Add new comment

While I do think that UZR is useful and it is a very good starting point for defensive discussions (leading to the incoorporation of other stats plus a judicious use of ones eyes of course), it certainly has its failings.  One of which is extreme bias in smaller samples.  It takes pretty much a full season, if not more, for UZR to start to get reasonable, so a handful of starts here or there, some at short some at 2b are going to be extremely volitile in UZR's eyes, thus I think that the -28 UZR for Scutaro needs to be thrown out pretty much altogether and seen as small sample size static, and the overal career UZR may be more useful, which makes Scutaro out to be a pretty average defender at short, which also lines up with my eyes as well.
We definitely put WAY to much stock into partial or even seasonal UZRs, career UZR for a MLB player that has been around a while is probably a uch more reasonable place to actually start putting stock into UZR analysis, anything less than that and UZR is just a nice guess and a good starting point to get your going in the right direction. (IMHO of course :) )