Hit Locations

103 views
Skip to first unread message

Chris St. John

unread,
Apr 18, 2013, 9:53:23 AM4/18/13
to baseball-sq...@googlegroups.com
Is it possible to use Retrosheet to create hit maps like at Katron? I want to test how many balls are hit to a certain area at each park. Is that data available on Retrosheet and if so where? (I'm finally getting internet at home soon and can start messing around with this stuff again).

Matt Bandi

unread,
Apr 18, 2013, 12:37:35 PM4/18/13
to baseball-sq...@googlegroups.com
I'm not too familiar with Retrosheet data, but I think it only contains the estimated zone of each ball in play. So you can get general hit locations, but can't really create a spray chart. (diagram)

MLB Gameday data contains coordinates for each ball in play, which is what katron uses. But they only record the location where the ball was first touched, so they're not really accurate. Ground ball singles will have coordinates where the outfielder fielded the ball, etc.

Chris St. John

unread,
Apr 18, 2013, 12:44:41 PM4/18/13
to Matt Bandi, baseball-sq...@googlegroups.com
How does one then get the Gameday data? Is that a more involved Perl type programming experience?

Matt Bandi

unread,
Apr 18, 2013, 1:04:04 PM4/18/13
to baseball-sq...@googlegroups.com, Matt Bandi
Yeah, it's in with the PITCHf/x data, so you would have to go through that whole process of downloading the data and getting it into a database. But if you're just looking for some specific data, I could probably run a query and get it to you in a CSV file.

Harry Pavlidis

unread,
Apr 18, 2013, 1:06:56 PM4/18/13
to Matt Bandi, baseball-sq...@googlegroups.com
if you gather that data you have to (a) convert to it to the proper coordinate system for each park [see Peter Jensen's work] and (b) caveat it because it's where the ball was touched by a fielder.
It's unreliable for fine grained work.

Sean Ahmed

unread,
Apr 18, 2013, 1:09:43 PM4/18/13
to Harry Pavlidis, Matt Bandi, baseball-sq...@googlegroups.com
Also, I believe there was an across-the-board change in 2012 that standardized the scale of the maps. Stringers click on maps, and the Gameday program translates where they click to x,y coordinates. In the past, infields could be disproportionately big, angles may not have been correct, etc. You may have more luck looking at only 2012, but I can't say first-hand.

Chris St. John

unread,
Apr 18, 2013, 1:10:48 PM4/18/13
to Harry Pavlidis, Matt Bandi, baseball-sq...@googlegroups.com
I was thinking of just looking at fly balls and home runs to hopefully get rid of the fielder caveat. I have a research idea but it's still in its infancy. I suppose I'll return to this when I want to pursue it more. Thanks, guys
Reply all
Reply to author
Forward
0 new messages