His 50th percentile projection from PECOTA this year has him at .245/.323/.383 which would put him at just about league average OPS (.706 vs .704 for the AL) for left fielders in 2011. I don't think he's the answer long term, but if he can provide near league average defense in left (might be a tall order), he could be a decent bench player.
As I understand statistics, the PECOTA percentile method was designed very precisely not
to be used in this manner. PECOTA's approach here recognizes that they are attempting to estimate an unknown mean which can, as such, only be estimated in terms of confidence intervals. This is to say, we consider ranges within which we are confident that the true mean resides, but that information doesn't let us say that we are more confident that it is on one side or the other, much less in the center.
From the BaseballProspectus write-up
What this means is that you can't look at a single stat (say, hits or strikeouts) and think that's the range of expectations PECOTA has for that skill. The percentiles are supposed to reflect what we know about the distribution of a player's skill, but they are in essence the average batting line we should expect from that player if he puts up that level of performance in that season. There are a lot of different shapes that performance could take, however, and that means there's more variance in any single component than is reflected in the percentiles. So the correct test of the percentiles is the overall level of performance, not the underlying components.
The more playing time, the more things "even out," but that's the functional equivalent of larger sample sizes giving you tighter confidence intervals for a given level of confidence, i.e. a tighter range within which you might say you are, say, 95% confident the true mean will lie--a range which will not necessarily converge on the 50th percentile numbers. It does not, however, mean that we can expect that results will converge upon the 50 percentile mean projections from PECOTA, (which, of course, is why they explain it as such).
Edited by Reverend, 27 April 2012 - 10:13 PM.