I did a little study regarding that using data from projected standings of mock drafts
that I ran. I am not claiming it is the best scientifical experiment in the world, but nonetheless you asked so here is what I have/did.
In my study I only looked at HRs, RBIs, Runs, SBs, Wins, Ks, Saves not the ratio categories, that would take some extra work.
What I did is averaged the totals from the top 3 teams in each category using a sample size of 10 mock drafts. I didn't count bench player stats towards projections, and I used the standard 12 team Yahoo league configurations.
like I said, not completely scientifical but just something I did when I had some free time last month. Perhaps others can chime in as well.
If you really want to know the lowdown on this stuff... you would talk to Todd Zola at mastersball and probably have to pay for their premium service, but if you have the loose $40 or so, it's a wealth of information. Todd has done indepth research on this exact topic and has written many essays on it. It's great stuff!!!