Dan Fox continues to work on his retrosheet-based fielding metric. It’s behind the paywall, so I won’t mention much. Fox does give the numbers in both UZR and SFR (Fox’s stat) for Giles. Here they are:

 … UZR … SFR
03: -3 …….. +0 (everything in runs)
04: +24 ….. +1
05: +20 ….. -20
06: +23 ….. -2
07: -3 …….. -22 (UZR through mid-season in 2007)

Basically, there’s quite a lot of difference there. Just take a simple average over the last 5 years – UZR has him at +12 and SFR at -8.6. Let’s take a look at some other numbers. Justin Inaz has converted the HBT’s zone rating data into a plus/minus stat. Giles was +9.3 runs there last year.

John Dewan’s plus/minus stat featured in The Bill James Handbook had him at +18 (plays, not runs, this time) from 2005-07.

How about the fans? After my quick calculations, that works out to about 3.5 runs above the average right fielder. Well, we’ve got data point all over the place. Above average, average, and well below. Where do you guys see Giles fielding-wise entering 2008?

Pizza Cutter responds to Brian Bannister’s interview. I’m sure this has been discussed before, but I think the reason why BABiP’s are a little higher in hitter’s counts is because pitchers are more apt to throw a fastball. When you’ve got a hitter sitting fastball and a 95 mph pitch coming to the plate, the ball is going to come off the bat faster, resulting in a slightly higher average on balls in play. That’s just my little theory based on nothing more than a little common sense. Anyway, Pizza’s study is interesting and there may be something to Bannister’s points. If there’s anything I’ve learned from sabermetrics, it’s to question everything about the game.  

Tango Tiger has created a sabermetric wiki that will likely become an great resource. Just to remind you, I’ve got my Sabermetric Studies site, as well, which I need to start adding more links to.

Ducksnorts’ Geoff Young takes a look at Bill James Brock6 system and Barry Bonds. Fun stuff. What I’ve always thought would be valuable to projection guys is to do something similar to this … like take a bunch of players from 1990 and project them for the next 5 years (i.e., 91, 92, 93, etc.) and see how accurate they are. Then you could tweak the projections for more long term accuracy, as that is really one major part of a projection (the long term aspect, rather than just one year). I’m not sure if there’s just too much work involved or if getting the right data would be a problem, but it’d surely be interesting.

Brian S. chimes in with some more thoughts on Khalil. I really do believe that the large home/road splits have something to do with the way Greene puts balls in play + the context in Petco (heavy air, outfield configuration, etc). But it’s more important to ask questions rather than trying to prove something (at least, to me), so until I know a better way to investigate the subject, I’ll shutup about it.

Corey Brock thinks CY has a shot to win the Cy Young Award next year. Hmm … I think he’s got a chance, that’s for sure. But I’d be worried about things like his BABiP finally regressing (at least a little bit) as well as his hr/fb% — which was a career low 4.1% last year. Still, the guy’s a fine pitcher.