Layering the PITCHf/x Leaderboards
Continuing my series (Part 1) looking at detailed pitcher stats made possible by PITCHf/x data, today I'm going to look at RV100 data by strike zone layer. Remember, RV100 stands for Run Value per 100 pitches and is compared to average.
What's a layer?
Think of the strike zone and surrounding space as a cake. Chocolate cake, angel food cake, whatever kind of cake you like. When you move between layers of a cake, you move up and down. That's opposed to slices of a cake, where you move side to side between slices. Layers pay no regard to the lateral location of the pitch, and slices ignore the vertical location. Here's a diagram for the five layers I use:
"Middle" pitches are the middle third of a hitter's strike zone. "Up" and "Down" are the top and bottom thirds of the strike zone, respectively, creating three equal layers. "High" and "Low" are above and below the strike zone, respectively.
The overall top and bottom of the strike zone for each hitter is defined by the average values across all plate appearances by that hitter. Aggregating their zones evens out oddities that arise from the top and bottom boundaries being manually set by human PITCHf/x operators.
Here are the baseline RV100 numbers for each layer, found by combining all PITCHf/x data going back to 2006. I mean everything, including the WBC, Futures Games, Exhibition, Playoffs and Regular Season. Lower numbers are better for pitchers.
High 1.84
Up -2.46
Middle -2.75
Down -1.76
Low 1.79
Here are the best and worst pitchers by strike zone layer, based on 2009 regular season games through May 14. A minimum 71 pitches was needed to qualify for a layer.
|
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
0 recs |
14 comments
|
Comments
I wonder if Banny knows how good his run value is on belt-high pitches?
If there is any pitcher out there that might, it would be Banny.
by Jeff Zimmerman (TucsonRoyal) on May 18, 2009 11:32 AM EDT reply actions 0 recs
not to mention opposing hitters in 2008
I'm not a sabermetrician, but I do play one at Driveline Mechanics.
by devil_fingers on May 18, 2009 2:09 PM EDT up reply actions 0 recs
Odd.
For as many “hanging” knuckleballs that get hit off of Wake, you’d think that he’d be a better pitcher when the ball drops like crazy. I’m probably missing something here though.
by bs.uf15bosox9bears23 on May 18, 2009 4:35 PM EDT reply actions 0 recs
Not saying you're missing it, but remember that it's not just batted balls that get counted here.
For RV100, small changes in run expectancy caused by changes in the ball-strike count are included.
So for Wakefield, maybe when he misses low, he misses really low and players can lay off it? Maybe hitters only really swing when the ball looks like it’s over the heart of the plate, and if that’s true, it won’t move enough to be well low? Just musing out loud…
Beyond the Boxscore // Calling BJ Upton lazy is lazy.
by Sky Kalkman on May 18, 2009 4:42 PM EDT up reply actions 0 recs
Oh...
Now I get it. Thanks Sky.
by bs.uf15bosox9bears23 on May 18, 2009 6:31 PM EDT up reply actions 0 recs
Next time ...
I’ll include stuff that will help with the “how does that number come about?”. Like called ball/strike rates, whiffs, slugging rate on contact etc. Also GB/FB/LD rates if you all behave yourselves. Either that or we get 27 hours a day for a couple days.
by Harry Pavlidis on May 18, 2009 9:49 PM EDT up reply actions 0 recs
Is that a threat...
or a promise?
by bs.uf15bosox9bears23 on May 19, 2009 4:22 PM EDT up reply actions 0 recs
A promise to be more threatening?
by Harry Pavlidis on May 19, 2009 7:12 PM EDT up reply actions 0 recs
I am suprised Wellemeyer is good at anything
Watching him pitching recently has been depressing. He has been over throwing his fastball and it has been going up. So maybe that has something to do with it. The batter assumes it will just be a ball but occasionally he gets a strike.
Stat Whore
by FlimtotheFlam on May 19, 2009 1:53 PM EDT reply actions 0 recs
How real are these numbers?
Have you seen run value by slice or layer be consistent for a given pitcher across time?
Definitely interesting stuff. I can’t believe Banny is almost a full win better than average belt high.
by Dan Turkenkopf on May 20, 2009 8:05 AM EDT reply actions 0 recs
I would also like to see how these numbers look from year to year.
Very interesting stuff though.
by lookatthosetwins on May 20, 2009 1:51 PM EDT up reply actions 0 recs
I really wouldn't want to go there yet with 2 1/4 years worth of data.
by Jeff Zimmerman (TucsonRoyal) on May 20, 2009 2:59 PM EDT up reply actions 0 recs
Good points all around
First, I’m not to excited about the pollution of batted ball data in here. I noticed some of the slice/layer leaders have unusually low BABIP. Based on that alone, I’d predict some instability in the measures when you slice them up. Is that anything more than the result of narrowing the sample, assuming that’s what happens, I dunno. I probably won’t look at that question until I work out the component based approach.
by Harry Pavlidis on May 20, 2009 3:38 PM EDT up reply actions 0 recs

by 














BtB on Facebook














