February 6, 2004
The Sabean Way
Vlad or 10-pack?lunatic fringe in the process. So it's no surprise that he faced his share of skeptical questions from Giants fans during his live chat on MLB.com earlier this week. But it was his answer to a fairly innocuous question that raised the most eyebrows among the "fringers":
Q: Did you ever make an offer for Vladimir Guerrero?
Yikes! There's nothing wrong with the general philosophy behind that answer. "We can't afford player A because we have to pay players B, C, and D" is something you hear out of front offices all the time--it's the kind of tradeoff GMs make every off-season. But normally when you hear an explanation like that, you expect players B, C, and D to be guys you actually want on your team. Aside from A.J. Pierzynski, Sabean's list is barely more than a collection of roster fillers. Is Sabean serious...they gave up a shot at Guerrero or Sheffield for these guys?
Salary-wise, the answer is pretty clearly yes. When you add up the 2004 salaries for the 10 players Sabean named, you get a Guerrero-sized bottom line:
Pierzynski $3.500M Snow 1.500M Tomko 1.500M Tucker 1.500M Eyre 1.025M Hammonds 1.000M Herges 1.000M Feliz 0.850M Hermanson 0.800M Brower 0.625M ------- TOTAL $13.300M
That's more than $13 million dollars for these 10 bottom-of-the-roster guys. If the Giants had gone with near-minimum salary players (say, $350,000) instead of these 10, they would have saved $9.8 million for the year, almost exactly what Guerrero is getting from the Angels ($10 million), not counting signing bonus. That assumes Pierzynski gets the $3.5 million he's asking for in arbitration--an assumption the Giants pretty much have to make in their budget planning anyway. Signing Guerrero would have required lots of money in 2005-2007 as well, of course, but those years would have been much easier to fit into the Giants' budget than 2004, as the expiration of the pricey Robb Nen contract at the end of this season will create some budgetary breathing room.
(I'm leaving Mohr out of the discussion since he's already a near-minimum salary player. It's not clear why Sabean mentioned him, since he's not arbitration-eligible and his trade was reportedly separate from the Pierzynski trade.)
Sabean putting Jim Brower & Co. on equal footing with Vlad Guerrero looks silly on its face. But maybe there's something to his claim that the Giants couldn't possibly be competitive without those 10 guys, even if they had Guerrero playing right field. Our disparaging comments above aside, most of these 10 players have some value, in that they're better than the alternatives immediately available to the Giants. But are they really worth passing on Guerrero?
We'll get a reading on the value of Sabean's off-season signings by looking at PECOTA's projection of each player's Value Over a Replacement Player (VORP) for 2004. PECOTA is Nate Silver's system for predicting player performance; you can read about the system's accuracy here. VORP, developed by Keith Woolner, measures the number of runs a player produces (or prevents, in a pitcher's case) beyond what a "replacement level," or freely available, player would produce.
Let's start with the pitchers. We'll compare the predicted performance of the five pitchers signed by Sabean to that of their likely alternatives if the Giants had pooled their free agent money for Guerrero or another big-ticket player. (Alternatives are projected to the same number of innings.) Note that in the Guerrero scenario, the Giants wouldn't have traded for the arbitration-eligible A.J. Pierzynski, so they would still have Joe Nathan around.
PITCHER PROJ. VORP ALTERNATIVE PROJ. VORP DIFF Brett Tomko 11.1 Kevin Correia 18.5 -7.4 Dustin Hermanson 7.3 Ryan Jensen 3.7 3.6 Matt Herges 9.8 Joe Nathan 5.8 3.9 Scott Eyre 6.1 Noah Lowry 4.7 1.4 Jim Brower 10.8 Chad Zerbe 7.0 3.8 ----- BENEFIT OF SABEAN SIGNINGS 5.3
By these numbers, the $5 million Sabean spent on 2004 pitcher salaries this off-season bought the Giants a measly five runs, or about half a win. That's partly because of the possibility that PECOTA sees of youngster Kevin Correia pitching better than journeyman Brett Tomko--not implausible at all if you look at Tomko's recent ERAs. Granted, PECOTA's outlook on Correia may be a little on the rosy side, but then again it may be taking too pessimistic a view of Joe Nathan.
Something this analysis doesn't consider is the "ripple effect" of not having Tomko et al. If Kevin Correia had to fill the Brett Tomko role for the 2004 Giants, then someone else--say, Jeff Clark--would have to fill Correia's role. And someone else--say, Merkin Valdez--would have to fill Clark's role, and so on. It's hard to know how much difference this makes, but it's worth acknowledging. On the other hand, the analysis also doesn't consider the very real possibility that the Giants could come up with better alternatives than some of the ones listed above through the waiver wire, a la Dustin Hermanson last year.
So the Giants' off-season pitching signings don't seem to have reaped much bang for the buck. But surely the celebrated A.J. Pierzynski trade must represent a big gain for the team, right? As it stands, the Giants will likely get around 450 PAs from Pierzynski and 200 PAs from backup Yorvit Torrealba. If they had passed on Pierzynski and his multi-million dollar price tag, they likely would have gotten around 450 PAs from Torrealba and 200 PAs from Alberto Castillo or Trey Lunsford, both of whom PECOTA sees as right around replacement level.
CATCHER tandem PROJ. VORP ALTERNATIVE PROJ. VORP DIFF Pierzynski/Torrealba 31.5 Torrealba/replacement 19.7 11.8
So the $2 to $4 million cost of bringing in Pierzynski projects to bring the Giants an extra 12 runs in 2004, or just over one win. That's a better value than the pitching salaries, but still not all that good. The reason that Pierzynski isn't more valuable to the Giants is that they already had a pretty good, and underrated, catcher in Torrealba. Pierzynski is an upgrade to be sure, but less of one than he would be for a lot of other teams.
Newly re-signed infielders J.T. Snow and Pedro Feliz are the most likely candidates to get playing time at first base for 2004, although another player could still be brought in before Opening Day. It's hard to name a first base alternative within the organization--the farm system is especially weak at the infield corners--but I'll go with a Tony Torcato/Lance Niekro platoon.
1B tandem PROJ. VORP ALTERNATIVE PROJ. VORP DIFF Snow/Feliz 8.7 Torcato/Niekro -10.0 18.7
I had to guess on the projection of a Torcato/Niekro platoon, since PECOTA projected both of them at positions other than first base. I tried to be conservative with the guess, giving Snow and Feliz as much benefit of the doubt as possible. Here the Giants would have plenty of options better than Torcato/Niekro if they really had to do without Snow and Feliz--trying to pick up a Graham Koonce-style Quadruple-A masher off the waiver wire or in a trade, for example. The bottom line is that the 18.7-run estimated value of the Snow/Feliz first base tandem is almost certainly on the high side.
So our estimated value of the Giants' off-season signings at pitcher, catcher, and first is 5.3 + 11.8 + 18.7 = 35.8 runs, or about three and a half wins. That leaves only the outfield. Here, instead of comparing the production of Sabean's new signees to the organization's alternatives, we'll compare them to the alternative use for the free agent money, Vladimir Guerrero.
RF tandem PROJ. VORP ALTERNATIVE PROJ. VORP DIFF Tucker/Hammonds 1.9 Guerrero 48.9 -47.0
So Sabean's safety-in-numbers approach this off-season may have bought the Giants around 36 runs compared to doing nothing, but using the same money to get Guerrero would have bought them--based on the projections above--quite a bit more: 47 runs by this estimate, 11 more runs, or just over one more win, than Sabean's approach. Plus it would have put them in a better position for the next few years, and would have given them a credible core for making the transition to a post-Bonds era. Gary Sheffield's not nearly as young as Vlad, but projects to have similar offensive impact for the next two or three years. The numbers confirm what our intuition told us: The Giants would be better off with one Vladimir Guerrero than with a whole bunch of Jim Browers and Pedro Felizes.
I don't want to make too big a deal out of a single chat comment. It may have just been a bad answer in the spur of the moment. There are plenty of other defensible reasons for not going after Guerrero. Maybe Sabean is especially wary of giving a long-term contract to a player with a history of back problems. Maybe he's keeping a slot free for a particular marquee free agent of the 2004-2005 class. Maybe he's determined to keep the right field spot available for Todd Linden. Maybe he doesn't want to give out any mega-contracts regardless of how good a player he's eyeing, with Giants' ownership breathing down his neck about heavy mortgage payments and managing risk.
But regardless of Guerrero's fit with the Giants, Sabean's answer highlights the problem of his off-season. While the contracts he's given to very replaceable players have all been small enough to escape scrutiny individually, they've added up. And they could have added up to someone who would have made a difference.