Tuesday, March 01, 2005

Is This A Dead Horse?

Imperabo, one of the Fourth Outfielder's favorite Primates, wrote this the other day during this discussion:

This is the one thing I do have concerns about with regard to Depo. He clearly puts a lot of weight on UZR (or something similar), and I'm afraid of it blowing up on us. The groundball staff won't be nearly as effective without good infield defense.

This is something I've considered as well. I thought I should take a closer look.

Let's assume the players at issue are Jeff Kent and Jose Valentin, since I don't think the Drew signing has seen much critique from a defensive standpoint. Firstly, how good do Kent and Valentin have to be at defense to be worth their contracts? If Kent is a +15 batter and Valentin a -10 batter (a low estimate, in my opinion), then with neutral defense they're +22 and -7 players when accounting for their positions. Baserunning makes those more like +20 and -5. Kent's getting an $8.5m average annual contract value and Valentin's getting $3.5m. For Kent, figuring $4m for an average player and $2m per extra win, his baseline is $8m, meaning he has to contribute 2.5 runs above average on defense to be worth the full $8.5m. For Valentin, since he'll sit a quarter of the time, his baseline is ($4m)*3/4 + (-.5)*3/4*$2m, which equals $2.25m, meaning he'll need to contribute 6 defensive runs above average to be worth the full $3.5m. At 3/4 playing time, that comes out to a rate of +8.

We'll deal with Valentin first. Valentin would have to be a pretty bad shortstop for anyone to figure him to be below average at third. Almost every defensive statistic likes Jose Valentin - He's been above average in David Pinto's Probablistic Model of Range in both 2003 and 2004. He has been above average in Davenports every season since coming to the White Sox in 2000. He has been above average in UZR in every season since 2000 except his -2 in 2001; he's also been at the top of the charts in UZR each of the past two seasons. He's been one of the best according to Studes' win shares calculations each of the past two seasons. While each of these methods has flaws, the fact that four different methods - including two using zone data and two not using zone data - show Valentin as being a very good defensive shortstop seems like a pretty strong preponderance of data. It's not just UZR that likes him.

Kent has certainly not been a favorite of most observers. However, we're not just dealing with UZR in his defense. Michael Humphreys' side-by-side comparison of UZR, ZR, Davenports, and Defensive Regression Analysis for 2001-2003 shows Kent as being above average by all four metrics, ranging from +2 to +12. Win Shares and Pinto's PMR were both down on him in 2003, but both had him near the top of the charts in 2004. That they were down on him in 2003 isn't terribly disconcerting, as the Davenport system has that as his only below average recent season (at -1). His 2004 Davenport was +10, and his 2004 UZR was +20. I also have Kent's 2004 ZR sitting around, and it's above average. Again, the consensus among several different methods seems to be that Kent is an above average fielder.

The other issue here is clearly one of misplaced expectations - most of us have seen that both Kent and Valentin dominated the 2004 UZR rankings and got excited. One season's worth of UZR, however, is not a sufficient sample size. A +20 over the course of one season doesn't mean that the player's true talent level is +20; it's probably closer to +5. Kent is probably in the +5 range defensively and Valentin at third is probably in the +10 or +15 range, with the caveat that both have shown an upward trend recently.

In summation, both Kent and Valentin look to be good bets to match their dollar value unless the available defensive evaluation systems are all pretty much junk. That's pretty good, considering that the average free agent signing is for above the player's actual value.

There's also another meta issue here. Paul DePodesta does not use any of these defensive metrics, or at least if he does he's using them to verify whatever proprietary methods he has going. If these signings (along with the Odalis Perez and Derek Lowe signings) signify that DePodesta's relying heavily on his defensive evaluation system, then we should be pleased since we have much evidence to indicate that his system is getting it right.

Comments:
Beyond the pure defensive aspects of the signings are the long term ramifications. If one listens carefully, one hears depo constantly saying "we are trying to build for the future while trying to win now, which is not easy to do." With three 3b prospects and Guzman not a year away, it makes sense to sign valentin for a bit more if he doesn't block Guzman next year. The extra we pay now will be offset by Guzman's dirt cheap contract for the next 6 years. Our second base prospects (Aybar and Young (and possibly LaRoche if Guzman sticks)) are probably a couple years away, so it makes sense for Kent to have a 2 year deal.

Factor in the "McCheap reputation". This year, our payroll is a little under 100 mill. In 2 or 3 years that could drop to 80 ish, but we'd still be fielding a competitive team, with all our young players. Maintaining a high payroll is important for this year for PR reasons. It may not be necessary to spend 100 million to field a contender in a few years.
 
That strikes me as a possibility, Roger, but whether you accept Tom's explanation for overpaying for some players, or my explanation (DePo got caught with his pants down -- like everyone else -- by the bull market in players this year), my gut feel is that Dodger payroll is likely to stick around the $100M mark, and maybe even nose up a tad.
 
It would be great if the payroll stayed at 100 million. Part of the issue is who will give all that money to next year? Next years free agent crop looks pretty lousy. With Werth, Drew, Bradley(?), guzman, izzy, kent, choi, and navarro under our control next year I don't see what we would pick up. There are no pitchers and hopefully some of the prospects will be ready to step into the back of the rotation.

I agree that nobody saw that the market would escalate as it did. We were essentially forced to pay Lowe, even if he does fit our ball park perfectly. I blame the mets, who were forced to resign Benson after their somewhat misguided trade for him last midseason.
 
Hey, you have a great blog here! I'm definitely going to bookmark you!

I have a accounting an introduction
site/blog. It pretty much covers accounting an introduction
related stuff.

Come and check it out if you get time :-)
 
This is an excellent blog. Keep it going.You are providing
a great resource on the Internet here!
If you have a moment, please take a look at my average professional athlete salary site.
Have a great week!
 
Post a Comment

<< Home

This page is powered by Blogger. Isn't yours?