[This is my first Diary post, so please forgive if I don't post it totally right]

El Senor Sabean has always seemed to me to be muy bueno at certain things : scouting and drafting pitchers, finding one last harrah out of vets who everyone else had given up on, getting offensive production out of the middle of the infield, and, before 2003,  fleecing other teams on the trading market.

And he has always seemed muy malo at certain other things : scouting drafting and developing position players (duh), understanding the value of OBP in building an offense or K/BB ratio in predicting future performance for pitchers, or letting anyone under 27 get a starting job.

The thing that bugs me the most about his game plan, however, has been his failure to get much production out of the corners of the infield.  In the last six years (2001-2006), all Giants third basemen have hit a combined to hit .259/.313/.406, and first basemen have combined for .262/.340/.418.  Ugggh.  Subtract Snow's immaculate miracle of a second half of 2004 out of the aggregate, and those first base numbers probably look kinda Nefi-esque.

At the start of each off season, I say to myself, OK, first base sucked this year.  But, next year, I am certain we'll have some sort of middle-of-the-order guy in there, if not Carlos Delgado or Lyle Overbay or Frank Thomas, then at least like Kevin Millar or something.  I mean - right?  Nature abhors a vacuum and all?

And each year, opening day comes, and the best that we can hope for from the first base guy is that hey maybe he'll crack 20 homers, or, if the stars all align, hit .270 and not .230.

In writing an article for today's baseball prospectus, Nate Silver didn't think that he was trying to get me mad at Sabean again.  But he did:

http://www.baseballprospectus.com/article.php?articleid=5848

He said:

Fortunately, writing about first base prospects gives me something of the day off, since there aren't very many of them. A lot of this flows from sabermetric first principles; first base is the endpoint of the defensive spectrum, which not only means that you have to hit a whole lot to make a name for yourself at the position, but also that there's nowhere to go if your defense gets any worse.

First basemen are also hurt by our use of the Upside metric, which only gives players credit for the probability that they're able to turn in a performance above league average. When Keith Woolner did his initial research for VORP, which was based in part on identifying the worst regulars in the league at any given time, he found that there was a particularly large difference between league average and replacement level for first basemen. Generally, if you look around the league at any given time, you'll find a few pretty bad starting first basemen (think Travis Lee or Jeff Conine or Doug Mientkiewicz), but the overall standard of competition is very high.

I suspect the reason for this stems from the fact that almost anybody can play first base. Teams sometimes use first base as a dumping ground when the roster has other problems: let's throw Travis Lee or Ty Wigginton out there ... hopefully he'll hit .270 ... we've got bigger fish to fry. But this solution rarely sticks once a team has the chance to jigger up its roster in the off-season--it's just too easy to find a Erubiel Durazo or a Chris Shelton. Or, you can move your aging third baseman over to first base, and target a solution at the hot corner instead. The work I did on freely available talent last year suggests that it's quite a bit easier to find an adequate first baseman for next to nothing than it is an adequate corner outfielder.

Put differently, perhaps the reason why there's a large difference between the very worst first basemen and an average first baseman is precisely because decent first basemen are so easy to find. When a team settles on an inadequate solution at the position, it's giving up a lot of ground; the cost of making inefficient decisions is high. Not coincidentally, Keith also found that there is a particularly small difference between league-average catchers and replacement-level catchers. Since it's very hard to find someone who can play catcher, in some sense it's more difficult to mess that decision up. It's sort of like this: the difference between the worst meal and an average meal at Chez Panisse is larger than the difference between the worst meal at McDonald's and the average McMorsel.