14 May 2013

Strikezone Analysis by Pitcher: Volume I

The off day gave me some extra time to go through and take a look at how each pitcher is fairing with the called strikezone, both Real and Typical, as well as borderline pitchers.

These numbers are likely to change as the season progresses. Hammel, Chen, Gonzalez, and Tillman have each seen multiple umpires, so their numbers are somewhat more reliable, although 35 - 50 IP is still a small sample size. Johnson (Jim), O'Day, Patton, Matusz, Strop, Hunter, and McFarland have enough appearances out of the bullpen to have seen quite a few umpires, but with less than 20 IP each, just a few pitches can skew their numbers significantly.

That being said, there are some interesting trends emerging. Wei-yin Chen and Jim Johnson have the most accurately called pitches, while TJ McFarland and Tommy Hunter have some of the least accurate called pitches. But there's more to this than simple accuracy. Chen and Johnson pay for their accurate zones, as their ability to get balls called strikes is much lower than other pitchers. On the other hand, TJ McFarland and Darren O'Day must be dazzling the umpires, because they are earning called strikes on pitches outside of the zone at a much higher rate.

Click through the jump for a more in-depth breakdown.


Real Strikezone
Correct Incorrect In Zone Out Zone BcS ScB
Hammel, Jason 88.68% 11.32% 44.71% 55.29% 10.46% 13.56%
Chen, Wei-yin 92.06% 7.94% 48.75% 51.25% 8.33% 7.02%
Gonzalez, Miguel 88.58% 11.42% 44.83% 55.17% 9.00% 17.95%
Arrieta, Jake 84.86% 15.14% 45.48% 54.52% 9.09% 26.67%
Tillman, Chris 87.47% 12.53% 47.28% 52.72% 9.89% 19.09%
Britton, Zach 92.31% 7.69% 48.51% 51.49% 2.56% 23.08%
Garcia, Freddy 81.52% 18.48% 45.81% 54.19% 18.75% 17.86%
Johnson, Steve 86.79% 13.21% 42.17% 57.83% 12.82% 14.29%
Stinson, Josh 88.64% 11.36% 50.00% 50.00% 14.81% 5.88%
McFarland, TJ 82.44% 17.56% 41.56% 58.44% 17.58% 17.50%
Hunter, Tommy 85.71% 14.29% 49.15% 50.85% 10.48% 23.81%
Burnett, Alex 100.00% 0.00% 60.00% 40.00% 0.00% 0.00%
Clark, Zach 80.77% 19.23% 50.00% 50.00% 17.65% 22.22%
Ayala, Luis 76.00% 24.00% 40.91% 59.09% 20.00% 40.00%
Strop, Pedro 86.81% 13.19% 42.08% 57.92% 12.50% 15.00%
Patton, Troy 88.51% 11.49% 50.57% 49.43% 11.46% 11.54%
Matusz, Brian 87.96% 12.04% 50.24% 49.76% 7.04% 21.62%
O'Day, Darren 86.71% 13.29% 46.39% 53.61% 14.29% 11.11%
Johnson, Jim 92.86% 7.14% 44.32% 55.68% 8.74% 3.92%

Here are some of the Rzone numbers that jump out to me:
  • Jake Arrieta thas thrown 54.52% of his pitches out of the Rzone, only slightly above the team's 53.65% rate. However, Arrieta is absolutely crushed by his 26.67% rate of strikes called balls, well above the team's 11.90%. The overall correlation coefficient between rate of pitches out of the Rzone and rate of strikes called balls is 0.445, which is one of the stronger correlations that I was able to find in this data.
  • TJ McFarland is an enigma wrapped inside of a mystery. He leads the team in throwing pitches out of the Rzone at a 58.44% rate, yet he gets 17.58% of balls called strikes. Could this be due to the extended outside corner of the strikezone for LHB? It doesn't appear so, as McFarland has faced 44 RHb to 27 LHB (he's not a LOOGY). However, he does have a 37.03% K/PA versus LHB, compared to 20.45% versus RHB.
  • Two pitchers with a ton of movement: Jim Johnson and Pedro Strop get very different results on their called pitches. Umpires call Johnson's pitches correctly 92.86% of the time, compared to 86.61% for Strop. Both pitchers throw significantly more pitches out of the zone than in.


Typical Strikezone
Correct Incorrect In Zone Out Zone BcS ScB
Hammel, Jason 87.03% 12.97% 59.74% 40.26% 0.84% 28.65%
Chen, Wei-yin 89.95% 10.05% 61.84% 38.16% 0.93% 22.22%
Gonzalez, Miguel 82.01% 17.99% 62.00% 38.00% 0.64% 38.35%
Arrieta, Jake 82.57% 17.43% 58.63% 41.37% 4.10% 34.38%
Tillman, Chris 85.64% 14.36% 60.38% 39.62% 1.38% 31.52%
Britton, Zach 76.92% 23.08% 61.39% 38.61% 0.00% 52.17%
Garcia, Freddy 84.78% 15.22% 63.23% 36.77% 6.12% 25.58%
Johnson, Steve 90.57% 9.43% 59.04% 40.96% 0.00% 22.73%
Stinson, Josh 93.18% 6.82% 64.47% 35.53% 0.00% 13.04%
McFarland, TJ 82.44% 17.56% 59.26% 40.74% 3.17% 30.88%
Hunter, Tommy 77.55% 22.45% 65.19% 34.81% 3.90% 42.86%
Burnett, Alex 100.00% 0.00% 80.00% 20.00% 0.00% 0.00%
Clark, Zach 92.31% 7.69% 58.33% 41.67% 0.00% 16.67%
Ayala, Luis 84.00% 16.00% 54.55% 45.45% 11.11% 28.57%
Strop, Pedro 86.81% 13.19% 57.92% 42.08% 0.00% 28.79%
Patton, Troy 85.14% 14.86% 64.37% 35.63% 4.00% 26.03%
Matusz, Brian 85.19% 14.81% 62.32% 37.68% 0.00% 32.00%
O'Day, Darren 87.28% 12.72% 61.44% 38.56% 2.22% 24.10%
Johnson, Jim 92.86% 7.14% 56.04% 43.96% 2.25% 13.85%

Here are some of the Tzone numbers that jump out to me:
  • The Tzone starts to provide some evidence for which pitchers are getting squeezed. Tommy Hunter has a 42.86% Tzone strike called ball rate, indicating a fairly tight strikezone when he's pitching.
  • On the other side, Freddy Garcia brought some black magic with him from up north, earning a 6.12% ball called strike rate.


Borderline Pitches
% of
Total
Strikes Balls Tz-Rz
Strikes
Tz-Rz
Balls
Hammel, Jason 12.68% 53.61% 46.39% 44.78% 55.22%
Chen, Wei-yin 4.18% 73.33% 26.67% 41.67% 58.33%
Gonzalez, Miguel 5.78% 57.58% 42.42% 32.73% 67.27%
Arrieta, Jake 9.04% 39.39% 60.61% 38.10% 61.90%
Tillman, Chris 7.07% 56.25% 43.75% 43.64% 56.36%
Britton, Zach 3.96% 25.00% 75.00% 10.00% 90.00%
Garcia, Freddy 10.97% 70.59% 29.41% 60.00% 40.00%
Johnson, Steve 8.43% 71.43% 28.57% 62.50% 37.50%
Stinson, Josh 6.58% 80.00% 20.00% 66.67% 33.33%
McFarland, TJ 9.47% 69.57% 30.43% 50.00% 50.00%
Hunter, Tommy 7.17% 52.38% 47.62% 28.57% 71.43%
Burnett, Alex --- ---  --- ---  --- 
Clark, Zach 13.89% 60.00% 40.00% 100.00% 0.00%
Ayala, Luis 13.64% 66.67% 33.33% 100.00% 0.00%
Strop, Pedro 7.92% 68.42% 31.58% 50.00% 50.00%
Patton, Troy 6.51% 64.71% 35.29% 38.10% 61.90%
Matusz, Brian 6.28% 38.46% 61.54% 38.46% 61.54%
O'Day, Darren 7.21% 73.91% 26.09% 51.72% 48.28%
Johnson, Jim 4.03% 81.82% 18.18% 50.00% 50.00%

Here are some thoughts on borderline pitches:
  • Jake Arrieta was brutally squeezed in his 4 GS in Baltimore. His 60.61% ball rate for borderline pitches was well beyond the team's 40.78% rate. On the other hand, his 61.90% T-R ball rate was only somewhat higher than the team's 56.79% rate on such pitches. Here's the thing, we tend to think of Arrieta as a pitcher lacking control because of his high 18.4% walk rate. But he threw a higher percentage of borderline pitches (9.04%) than the team average (7.57%) and was crushed by the umpires for his trouble.
  • Jason Hammel may have the best command on the team. He's thrown, by far, the most pitches for the Orioles in 2013 and is maintaining an excellent (at least compared to other Orioles pitchers, I don't know about non-Orioles) 12.68% borderline pitch rate. Unfortunately, he does not get as many called strikes on the edges as some other pitchers. Despite this, he still has almost twice as many called borderline strikes (52) as the next Oriole, Tillman (27).
  • Certain Orioles pitchers are classified as having good "stuff", that nebulous combination of velocity. movement, and deception. In this group would be guys like Jim Johnson, Jake Arrieta, and Pedro Strop. On the other side of the coin are the pitchers who lack "stuff" and rely on command to induce weak contact or confuse a hitter into taking a strike. In this group would be guys like TJ McFarland, Steve Johnson, and Miguel Gonzalez. The top 3 pitchers for percentage of pitches thrown within the Tzone, but outside of the Rzone are TJ McFarland (29.86%), Steve Johnson (28.57%), and Miguel Gonzalez (27.68%). At the low end of T-R pitches are Brian Matusz (19.38%), Jim Johnson (20.92%), and Zach Britton (20.97%).

6 comments:

Peter Tebin said...

Great post! Love your blog!

Patrick said...

The majority of the strike zone analysis stuff up til now was interesting in a non interesting kind of way. This article is very interesting, would be cool to see some other teams investigated, not sure if you can get ahold of those data sets.

Lou Proctor said...

It's very easy to get the data for other teams. The problem is the time involved. It takes me about 30 minutes to manually pull the data from Brooks Baseball and plug it in to my Excel for analysis. Without an automatic way of doing this, which I don't have, I simply don't have the time to run the data for other teams.

oriolepaul said...

Question: Does the strike zone box reflect the location of the pitch at the instant that the catcher catches the ball? or the ball's location as it passes over or outside the plate? There are many pitches which pass over the plate but are caught at a location appearing to be off the plate. Please explain.

Lou Proctor said...

PITCH f/x records the location of the pitch as it crosses the plane extending vertically from the front of home plate.

Triple R said...

I read a very interesting article on Grantland today about the effect a catcher framing pitches can have on a team's pitchers. Would you know how good Wieters is with pitch-framing, and how this affects the borderline-call%?