Jeff Sullivan featured at Baseballanalysts.com

by vj on Jan 19, 2006 11:44 AM PST

Since Jeff seems to humble, here's a plug for his article over at the Baseballanalysts site:
"A Quantitative Approach to Studying Release Point Consistency" features the analysis by screenshots, that we are familiar with.

Here's the link:

http://baseballanalysts.com/archives/2006/01/a_quantitative_1.php

Tweet Comment 32 comments | 0 recs |

Do you like this story?

Comments

Heh.

You're confusing "humility" with "was up in Los Angeles and away from computers for a day and a half."

by Jeff Sullivan on Jan 19, 2006 12:53 PM PST reply actions

Man....

you're piece really brought out a couple of the big names. I think tangotiger and Will Carroll are going to rumble ;-)

by kenshin1 @ Lookout Landing on Jan 19, 2006 1:32 PM PST reply actions

<sigh>

that should read "your piece"

by kenshin1 @ Lookout Landing on Jan 19, 2006 1:33 PM PST up reply actions

Interesting...

Will Carroll's second comment notwithstanding, this caught my eye:

Jeff's attempts at "open sourcing" this info are nice, but ignore the fact that others have done this better elsewhere. He also ignores the research that shows that release point itself is all but irrelevant, rendering any of this useless.

I'm not sure what your relationship with Will is, Jeff, but I'm curious to know if you knew of this research. I read Will's comment (however harsh it is) and take it that he's assuming you a) knew of both the release point and the Dartfish research in the first place and b) were actively ignoring those two sets of data. I'm assuming that the whole point of this study was because you didn't.

Either way, that's a pretty lofty, snooty comment from Will. Tangotiger's damn straight in putting Will in his place.

Go M's! Mariners Morsels (aka Free George Sherrill).

by PositivePaul on Jan 19, 2006 1:55 PM PST up reply actions

Heh

Yeah, I agree.

Good article, surprising snootiness from Will Carroll. Either way, good article. :) I always enjoy reading your pitching analysis posts, so it's good to see them getting out there in the world.

Marinerds - a different daily dose of baseblog.

by Deanna on Jan 19, 2006 2:03 PM PST up reply actions

As I told Will in an email...

I have to plead total ignorance on the Dartfish system and the ASMI research. Any overlap was completely unintentional.

by Jeff Sullivan on Jan 19, 2006 3:25 PM PST up reply actions

What do....

you make of "uselessness" of release point info. I am by no means an expert of pitching mechanics (in fact I am pretty much completely ignorant) but that seems counter-intuitive.

by kenshin1 @ Lookout Landing on Jan 19, 2006 3:51 PM PST up reply actions

Re: What do....

That is something that I'd love to read up on.

by Jeff Sullivan on Jan 19, 2006 4:01 PM PST up reply actions

the most on-topic research

I can find at ASMI's site doesn't exactly help.
They've got a one paragraph summary of a study on changes in pitching biomechanics within a game (seems pretty analogous). This is basically tangotiger's point: this stuff isn't available. You've got a brief note about how their looking at biomechanics at release (though, to be fair, not at arm placement at release), so it's not like we can conclude much. This is either a big research topic through the specialist journals, or its proprietary (Dartfish) - either way, that's basically irrelevant for fans.
I'm betting Will thought Jeff was trying to market or sell this somehow. That's clearly not Jeff's point; like the pitch-by-pitch data discussed here earlier, this stuff costs a lot and so we regular fans need proxies. Like, say, 'Under the Knife,' is a proxy for ASMI research on recovery times from various injuries. Even if Will thought Jeff was trying to make a fast buck, it came out incredibly dismissive and rude.

by marc w on Jan 19, 2006 4:13 PM PST up reply actions

"Uselessness" of release-point info...

++ What do you make of "uselessness" of release point info. I am by no means an expert of pitching mechanics (in fact I am pretty much completely ignorant) but that seems counter-intuitive. ++

Very counter-intuitive :-)

Just as an example, if an individual pitcher tends to drop his arm as he fatigues, then of course you could track this during the game, and make pitching switches that prevented injuries...

========

About 4 times in 5, when a "published" sabermetrician declares a end-the-discussion dogmatic blanket such as "RESEARCH HAS ABSOLUTELY SHOWN -- X" without sourcing it ... he's bluffing or misinterpreting.

Research has absolutely shown very few things in baseball, other than that if you score more runs on Friday night than you allow, you're going to win.

I'd bet 100:1 that there are all kinds of uses for release-point info, and I'd start with tracking pitchers year-to-year and within games against their own data. Carroll's shameful reply to Sullivan's work is an example of what has a bee in my bonnet about that class of BP-style authors.

B'lee DAT,
Jeff C

by DrDetecto on Jan 20, 2006 11:46 AM PST up reply actions

BTW ...

Even when initial research HAS suggested the null hypothesis, very often you've got a general-vs-specific thing going.

In other words, some amigo might run a study showing that NL-AL crossover players didn't show an "adjustment" year followed by a "bounce" year. But they'll forget that they're measuring a group ... and that an individual player (say, Adrian Beltre) or type of player might indeed show the effect.

Of course, when only 1 or 5 or 12 players show the effect, they'll cross it against the group, declare "sample size" issues and blow off the exceptions to the rule.

In other words, if Mark Prior drops his arm slot 10 pitches before getting lit up, you still might not see the tendency across the group of all MLB pitchers. Does that mean that the info is useless to the Cubs?

=======

As Jeff S is well aware, as an ex-pitcher, the state of pitching is so complex that it may be literally impossible to get our arms around it.

You simply can never isolate the variables. Do 120 pitches (or a dropped arm slot, or a wobbly release point) matter for young pitchers? Left-handed pitchers? Curveball pitchers? Pitchers who have thrown 150 innings that year? 112 pitches that day? Pitchers who have been in a close game that day? Tall pitchers? 94 mph+ pitchers?

The biggest problem with Carroll's general reaction (as to uselessness of release-point info) is that he shows little respect for (1) human intuition (Sullivan zero'ing on Prior's fatigue) and (2) the complexity of the problem in general.

But then, this debate about dogmatism in baseball didn't start this week. :-)

by DrDetecto on Jan 20, 2006 11:56 AM PST up reply actions

Dr D

you should take those comments and send them in a email to Will Carrol. I would like to here his response :)

by Scruffy Lefty on Jan 20, 2006 4:02 PM PST up reply actions

Re: Dr D

Carroll and I spoke today, and he apologized more than once for coming off sounding so harsh. Besides, I never took issue with what he had to say to begin with, so I'd appreciate it if nothing would come of this.

by Jeff Sullivan on Jan 20, 2006 4:41 PM PST up reply actions

Yeah

If he regrets it, fine, end of story. I need about 12 mulligans a day, myself. :-)

Just the same, dogmatism continues to reign as the Orwellian enemy of sabermetrics (or any kind of scholarship). ;-) The entire industry needs a large, and quick, injection of "what don't I know here?" attitude.

Such as the attitude found at Lookout Landing.

Cheers,
Jeff

by DrDetecto on Jan 20, 2006 4:56 PM PST up reply actions

Sure thing, Jeff!

I'm pretty sure I speak for most of us 'round here -- just wanted to let ya know:

We got your back!

Go M's! Mariners Morsels (aka Free George Sherrill).

by PositivePaul on Jan 20, 2006 6:54 PM PST up reply actions

Ya Jeff

you might not realize but I am pretty sure you have started a cult :)

by Scruffy Lefty on Jan 21, 2006 4:42 PM PST up reply actions

WWJD!

Er, wait.

Go M's! Mariners Morsels (aka Free George Sherrill).

by PositivePaul on Jan 21, 2006 9:37 PM PST up reply actions

A cult?

That didn't take nearly as long to accomplish as I thought it would. At this rate, I'll have you guys committing ritual blood sacrifices by March.

by Jeff Sullivan on Jan 21, 2006 10:34 PM PST up reply actions

Can we start with Thorton?

When I was a kid my parents moved around alot.But I always found them. ~Rodney Dangerfield

by Goose on Jan 21, 2006 11:01 PM PST up reply actions

I Wish...

But, then again, Jeff DID lead by example...

Where's my car keys!

Go M's! Mariners Morsels (aka Free George Sherrill).

by PositivePaul on Jan 23, 2006 12:34 PM PST up reply actions

Let's make that a little bigger.

http://images.lookoutlanding.com/images/admin/car.JPG

by Jeff Sullivan on Jan 23, 2006 1:49 PM PST up reply actions

Heh heh....

You DIDN'T go all Arthur Anderson and destroy that evidence :-) I'm sure you tried, but then I caught ya red-handed.

LOL!

Go M's! Mariners Morsels (aka Free George Sherrill).

by PositivePaul on Jan 23, 2006 3:10 PM PST up reply actions

Yeah, we're looking out for you (and landing)

When I grow up, I want to be Jeff Sullivan.

Marinerds - a different daily dose of baseblog.

by Deanna on Jan 22, 2006 2:14 AM PST up reply actions

I must say

that I lost a lot of respect for Carroll with that comment.

by Corco on Jan 19, 2006 7:33 PM PST up reply actions

Am I reading this a lot differently from you guys

or does Carroll basically say "that's nice, but it's been done before and isn't useful" (which is true on the first two parts, the third is opinion) and then he realizes that the studies he was refering to were not open to the general public and he amends his statements for making that assumption?

Am I really missing something here?

by Matthew on Jan 19, 2006 7:53 PM PST reply actions

But

I get the expectation that he expects the average dude blog reader on the street to have access to these things by calling Jeff "not up to speed."

For regular people who can't see top secret stuff, Jeff's work is really cool.

by Corco on Jan 19, 2006 7:59 PM PST up reply actions

And he had that

"I am cooler than you tone"

by Scruffy Lefty on Jan 19, 2006 8:47 PM PST up reply actions

irrelevance

The strangest thing Carroll said is that release point is not important. In and of itself it may not be, but everything a pitcher does leads up to the release point. that makes it an indicator at the very least. until you've studied the pitcher enough, you may not know exactly what it indicates (his release point drops when he gets tired or speeds up his mechanics or flattens his curve ball, say). But that's just a matter of study time. And the above comments are exactly correct: if there's some proprietary analysis along these lines out there, it certainly isn't common knowledge, and jeff's parallel "discovery" of it IS useful. But Carroll is really contradicting himself: 10 clubs are already doing this; and it's irrelevant. Well, why are 10 teams doing it then?

rightly, in every age it is assumed we are witnessing the disappearance of the last traces of paradise... Cioran

by toonprivate on Jan 19, 2006 8:55 PM PST reply actions

It seems

from the ASMI study that there are factors that ARE important at the moment of release, but that, say, lead knee flexion at release is more important than where the hand is. That's interesting, and maybe we need some follow-up there, but all I can find is a paragraph summary of a study. It's shocking, but I don't subscribe to Sports Biomechanics.
I think we need to figure out what signs indicate fatigue and are warning signs for elbow/shoulder damage and then use Jeff's method to look for them. If they're not release point, narrowly defined, then fine - what ARE they, and how do we measure them. Jeff's already got something of a head start in looking at things like the angle of a pitcher's torso at release - that's in the ASMI thing too. Now we just need a way to operationalize that using freely available tv feeds (those behind CF cameras may not be as useful).
I think Jeff's basic method is what's so important - it almost doesn't matter if you're looking at the hand or the knee or whatever. That he used overlaid images and tried to measure variability in them is the real innovation here. I don't give a crap if teams are purchasing studies using 160 high-speed cameras and measuring eleven 'kinematic parameters' at a cost of several thousand dollars. That means very little to anyone reading this blog.

by marc w on Jan 20, 2006 9:46 AM PST up reply actions

Jeff's study is relevant

If I remember my Biology 101 class correctly - and I might not, because that was in 1990 - Darwin and some other dude both came upon this speciation idea at about the same time, independantly.
Who was that other dude?
I can't remember...
and thats the point.
If Jeff finds a way to do it better than anyone else - and he may, since he is obviously coming at it from a different angle, so to speak - or cheaper than anyone else, or more available to the Mariners fans than anyone else, then he will have carved out his own niche.
And that is relevant.

by KC @ Lookout Landing on Jan 19, 2006 9:23 PM PST reply actions

Yeah....

That would be Alfred Russel Wallace, who came up with the idea some time after Darwin's jounrals had been gathering dust, unpublished. As soon as Wallace started contacting some people (Darwin included, IIRC) about it, Darwin rushed to publish his own findings. I think there was co-authorship for some of them, with credit being given to both Darwin and Wallace. Of course, few give him credit these days....

anyway, that was your brief, tangential history lesson of the day.

"Ever tried? Ever failed? No matter. Try again. Fail again. Fail better." -- Samuel Beckett

by JY on Jan 19, 2006 11:20 PM PST up reply actions