Mench

This one's bothering me, kind of. I've now read several places that, following yesterday's trade, have alleged something along the lines of 'Kevin Mench can't hit outside of Arlington.'

This is really ridiculous. For some reason, it's not uncommon for people to use a player's road numbers as if they were the true sample, and the home numbers are treated as merely taint. Aside from the obvious statistical sampling problems this represents - you are cutting your sample size in half, most obviously - this practice neglects baseball's homefield advantage, which is quite significant. Mench has hit .263/.314/.454 outside of Arlington, and the AL average for road games in 2002-2005 was .264/.326/.421. So Mench has been above average on the road, especially if you consider that most of the league got to play its road games in Arlington. He hasn't been great on the road, no doubt, but we're dealing with less than 1000 PA here, and it would be ridiculous to assert that his 930 road PA tell us more about his ability to hit outside of Texas than a projection which regresses his data and incorporates the 1053 home PA would.

People sometimes use splits just because it's, I don't know, cool? I suggest that if you want to use splits to make a point about a player, the first split you should look at is how many plate appearances a player has against LHP and RHP. This will tell you more about how to adjust a player's performance for his future role than a home/road split. And in the case of Mench, it just so happens that it's somewhat illuminating. From 2002-2005, Mench only had 29% of his PA against southpaws, and this season it's been less than 26%. I don't know about you, but when I see a right-handed fringe regular like Mench, I tend to assume that he's being used against LHP each time out and is splitting time against RHP. And while the Rangers have largely treated Mench like a fringe regular in that sense, he hasn't had the platooned out playing time you would expect. Now cut to Milwaukee, who have lefties Gabe Gross and Geoff Jenkins in the outfield. The latter in particular would look a lot better if he hadn't been hitting against southpaws so often. Sure, if Milwaukee uses Mench in the same half-assed way as Texas, this doesn't matter. But the point is that Mench's value hasn't really been maximized, and a team with a setup like the Brewers can easily leverage his value a lot more efficiently.

And though the argument that Cruz could have done close to the same at a cheaper price is somewhat compelling, if the Brewers haven't used Cruz on the big league club thus far then it's pretty reasonable to deduce that their scouting opinion of him isn't very high. Given that his issue is supposed to be breaking balls, I don't think it would be a surprise to discover that players breaking out in their mid-20's who don't hit the curve won't contribute much in the majors, and if the Brewers have done research to confirm that, this all looks pretty reasonable. Moreover, a team like Milwaukee isn't very long on options for improving the club in the short term; if the market for Carlos Lee will ostensibly be more than 4/48, paying $4m to Mench in 2007 to be not much more than a win better than Cruz is not as inefficient as it would otherwise seem. The Brewers only have a finite number of positions they can reasonably upgrade in the offseason, and I'm getting the hunch or impression that the free agent market will be pretty brutal this offseason.

Don't get me started on the "Cordero is very hot right now" claims, which aren't meaningful and certainly aren't necessary for anyone willing to look at his numbers beyond the ERA.

Whether Milwaukee could have gotten a better package for Lee is of course debatable, but it's not like anything the Brewers did only helps them for this year. Ned Colletti's irrational discards notwithstanding, I don't know that there really are a lot of top prospects being readily wagered by GMs on deadline deals, and once you get beyond the level of top prospects, you're looking at players where you should be happy to end up with a Kevin Mench.