Which stats correlate the most with overall success?

I've played in 14 fantasy baseball leagues in the past two years and I use a lot of the data from those leagues to help determine strategy for future leagues. Recently, I put all 168 teams from those 14 leagues into one simulated rotisserie league. The teams were scored just as in any normal rotisserie league, earning 168 points for a first place finish in a category, down to 1 point for a last place finish, then adding up all their points and ranking them.

I thought it might be interesting to see how the teams' rankings in individual categories correlated with their overall ranking. Here are the correlation results (1.00 is perfect correlation) for hitting stats:

R .91
HR .77
RBI .85
SB .46
AVG .61

Runs, homers, and RBI are the leaders for one simple reason: they're highly correlated to one another. That is, if you draft a guy who hits a lot of homeruns, chances are he's also going to drive in a lot of runs and score a lot of runs.

While steals should probably never be completely neglected, they don't correlate well with any of the other categories, and thus, guys like Juan Pierre help you in one category while simultaneously dragging you down in all the others. This also points out why players like Grady Sizemore are so valuable - they get some steals but don't hurt you in the other counting categories.

I have a feeling that the correlation for batting average is artificially low due to people not paying enough attention to their team. They have good players who post good batting averages, so they do well in that category. But because they don't replace injured or crappy players quickly enough, they struggle in the counting categories and end up doing poorly overall.