There were a couple comments in the game thread during yesterday's Angel/Yankee game about the two different strike zones; A.J. Burnett getting a wider zone than Scott Kazmir. Was Kazmir getting squeezed while Burnett was benefiting from a huge strike zone?
With this website (BrooksBaseball.net), it's real easy to check out the data from any game. Below are two plots from yesterday's game (9/23/09) showing the location and outcome of both starter's pitches.
A.J. Burnett's pitch data (click on the plot to enlarge):
The only questionable strike calls were on four pitches off the plate (either inside to RH batter or outside to a LH batter). It also looks like the umpire missed a couple low pitches that were called balls which were actually strikes. Otherwise, looks like the home plate umpire did a good job with Burnett, but how does it compare to Kazmir's zone...
Scott Kazmir's pitch data (once again, you'll need to enlarge the plot to see the data more clearly):
When you compare the two pitcher's strike zone, home plate umpire Mike Everitt was very consistent with each pitcher. Kazmir got the same inside strike (to RH batters) that Burnett was getting, and there was only one Kazmir pitch which was within the zone that was called a ball. Overall, Everitt did a great job for both pitchers.
I think this kind of data is pretty fascinating, and while looking at it I noticed something interesting with the location of the pitches for each of Kazmir's hits allowed. Go back to Kazmir's pitch plot and look for the light-blue squares. Those are the pitches which resulted in hits (all from right-handed batters). Each of the hits were approximately a half foot from the center of the plate and between 2.2 and 2.7' above the plate. Roughly the size of a Burger King Whopper. Kazmir might want to stay away from that spot, at least when he pitches against the Yankees next month as it looks like they tend to feast on that location. Also, look at the number of dark red squares (foulballs) Kazmir has on his plot compared to Burnett. There were too many of Kazmir's pitches that were fouled off for me to count compared to Burnett's approximately eight foulballs. The difference between Kazmir's four strikeouts and Burnett's eleven.
0 recs | 20 comments
wow awesome site
I didn’t get to watch the game to know enough from that to complain (or argue in favor of the umpire), but this site helps a lot.
I love the one for Burnett that was 2 feet outside, yet a strike. huh? to be fair, Kazmir had at least one of those too and a lot of low strikes, but not many green (called balls) in the box. how many of those plotted as strikes were just ones that guys swung at vs. called strikes? but either way this Ump seemed a little bit clueless or something on a few pitches.
they still publish the wrong name for the Angels, but I figure some of that may be laziness and not wanting to change past data. so they stay with it.
Brian S (brianguy) - September 24, 2009
nevermind
I was looking at the default charts on the website, rather than the more customized ones above. how do I get it to show the “called strike” option rather than just B/S/X(contact)?
Brian S (brianguy) - September 24, 2009
ahh
“extra detailed”. got it
nevermind, again.
Brian S (brianguy) - September 24, 2009
Very cool.
Is it updated live, or after the game? That would be so sweet if it were live, but I doubt it. That way, you could immediately justify your outrage at a bad call, or conversely eat your words on a good call.
Rally Manatee - September 24, 2009
Izzy in the 8th
I was following the game on my PDA using MLB’s applet. Strike #2 to Izzy in the 8th was low, and outside off the plate by 4 inches. Stunningly bad.
Stirrups - September 24, 2009 via mobile
This is that at-bat.
That was the worst called strike for any Angels player in the game, with 1 out and 2 on, down by a run.
But Bulger got a called strike on an equally bad pitch.
Bulger’s pitch came in the 7th with one on and one out, and it was strike 1 in what turned into a three pitch strikeout. No doubt the Izturis pitch was in a higher leverage situation.
But it turns out that the walk by Figgins which followed, and which loaded the bases with two outs, included a called ball that is supposedly in the strike zone. I wasn’t watching, maybe Figgins is just so short that that was really a high pitch, I dunno, but that what’s on the chart.
So what does all this mean? There probably wasn’t a conspiracy by the home plate umpire to cause the Angels to lose. But the Angels certainly didn’t get any Boston style home field advantages.
AlanFalcon - September 24, 2009
For my part, I believe that the top and bottom of these zone plots
are baseball’s equivalent to crop circles. Just because there are so many data points and so much data, doesn’t make them real.
Stirrups - September 24, 2009
Probably a stupid question...
… but do these charts reflect where the catcher catches the ball, or where the ball is crossing the plate?
clover_black - September 24, 2009
I'm under the impression from things I've read...
that the view is from the catcher’s perspective.
WiHaloFan - September 24, 2009
Seems like there would still be the question
of where the ball is actually at…in the catchers mit or above the plate?
If these graphs are made by some dude who puts a dot on a graph of where the ball theoretically lands, then they’re flawed from the get-go.
clover_black - September 24, 2009
Welcome to my POV.
This stuff is flawed from the get-go. I am biding my time and enjoying the season in progress. In the offseason, while everyone else will be tossing up trade ideas and signing opinions, I will start to unravel the years I have spent thinking about electronic strike zones in a series of posts.
As an example, these points are plotted in only two dimensions. The zone is a three dimensional area in space. The chart plots must assume that each ball took a straight flight path through the strike zone parallel to the ground and on an absolute straight line from the pitching rubber to the back of the zone. Both of these are rarely true. This is but one way that the chart is flawed.
The chart identifies where the pitches actually ended up. It is totally possible via technology to determine that result in space. And, since home plate is fixed, it is also possible to determine the left of zone and right of zone. But the top and bottom of the zone? Something that varies from player to player? Note that these charts ALWAYS have a fixed line for top and bottom of the zone and, since they are capable of including pitches for multiple players, they are assuming that those players have identical top/bottom zone lines. Flawed. The top of Vlad’s K zone is NOT the same as that of Willits. Trying to “normalize” away that problem through raw quantity of data? Baloney.
Stirrups - September 24, 2009
I'm looking forward to reading your off-season posts
Here’s something that briefly talks about how the pitch fx system is set up:
The entire article is HERE
It’s more than just a guy putting dots on a graph. There are graphs showing the flight of the ball from the pitchers hand to (through) the strike zone. As well as velocity and break.
One question that I haven’t found an absolute answer for is; how deep into the strike zone is the data point taken? I had read somewhere that it’s taken from the front of the plate. That’d work only if it’s the same location the umpire uses in determining balls and strikes.
I’m making a big assumption here, but I would think MLB and Sportvision would have thought of all the requirements for a system like this before broadcasting the results. Especially with the number of
statheadsanalysts who would be looking at that data.WiHaloFan - September 25, 2009
Thanks. That is really good stuff.
Except for the mechanical process that goes on pre-game, that is almost exactly what I have been able to bolt together myself over the years concerning the process. And, yes, there are still huge holes in the story (Such as where along the flight of the ball is the data point taken? And what happens when a batter squares around to bunt?).
And, of course, the phrase “the crew in the truck sizes each player during batting practice” is THE precise concept that I have been eluding to in recent comments around here. People presume that robots will solve everything, because they will remove the human factor. In fact, all it does (in the current setup) is relocate that human factor from the human behind the plate to humans in a production truck.
Stirrups - September 25, 2009
That one strike that the ump missed resulted in a walk and Kaz's 3 run inning...
crazy how one bad call can shift a game.
BryanHarvey'sMoustache - September 24, 2009
Great resource
With the notable exception of the 4-strike walk at Fenway, the persistent complaining about balls and strikes on this site is rarely justified.
Brody - September 24, 2009
In defense of the people who complain regularly (and I was one in the game yesterday)
I think the camera angle in most broadcasts makes it hard to tell exactly where the ball crosses the plate. I’d much prefer an angle directly behind the mound, like the Rays have and I think the Red Sox have on their HD channel.
Here’s a comparison between the two angles
Matt UK - September 24, 2009
This is absolutely true. In fact, it is incredibly more complex than that.
The strike zone is a three dimensional box. It does not matter if the ball cross the front of the plate outside the zone, which is where we tend to visualize the pitch on camera, if that same pitch then moves into the zone before it gets to the catcher’s mitt. There is no way to pick that up from a TV monitor.
Stirrups - September 24, 2009
Broadcast in 3D
;)
Downing Rules - September 24, 2009
Like this
LINK
eyespy - September 24, 2009
Yeah...I looked that pitch your talking about up...
and it was a strike.
Angel Hawker - September 24, 2009
You must Login with your SB Nation account and be a member of Halos Heaven to post a comment.