299093 | 20% | 44% | 36% | .301 | .303 | -0.00185 |
Looking at the table, very few Rays that have been with the team for much of these innings fared well. This didn't suprise me when looking back at the defense the Devil Rays had for so many years. Unfortunately, we don't know how much this (or the turf) plays on a pitcher's "hittability." Sonnanstine doesn't fare well, but neither does Kaz (and nooone talks about him being "hittable"). It will be interesting to see how the improved team defense has a role in helping these pitchers out, and if Sonny keeps being "hittable."
I have a ton of data, so feel free to ask any questions you may have.
This post was written by a member of the DRaysBay community and does not necessarily express the views or opinions of DRaysBay staff.
35 comments
|
3 recs |
Do you like this story?
Comments
So this shows that Sonny is hittable?
Although I think the word hittable is clearly a bit misused. Either way this sort of validates the claim that hitters essentially make better contact against Sonnanstine. Clearly the same is true with Kazmir to a lesser extent. Sure it isn’t as evident to the naked eye, but numbers do not lie.
Plus factor in a lot of Sonnys IP is during the good defense time whereas Kaz had a stint with some horrible defenses.
That was one thought I had as well.
How much of Kaz’s was due to his having a terrible defense behind him? >1/2 of Sony’s were in that circumstance, so it’s hard to say. Either way there is so much noise in all this that it makes it nearly impossible to put too much faith in it. Maybe in 10 years when we can use 15+ years of data when guys have played for multiple teams in multiple statiums with multiple defenses it might make a difference.
When I think of "hittability"
this analysis a big part of what I would consider. However, I’ve also been assuming strikeouts play a significant role. I don’t consider Kazmir (old Kaz) nearly as hittable as Sonny. I think a big reason for that is their BABIPs may be similar but Kaz strikes out far more batter then Sonny and that plays into hittability as well.
I defs agree.
But when people on this site call Sonny “hittable” they use it as a way to say that’s why he consistantly underperforms his FIP as compared to his ERA. Kaz is hurt a lot less by his relative “hittability” because he has much fewer BIP generally (old Kaz obvi).
When I talk about "hittability"
I mean in the sense that hitters generally make stronger contact versus him than the average pitcher.
103
Player IP H BIP BIP LD GB FB LD% GB% FB% BABIP xBABIP Diff
Josh Beckett 103 1162.1 945 3120 610 1376 1134 20% 44% 36% .303 .299 0.00375
I meant like, # wise.
Like with Sonny he’s 2.
And same question with Felix.
by R.J. Anderson on Jun 5, 2009 3:42 PM EDT up reply actions
Can you post the top 10.
Just the rank/player name?
by R.J. Anderson on Jun 5, 2009 3:43 PM EDT up reply actions
I'll do you 10 better:
Player
1 Boof Bonser
2 Andy Sonnanstine
3 Sergio Mitre
4 Scott Kazmir
5 Ian Snell
6 Jorge de la Rosa
7 Dennys Reyes
8 Brandon Claussen
9 Edwin Jackson
10 Mike Wood
11 Felix Hernandez
12 Matt Belisle
13 Kelvim Escobar
14 Kyle Davies
15 Jaret Wright
16 Matt Garza
17 Adam Bernero
18 Kameron Loe
19 Zach Duke
20 Curt Schilling
There's a few really, really bad pitchers, but almost as many good ones.
by R.J. Anderson on Jun 5, 2009 4:52 PM EDT up reply actions
this
11 to 9: bad pitchers to good. Give or take a pitcher or two. Also, I think Zach Duke is good… go figure
Which is why BABIP doesn't quite work right
If we’re concerned about outs, we should be counting strikeouts as well. It should be straight batting average (but without removing sacrifices), I think, and a 4th column for K%.
Bad Left Hook - The SB Nation boxing blog
"Baseball is played on the field, not on a calculator."
BABIP does work as a measure of luck and predictive value.
Does it work as a talent indicator? Absolutely not, and that’s not its intended purpose.
Pitcher BABIP is a lot more luck driven than hitter BABIP.
by R.J. Anderson on Jun 5, 2009 7:34 PM EDT up reply actions
Absolutely
But it’s not completely luck driven.
Bad Left Hook - The SB Nation boxing blog
"Baseball is played on the field, not on a calculator."
I still suspect
That you’d see more of a difference between good and bad pitchers if you only checked these numbers against high batting average hitters.
Bad Left Hook - The SB Nation boxing blog
"Baseball is played on the field, not on a calculator."
Excuse me
Not good and bad pitchers, but more hittable and less hittable pitchers. There are plenty of good pitchers who are hittable and bad ones who aren’t.
Bad Left Hook - The SB Nation boxing blog
"Baseball is played on the field, not on a calculator."
in this study its a bad thing
Data goes back years, working on a team BABIP vs UZR comparison. Using that to refresh some old excel skills before i tackle some indiv stuff. This could bug me for awhile
Follow Me on Twitter @FreeZorilla
If you have any excel q's and need to shake off some rust let me know.
I can’t access my gmail account from work, but I’ll be around here for the next hour or so until I leave.
Also
I think I did something similar w/ GIDP last year, and it was pretty cool.
One thing that might be a nice addition is do breakdowns by OF/IF, COF/CF, MIF/COF, and then run some regressions. Maybe compare it to pitcher BB-types or something. Good luck.
Good stuff
I’d like to see swinging strike rate thrown in there as well though, since that’s a separate measure that might be a proxy for hittability. I mean, just because balls are hit in play more often, doesn’t mean someone’s more hittable if they’re striking guys out at a higher rate. Maybe it should just be BA/xBA (as opposed to BABIP, but excluding the effect of sacrifices) with K% thrown in there as the additional column.
Bad Left Hook - The SB Nation boxing blog
"Baseball is played on the field, not on a calculator."
Also
It’s a bit surprising that so many Rays are so high, considering team BABIP was only around .290 last year. FWIW, Kazmir historically has had a high BABIP relative to his teammates, but a lot of that is because of his demeanor. Nolan Ryan had a high BABIP too. High-K/High-BB pitchers usually have higher relative BABIPs for whatever reason. Not sure why though.
Bad Left Hook - The SB Nation boxing blog
"Baseball is played on the field, not on a calculator."
That's why I said "relative to his teammates"
Obviously there’s variation, but there shouldn’t be tons of it.
Bad Left Hook - The SB Nation boxing blog
"Baseball is played on the field, not on a calculator."
Or do you mean guys who allow fewer defensive plays overall
Get the defense cold, so they don’t perform as well?
Bad Left Hook - The SB Nation boxing blog
"Baseball is played on the field, not on a calculator."
Guys who allow less balls into plays have higher chances at variation.
That’s where the whole 1,500 rule comes into play. R/R+1500 is the formula to figure out how much you regress.
by R.J. Anderson on Jun 6, 2009 6:24 PM EDT up reply actions
That is true
But, while I could be wrong, I thought I remembered seeing something somewhere saying that high-BB high-K pitchers have higher BABIPs generally, regardless of higher variation. Over a large sample, it would imply there’s a reason for that, either that defenders aren’t as sharp, or the pitching style means that when they miss a pitch it’s more likely to end up over the plate, or something.
Bad Left Hook - The SB Nation boxing blog
"Baseball is played on the field, not on a calculator."
Comments For This Post Are Closed
Secondary Sidebar
User Tools
Please read our Community Guidelines.
FanPosts
Community blog posts and discussion.
Recommended FanPosts
Recent FanPosts
DRB Suggestion Box
Certain photos copyright © 2012 by Associated Press or Getty Images. Any commercial use or distribution without the express written consent of Associated Press and Getty Images is strictly prohibited.
