THE BOOK--Playing The Percentages In Baseball

Monday, July 26, 2010

Ubaldo = Regression, using PITCHf/x

Ricky: .247 wOBA on contacted PA in his early season run, and .378 on his current run.

Thursday, July 15, 2010

Fastball speeds at the ASG: average of 96mph!

Whoah:

Of the 272 pitches thrown, 190, or 70 percent, were fastballs. The average fastball speed in the game was 96 mph. Wow.
...
The average fastball speed boost relative to the regular season was 1.5 mph for starters and 0.4 mph for relievers.

Wednesday, July 14, 2010

wOBA by Strike Zone location

Harry at WSJ with more talk at BtB.

Wednesday, July 07, 2010

PITCHf/x Summit 2010

http://baseball.sportvision.com/summit

David Wright’s swing zones

Just beautiful:

The colors indicate run values, so closer to red the better, and the contours show Wright’s swing tendencies. He swings at 75% of pitches within the smaller circle, and 50% within the larger one, in other words.

Stolen Bases and PITCHf/x

Great stuff from Jeremy.

It’ll feel good to retire one day, knowing there’s some a great group of talented and hardworking analysts out there, who have jobs where they can sneak in all this great work. 

The internet is as much about adding value to corporate america’s efficiency as it is about removing value by surfing and posting.

Tuesday, July 06, 2010

Using PITCHf/x to see who crouches

Love it.

Wednesday, June 23, 2010

Strasburg’s PITCHf/x after 5 games

Great stuff from Mike, especially this:

To RHB, he’s targetting the ball down, below the knees.  To LHB, he’s targetting either down, or away (along his pitching-hand).  A substantial portion of these changeups are outside the strike zone.  And yet the batters can’t help themselves. 

If I had to guess, batters are used to seeing 91mph pitches as fastballs or sinkers, and so, they expect a ball that is coming in mid-to-low to remain somewhere above the knees.  But Strasburg 91mph is a changeup, which means it has extra drop on it (18 inches, rather than 6 or 12 for Strasburg’s fastball / sinker).  And so, the batters can’t lay off it.  Yet.

I’d like to see a similar chart for Lincecum, if Mike is up for it.  And, if you are looking for extra work, how batters respond to Lincecum’s changeup the second game and third game they face him.  That is, do they learn?

I’m also surprised to the extent that his velocity is down throughout the game.  Mike, two things: is his location also off?  And does his movement also diminish?

Tuesday, June 22, 2010

Scientific Baseall

A standard Brian Bannister article, but then this is a new one for me:

However, there is one sign of growth at the lower levels at an Oklahoma City training facility.

Glenn “Butch” Schoenhals, a retired neurosurgeon and longtime amateur coach, is the owner and president of Scientific Baseball, which operates the only PITCHf/x system outside a major league ballpark. He’s using the system to teach players as young as 9, and even to train umpires.

“This simplifies the process,” Schoenhals says. “It converts what we think to what we know.”

“That’s great,” Bannister says. “This is where I foresee all this going.”

Monday, June 21, 2010

SABR 40 - Alan Nathan, Dave Allen, Josh Kalk, et al

For those in the Atlanta area on Aug 7:

3:00 p.m. New Technologies in Baseball, panel moderated by Alan Nathan

Alan Nathan moderates a discussion of the latest developments in Sportsvision’s PITCHf/x, HITf/x and FIELDf/x, and TrackMan’s radar technology used to measure ball flight.

Dave Allen is an expert in spatial statistics and graphical analysis. He is a staff writer for Fangraphs and Baseball Analysts. He will talk about PITCHf/x analysis.

Josh Kalk is the Baseball Operations Analyst for the Tampa Bay Rays. Prior to that he was one of the leading PITCHf/x analysts and a writer for The Hardball Times.

Greg Moore is in charge of marketing for Sportvision’s baseball products, including PITCHf/x, HITf/x, and FIELDf/x.

Rob Ristagno is Director of Business Development for TrackMan. He seeks new markets in which to apply the TrackMan radar technology, including professional baseball.

Alan Nathan is an expert in the physics of baseball, with experience using both PITCHf/x and Trackman for trajectory analysis.

Thursday, June 10, 2010

Figuring out command

I’’m glad there are others also thinking about this:

With fastballs, you either go high heat or throw at the knees. With sliders, there’s back foot or back door. Curves are intended to be thrown either anywhere in the dirt or anywhere in the zone. Anyway, those are the assumptions you need to make if you believe clustering makes sense. Furthermore, if you’re limited to k-means clustering, you might as well assume that all pitchers have two intended locations for their fastballs. That’s what I did, anyway. So I gave each pitcher his own two separate cluster centers, and found each pitch’s standard deviation from those centers, grouping by pitcher. Here were the leaders:

I’m not sure about a fixed “2”, but it’s a good start.

Monday, June 07, 2010

PITCHf/x, R, and XML Package

Here you go.

Glove-slap: Tommy

Wednesday, June 02, 2010

The genius of Brian Bannister

I work with a pitch f/x system to try and get as much sink as I possibly can on my pitches with my arm-slot and my talent, and hopefully get that groundball rate up around 50%, keep working on expanding my strikeout to walk ratio and keep my homeruns down. That’s my gameplan. I’m shooting for a low-4 ERA/FIP, and beyond that I just hope I get lucky in a given year.... I use brooksbaseball.net--I’ll throw pitches in a game every now and then in a low key, low-leverage situation where nobody in the world is paying attention to what I’m doing; I’ll try a new grip, throw it. After the game, [I] check it to see what it registered on the pitch f/x, see if it was better or worse, and I’ll work off of that.

Brian Bannister on BP Radio (MP3 podcast on May 28)

Glove-slap: Jeff.

Launch angle + Speed off the bat = trajectory

Courtesy of Greg (click to make bigger):

Tuesday, June 01, 2010

Standards for PITCHf/x viewpoint: catcher or pitcher?

HITf/x will undoubtedly be from the hitter’s perspective.  For that reason alone, I’d like to see PITCHf/x charts from the analysts from the hitter’s viewpoint.  Not to mention that we sometimes see the charts as it relates to catchers and umpires, and those charts are from their vantage point as well.

Given that the choice for the PITCHf/x charts is rather arbitrary, am I alone in thinking that I’d prefer that they match the eventual HITf/x, and that the vantage point of the hitter, catcher, and umpire is preferred to the pitcher? Even for stuff like FIELDf/x, we’re not going to see the data from the fielders’ perspective, right? 

So, make standing at home plate be the vantage point for all these charts.

(Someone is obviously thinking:who am I to impose my will on anyone.  Duly noted.)

Sunday, May 30, 2010

Perfect Game, PITCHf/x style

ESPN notes:

Halladay got 26 called strikes in this game and struck out six batters looking. Using Pitch F/X technology, a technology set up in conjunction with MLB, we had Doug Kern of ESPN Stats & Information look at whether those strikes were really strikes.

Of the 26 called strikes, seven were, via Pitch F/X standards, out of the strike zone, or close enough that they would be “borderline strikes.”

There was a pattern:

Six of these called strikes were off the plate to the left side (looking in from the pitcher’s view). But more notably:

Of Halladay’s six strikeouts, five came on pitches that were not in the Pitch F/X strike zone.

* Strikeout pitch to Chris Coghlan in the first inning was inside.

* 2-1 pitch to Josh Johnson in the third was low.

* Strikeout pitch to Hanley Ramirez in the fourth was outside.

* First pitch to Johnson in the sixth was inside.

* Strikeout pitch to Coghlan in the seventh was inside.

* Strikeout pitch to Ramirez in the seventh was outside.

* Strikeout pitch to Wes Helms in the ninth was outside.

Eric Polsky gets into it as well, using Brooks Baseball.

Ideally, the comparisons would be made against Halladay’s strike zone, and the Mike DiMuro’s strike zone.

Tuesday, May 25, 2010

Trading speed for location

I can’t see all the charts, and this article is not about the actual trading, but comparing those who rely on speed vs those who rely on location.  Location, speed, movement.

Friday, May 21, 2010

Gameday Mapping XY Coordinates

Something interesting about the hit location mappings on MLB.com for the pixels.  A pixel is granular, an integer.  There is no such things as 0.24 pixels.  When an entry is plotted, it should be marked at coordinate 125,160 or whatever.  It can’t be marked at coordinate 125.64 and 160.42.  But, that is how it is being marked.

More strangely is the distribution of how often those decimals occur.  They should be somewhat random and uniform.  Now, seeing that we shouldn’t have those decimals to begin with, there must be some conversion happening in the interim (pixel to something to pixel).  And in that conversion process, decimals appear.  And they appear in a non-random systematic way.  Here is the frequency of how often they occur:

You should note two things:
1. How much the x and y mirror each other
2. How there is almost no data points below .10 or above .90.

And, the pairing of the X and Y are not-random.  For example, here is how often the X decimal value appears for Y value decimal of .67:

Read More

Wednesday, May 19, 2010

Scouting-data for forecasting

Using PITCHf/x:

+------------------+-----------+
| Player | Sim score |
+------------------+-----------+
| Scott Richmond | 0.90 |
| Zack Greinke | 0.90 |
| Ricky Nolasco | 0.89 |
| Joba Chamberlain | 0.89 |
| Matt Garza | 0.88 |
| Takashi Saito | 0.88 |
| Brett Tomko | 0.88 |
| Kevin Gregg | 0.87 |
| Seth McClung | 0.87 |
| Sean Gallagher | 0.87 |
+------------------+-----------+

Friday, May 14, 2010

Brooks… Doctor Brooks

Congratulations to Dan Brooks, baseball analyst and PITCHf/x maven, on successfully defending his Ph.D. dissertation in psychology at the University of Iowa!