Improving Plate Discipline From Longoria
Editor Note: Too good for the side-bar. RZ's graphs are amazing as always.
I got the idea for this article from Tommy's post on Longoria. Looking at FanGraphs data, the only major changes in Longoria's plate discipline from his first two season and this season is that his swing rate has gone down a tad and his contact rate is up to the highest of his career. These two stats will likely trend one way or the other as he continues on with his career, more so in his early seasons as continues to improve his batting skills.
I then made a couple contour graphs to visualize his changing contact rate and swinging rate.
First up, swinging rate. In this chart, the solid line denotes Longoria's 55% swinging rate (AKA the location where he's swung at 55% of the pitches thrown) and the dotted line marks his 40% rate, which is closer to his overall career average.
Longo is clearly laying off the high stuff much better this season while remaining the same on pitches away.
Next, his contact rate. Solid line represents an 85% contact rate (AKA where he's made contact on 85% of his swings) while the dotted represents 70%.
Longoria's 85% contact rate for this season seems to be centered in the pitch f/x strike zone. He's making less contact on the inside pitches, but he's also swung at less pitches located there (as seen in the first graph).
In the contact rate graph, the contour lines look like the swing path of a right-handed hitter. But then if you look at the first graph, the swing contour lines look like the reverse. Could this be normal, or is it something unique to Longoria?
I was going to do this by count but the samples for this season were pretty small.
This post was written by a member of the DRaysBay community and does not necessarily express the views or opinions of DRaysBay staff.
5 comments
|
6 recs |
Do you like this story?
Comments
Maybe it is proprietary, but can you give a little info
on how you come up with the circles. For the top graph, you call one of the top lines a 55% swinging rate. I assume that means that 55% of his swings occur in that circle. Okay, I’m sold so far, but how do you come up with the shape of the graph. What I mean is, it seems the shape of the graph could be almost anything if you select an arbitrary 55%.
Say he swings a 5 pitches in the strike zone and 5 pitches out of the strike zone. The graph could be a circle in the strike zone, or could probably also be a graph of the top half of the zone and several pitches out of the zone, or some such.
I guess I’m asking how you make the shape. Is it a "start in the middle of the circle and just start moving until 55% is grabbed?
nothing in this is done manually.
The contour line is plotted from a local regression analysis. I don’t know how the process of the regression analysis works exactly, but basically I created a formula that finds out the likelihood of, in the case of the top graph, of a swing based on the plate location data from pitch f/x for Evan Longoria. So a 55% contour line should be the line where anything inside is likely to be a swing 55% of the time by Longoria.
The number 55% is arbitrary. Most of the graphs I make I have to play around with the numbers to get a good looking graph. In this case, there was no contour line for Longoria 60% or up which would be really high compared to his total swing rate this season which is roughly around 45% right now.
The shape varies depending on the data and the settings to the local regression analysis.
Fuzz
View perspective
Are we looking at this from the view of the pitcher or from the view of the catcher/umpire? For instance, is -1 the inside or outside corner of the plate?
always from the catcher's perspective unless noted otherwise from my graphs
The negative numbers is to the left of the middle of the plate like a batter facing towards the pitcher. This is in feet and the called strike zone as determined by various experts is about a foot each side from the middle of the plate for a right-handed hitter. So -1 is a foot inside from the middle of the plate and the actual edge of home plate is around + or – .708 feet on the chart.
Fuzz
by RZ on May 25, 2010 3:17 PM EDT up reply actions

by 























