The reason Duncan doesn’t look so bad on the BP metrics is, in part, because the replacement level is set far too low for FRAR. If you want to use their metrics, and I’m not a fan of them, use the FRAA as your replacement level. Also remember that those are counting stats so Adam Dunn is going to look worse since he played more games. You can use their adjusted games stat to turn FRAA into a psuedo rate stat.

The 2007 data for UZR hasn’t been released yet (if it will be at all).

The RZR stats are at The Hardball Times but a blogger by the name of Justin Inaz (goes by jinaz at The Book blog) who blogs about the Reds has converted all the RZR data into runs because RZR is a rate (plus there are out of zone plays to account for in terms of RZR). Anyway, RZR put him at around -13 runs last season.

PMR is in terms of plays and as erik noted 1 play = roughly .8 runs. Tango explains this in detail over at The Book blog in terms of linear weights. Anyway, PMR had Duncan as -18 plays or about -14 runs.

A newer system that is a psuedo-play-by-play defensive metric which looks promising is being done by Dan Fox at BP and it’s called SFR (Simple Fielding Runs) you can read about it here. That system pegged Duncan at -16 runs. All the advanced defensive metrics agree that he’s about -10 to -15 runs with the glove.

If you want to read a more detailed description of the fielding stats (with the exception of SFR) written in plain english, I’d check out parts 3a and 3b of Justin’s Player Value series. They’re good primers.

Anyone think he’s going to outproduce those lines given that he’s going to be 3 years removed from 2005 and 2 years older than his 2006 line? 2bman don’t typically age well. I think you’re underestimating the aging curves for 32 year old 2B. Tango’s got a nifty aging curve plot near the end although this study is a little old. Here’s another one from Nate Silver that has a specific graph for 2nd basemen. Note that they have a steeper decline phase than the average ballplayer.

I’m not fluent enough in the various defensive metrics to know which is best, but it sure seems like there’s at least some evidence that Duncan is passable in the field.

As for Dunc’s ability to get better. I am firmly behind the guy, definitely inexact, but I know in going to batting practice every time I see again, there is no one working harder on getting better than Dunc. He is miscast as a LF, but his bat is special if you believe in his first half…

As for Kennedy he is not a world beater, but I’d bet on him being league average or better next year, that’s way better than the prospects (next year) or anyone else on that list…

And .290/.380/.550 first half would place his trade value a lot higher than it is now. He’s perfectly capable of doing that, he has proven he can. Playing through his injury killed his numbers last year.

i personally like duncan, but players of his ilk are a dime a dozen. a couple of dunc’s batting comps at bbref are #1 luke scott and #2 thru age 26 is kevin mench. also, he has a .632 career ops versus lefties, the man needs to be platooned. he’s also not going to be much more valuable in a trade, so now would be a good time to do see what he could fetch.