Friday, March 02, 2007

THT has new fielding stats!

Amongst all the fun news about the first spring training game of the year, I overlooked this important item. At least, it's important to people like me:

The Hardball Times has new fielding stats! As initially mentioned in this article by David Gassko, THT has purchased fielding stats from Baseball Info Solutions, who are also responsible for the Fielding Bible stats that I relied on so heavily last season.

These new stats are not, unfortunately, the +/- system that was featured in the Fielding Bible. Those stats are not available to the public for the 2006 season (I actually inquired with BIS about purchasing them, but they wanted a fairly ridiculous price for them...and would only release them for my own personal use, which I'm sure does not include posting them on this blog).

What these new stats are, however, an improved version of Stats, Inc.'s Zone Rating. ZR has been available to the public for quite a while on sites like ESPN.com, and has been successfully translated by Chris Dial into a runs saved above average statistic. The basic idea behind it is that the field is divided up into zones, and each fielder is responsible for a certain proportion of the field. The statistic is calculated (mostly) as the proportion of all balls hit into that zone that are converted by the fielder into an out.

Nevertheless, it suffers from some drawbacks due to some additional tweaks that were made to the original ZR:
  • Double plays were counted twice (I guess because two outs resulted from it?), making the stat particularly probematic for infielders--especially on teams that received a lot of ground balls. This is a major problem that seemed almost unknown among stat circles prior to the publication of the Fielding Bible.
  • Plays made outside of the zone by a fielder were added to both the numerator and the denominator. To be honest, I don't see this as a major problem. It does "bastardize" the stat a bit (MGL's words), but leaving them out ignores the contribution of a fielder with outstanding range.
    • There is an argument that these plays should only be added to the numerator, and that may have some merit. Effectively, that modification would be saying that when a player makes 1 out of zone play it makes up for his missing one play in his zone. Seems appropriate. Original way of doing it, putting the out of zone play in both the numerator and denominator, gives the player credit for making the play, but also indicates that he should have made that play by effectively placing it in his zone. This doesn't distinguish between an out of zone play and an in-zone play.
The version of ZR that Hardball Times purchased from BIS is an "improved" version of ZR that removes those extra tweaks. Perhaps even more importantly, however, THT also is posting the raw data on balls hit into the zone, plays made in the zone, and plays made out of zone. This will allow individuals to make their own decisions about how to treat these events.

No comments:

Post a Comment