Also, excession: I am not a mathy guy, so i dont know the correct probablistic answer to this, or if there even is one- i think i understand what you mean by decreasing variance, because more tables = more hands = getting you to the 'long run' sooner, but at the same time, multitabling still increases your exposure, like 3 tables of 10/20 is like one table of 30/60 in terms of how much you can lose if you run bad, right? ANd i mean, even if you play 4-5 tabels for like 4 hours, your gonna get like 1200 hands maybe, which is still a tiny sample size, and very subject to variance, right?

I am no mathy guy either, but from my limited understanding of statistics, bigger sample size means more convergence towards the mean.
If you are a winning player (and making the rather simplistic assumption that your win rate is unaffected by 3 tabling as opposed to 1 tabling) this means the more hands you play in a given time period the more you are likely to win and the more your good and bad luck will balance out (as variance due to luck will presumably converge to zero over enough hands, unless of course you are Chris 'God Bless Chris' Moneymaker ).

So by 3-tabling you should increase your actual winnings threefold with and achieve lower variance per hand played (sure you play more hands and if you are playign badly you can flush money away real fast, but if you are playign well enough to have a +ve expectation then it really must be the best thing to do in the long run)

............

.............

................

...............

...............