you guys remember when zack cox was in the cardinals' system, and the fourth law of the internet (after "there shall be porn," "no one will properly use 'there' and 'they're' and 'their,'" and "nobody is really going to give you a free iPad just for clicking on a link") appeared to be that within four comments of any mention of zack cox's name, somebody was required to ask when he'd start playing second base?
well, our situation has not improved, although we at least know the answer to that perennial question ("never, at least for the cardinals").
i mean, i suppose it has improved in the sense that daniel descalso has become a tolerable place holder at second base, good for half a win to maybe a win above replacement value. but we're hardly set at second base. for the second time in a year, i had to look up chone figgins's stats and confirm that, no, he would not be an acceptable option at second.
so, the latest experiment is that matt carpenter, long-time third baseman will try his hand at the keystone. it is unlikely that carpenter, brand new at the position, will be affirmatively good at the role. instead, it seems like the question is:
just how bad can carpenter be defensively than descalso, yet still be better than descalso overall?
the numbers are out now, on various projection systems.
last year, matt carpenter put up a .355 wOBA, which is pretty good. most of the projection systems don't like him to repeat, though. ZiPS is most pessimistic, labeling him with a .324 projection, while the other projection systems all like him to end up with numbers closer to last year's: bill james thinks he's a .353 hitter, while oliver looks for a .350 wOBA from carpenter. steamer splits the difference and calls him a .335 hitter. if you average those out, you get a .341 wOBA from matt.
daniel descalso is not as difficult to project, since he's been fairly consistent and has a longer major league track record. last season, he had a BABIP-affected .278 wOBA; he has a career .288 wOBA. not surprisingly, most projections aim for a .300 wOBA for him. both steamer and ZiPS call him .300 on the nose, while bill james takes the under at .299, and oliver thinks he's going to start slugging more and proposes a .309 wOBA. so, average that out and you get a .302 for daniel.
descalso has a modest net negative ranking as both a second baseman and a third baseman by UZR. he doesn't have enough playing time to really make any conclusions about his overall aptitude as a second baseman. the conventional scouting wisdom seems to be that he's average-ish at second base. so, let's assume that.
the calculation for WAR is pretty simple. it's just wRAA, UBR, and UZR added up, with a positional adjustment thrown in. (UBR is the baserunning metric, which we can probable safely ignore, since neither descalso nor carpenter appears to pose an substantial base running advantage or disadvantage).
the real question is how carpenter's wRAA offsets his defense. descalso has an average wRAA projection for being -4.25 runs below average with the bat. carpenter has a wRAA projection that averages out to being 8.5 runs above average with the bat.
if descalso is average (and we ignore base-running), then matt carpenter would have to have a -12.75 UZR at second base to be below descalso's performance. he'd pretty much have to go full schumaker.
now, i've just committed one of my least favorite statistical errors, that of unwarranted precision. while arithmeticly correct, we are not sure to the hundredths place of carpenter's value. we probably shouldn't even call that number -13. it should probably be -10 to -15-ish. none of the projections are terribly precise. we aren't really sure that descalso is exactly average. maybe he's a +3 or -4 second baseman.
this is our best ballpark number. you can adjust it up, if you think descalso is above average at second, or is going to steal 20 bases, or if you think ZiPS projection for carpenter is better than the others. you can adjust it downward if you think descalso is below average at second, or carpenter's ZiPS projection is too pessimistic. however you put the numbers together, it's pretty clear that even the most pro-descalso and anti-carpenter theory should require that carpenter be a few runs below average at second.
to the extent one can rely on the scouting reports coming out of spring training, they seem to be guardedly positive about carpenter's defense. he's widely and I think sincerely described as picking up the position better that schumaker. he has turned several double plays more smoothly than we saw in, say, Allen Craig's emergency appearances at second in 2011. Carpenter was a decent to good third baseman, and the skills that make a good third baseman should also make a good second baseman, even if the angles and footwork are different. I'm fully prepared to take "not as bad as schumaker" as a reasonable goal.
[ed. - to provide context, dan uggla, rickie weeks, and neil walker all had UZR/150's around -5 over the course of the 2010-2012 seasons, which should be statistically significant. skip schumaker had a -10.9 rating over four seasons at second base. to translate this roughly, if we think that carpenter is roughly schumaker-ish at second, he's probably a better option than descalso. if we think carpenter is roughly uggla-ish at second, he is almost certainly a better option than descalso.]