[nat]

"there are still several situations I'd much prefer a guy who aggressively puts bat to ball, as opposed to taking a walk... like any 2 out RISP situation."

Of course there are situations like that. And it might be a good idea to have somebody on the bench who can do just that, so that you can use him as a pinch hitter in those situations. But, on average, a point of on-base percentage will do more to win games for you than a point of slugging percentage. The evidence for that is that that is precisely what has happened. The people who looked into this ran regression analyses to determine the strength of the correlation between slugging/on-base and runs, using historical data. It turns out that the correlation is stronger for on-base percentage than it is for slugging percentage.

As far as RBI go: they are a very crude measure of the quality of a batter. They depend very heavily on the ability of the guys in front of you to get on base. It's not a coincidence that the guys batting behind Wade Boggs (to take one high OBP guy) had lots of RBI. Now, RBI are also reflective of a batter's ability to drive the ball, but they don't reflect it very well. Slugging percentage does a much better job.

Edit: Lots of splits don't mean anything, in that they are not predictive. Sometimes people summarize this point by saying "there's no such thing as clutch hitting"; what this means is that a player's performance in the clutch at one time does not make a future clutch performance any more or less likely.