THE BOOK--Playing The Percentages In Baseball

Saturday, April 28, 2012

Competitive advantage for the Giants with FIELDf/x?

I’d be interested to hear who the Giants have to mining and analyzing the data, and what they’ve been able to find. 

Thursday, April 28, 2011

Out rate by location and hang time

I love anything with hang time because it makes it quite clear how determinant it is of the out rate.  Dudek’s initial article way back in the first Hardball Times Annual with limited number of games is all that we needed to know how powerful hang time is.  BIS is finally publishing snippets of it, which we see here courtesy of Mark Simon’s article at ESPN (with an appearance from Ben):

Wednesday, March 16, 2011

Tracking players’ movement

I’ve seen this for soccer, in an automated fashion.  I don’t remember what system is used in soccer (trackman?).  Anyway, this enterprising fellow is doing this by hand in hockey.  When FIELDf/x rolls out, you will see charts like this as well, especially for outfielders, but also for infielders with guys on 1B.

Monday, December 13, 2010

Preparing for FIELDf/x

Albert gives us one scenario:

You can already start imagining how FIELDf/x can inform the Red Sox about their outfield. For instance, let’s look at Crawford’s case. With a full FIELDf/x database on a server dynamic enough to display timed animation of fielding routes, we can draw a baseline for True Defensive Range (coined by Greg Rybarczyk). Looking at away teams, we can estimate how a leftfielder’s TDR is affected when playing at Fenway versus when playing at other ballparks. Distribution curves of varying TDR values can be plotted for leftfielders at Fenway versus that of leftfielders at other ballparks, giving us an idea of how Crawford’s TDR should be affected by Fenway if we look at his TDR comparables.

For Drew, we can improve our assessment of arm strength by accounting for 1st-to-3rd baserunning situations and runners thrown out at the plate while accurately categorizing every throw a right fielder makes at any ballpark. Those situations can be compared with would-be 1st-to-3rd baserunning situations that are hit to left field instead (where the runner at 1st doesn’t think twice about passing 2nd). In whatever method is chosen, throwing ability can be more accurately assessed, whether it’s comparing its value in LF and RF at Fenway or comparing rightfielders across baseball. The theory is then that Crawford’s plus-plus range and plus arm versus Drew’s plus range and plus-plus arm can be quantitatively assessed in order to figure out who is a better fit in LF or RF at Fenway.

This seems like a lot of work in order to make one decision, deciding on switching two corner outfielders who are both already good at defense. However, what can be taken away from this FIELDf/x thought experiment is the breadth of questions that FIELDf/x can be utilized to answer — if assembled and analyzed correctly.

Monday, November 29, 2010

Gameday slice bias

How much is the hit location biased based on where a fielder is normally positioned?  The following is a starting point, and as a result, will use a crude estimate.

From 1B to 3B line is 90 degrees.  The circumference of a circle is 2PI*r, or PI*r/2 for a quarter circle.  If we treat the radius as around 115 feet (meaning a spot somewhere between the 1B/3B bags and the 2B bag), then the distance from 1B to 3B bags, along a circular path is 180 feet.  Or, 1 degree = 2 feet.  I know it’s not a circle, but, we just need crude approximations.

Also remember that a SS / 2B are positioned around -16 / +16 degrees, where 0 degrees is 2B. 

Peter provided us with this data from HITf/x (FX) and GameDay (GD), based on spray angles in 4 degree slices:

rate is FX / GD.

We see that at -16 (meaning -14 to -18 degrees), groundballs are recorded by Gameday far more frequently than HITf/x is recording.  We see an enormous bias in the holes as well.

Now, let’s try an experiment.  Let’s say that the Gameday scorers agree with HITf/x perfectly on half of the batted ball locations, and are off by 4 degrees (8 feet) on the other half?  Let’s start with the 314 balls up the middle (-2 to +2 degrees).  Gameday marks 157 of those up the middle, and for the other 157 record half to the left (-6 to -2) and half to the right (+2 to +6).

At the -4 degrees (-6 to -2), HitF/x had 329 balls, of which half Gameday agrees with, and the other half are all more toward the SS side (at -8 degrees). 

So, this is what we have at -4 degrees as per Gameday:
164.5 balls that HITf/x marked at -4 degrees
78.5 balls that HITf/x marked at 0 degrees

That’s a total of 243 balls marked by Gameday at -4 degrees under this illustration, compared to the 329 actally recorded by HITf/x… but still a far away from the 204 actually recorded by Gameday.

The same thing happens at -8 degrees: of the 328, 164 are properly marked by Gameday, the rest are marked toward the SS side (at -12 degrees).  And we go on until we get to -20 degrees, where the shift happens toward the 2B bag.  This is the result of all that:

My illustration here shows that my model bridges some of the gap.  The standard deviation of the original FX/GD is 42%, while the Tango/GD is 34%.

And trying different inputs didn’t make much better difference.  If I treat anything between 50% and 75% of the HITf/x data as being perfectly recorded by Gameday, the remain balls in play are about 4 degrees (about 8 feet) biased toward where the fielder is positioned.

I think we can try to construct a more elaborate model, and we’ll probably end up at the following: about half the data from Gameday will match HITf/x, and the other half will be off by 2 to 8 degrees (4 to 16 feet).  The amount it will be off will be biased by either where a fielder is normally positioned, or whether a play was made or not, or how much space between fielders (the holes).

This is the framework I’m proposing.  Implementations will vary.

Tuesday, October 26, 2010

Fielder positioning

You could do alot with $1 stopwatch (hangtime) as we know.

Now, this is what you can do with a 50$ camera (positioning):

Wonderful, isn’t it?  It would be beneficial to know how a player’s fielding value breaks down between his positioning (of which he may or may not control, and if you see him play the rover position, you are inclined to believe it’s manager-influences), and his legs/arms.  Furthermore, there’s simply the question of regression, which is why we’d like to know how much of a hitter’s skill comes from BB, SO, HR and how much from BABIP.

I know we’ll get all this positioning from the FIELDf/x jackhammer, but we should have been getting this from the scorer’s hammer for decades.  (Hire more scorers.)

Saturday, October 02, 2010

Why we want hang time

Dudek’s article should have been enough.  Well, Greg, makes the point graphically even clearer. 

Go to the 15 minute mark of the presentation.  (You select the presentation on the right side.)

That’s EXACTLY why you want hang time.  Greg shows the “landing point of ball from starting point of fielder” on one axis, and hang time on the other.  This is so obvious, it makes you wonder why hang time (and initial fielder position) is not tracked for 20 years now. The NHL tracks who is on the ice every second (the NHL employs multiple scorers every game… 3 or 5 of them or something, and they are not swimming in money like MLBAM).  We can’t get an extra stringer to show where a fielder is positioned with a stop watch in his hand?  The 30 MLB teams are not demanding this from their subsidiary?

Anyway, his chart shows the obvious and expected clumping of hits and outs.

Great job from Greg in laying out the foundation here as to how things work with UZR and why it works like that, and what the uncertainties you have based on not having hang time. 

Saturday, August 28, 2010

PITCHf/x Summit 2010 - Recaps

I’ll update links as they come in.  First up is Colin:
http://www.baseballprospectus.com/article.php?articleid=11868

Ben:
http://www.baseballprospectus.com/article.php?articleid=11869

Dave:
http://baseballanalysts.com/archives/2010/09/pitchfx_2010_su.php

Friday, February 12, 2010

FIELDf/x

Nothing new here except for one little tidbit, but the picture is so cool:

“After an amazing catch by an outfielder, we can compare his speed and route to the ball with our database and show the TV audience that this player performed so well that 80 percent of the league couldn’t have made that catch,” says Ryan Zander, Sportvision’s manager of baseball products.