Juicy game but high variance... what it do?

[Craggoo]

I felt this topic belonged in the probability forum. I was discussing this with someone on aim the other day to see if he agreed with my line of reasoning. Now I ask you the same and also if it seems like the variance in this game is extreme. I'll try to use an example that is typical of the game im describing. Lets say hero has QQ and gets 5 callers when he raises preflop (completely standard). Lets say the flop is all cards below a Q. I want to figure out how often your hand should hold up given 4 situations:
1) board has a fd on it
2) board has a straight draw on it
3) board has flush and straight draws on it
4) completely rainbow drawless board
Now, typically people will call your bet on the flop with any pair and even with a single overcard. Lets say that the 5 people we are up against have the following general hands: some pocket pair, top pair, middle pair, Ax, Kx. I figure the minimum amount of bad cards that could come is 17(3A's, 3K's, 7 cards for middle/top pair and their kickers, 2 for the middle pocket pair, 1/2 out for backdoor draws for each hand except the PP). Now this is on a completely drawless board! Imagine now that their are draws out there as well meaning everybody that is drawing has a total combined 23+ outs. Given how loose people are playing what is a reasonable % range that your premium hands (AJ-AK, TT-AA) holding up? I ask because I've been averaging somewhere in the range of 15-18% and this just seems ultra low (probably just wicked variance but its been going on for quite some time now).
In addition, considering how loose these games are what kind of hands are generally most profitable in these type of games (and does position even matter?). BTW, this is $3/$6 limit and people are NEVER folding when its blatantly obvious they're behind. I dont have a big roll to sustain horrible variance so might it be better to play a different game? The casino also hosts a $3/$6 omaha hi-low game on the weekends with a full kill that has the same general group of horrible players. I figure this might actually be a better option (with far less variance) since I do have a good idea of the type of hands to play (suited aces, A2-A3 with a PP [your best chance at quartering someone], a bunch of connected cards like QJT9)

[pzhon]

PokerStove

When a game is soft, your win rate should be much higher than in a tough game where people fold. Often, this compensates for the added variance from the bad calls. If (standard deviation^2)/win rate does not increase, then you don't need a larger bankroll to play.

If you don't know how to play in these games, read SSHE. Misguided people will tell you to play like a rock to decrease your variance. SSHE says otherwise.

[DiamondDog]

If Hero's raise is getting five callers, Hero's raise isn't big enough. Get more money in pre-flop. Against fishy loose callers, bigger raises are ++++++EV.

Variance is the price you pay for the huge winrate you can expect from such a loose game.

[Craggoo]

Perhaps i forgot to mention its limit (maybe somday the casino will get a clue and start hosting no limit games)

[GoogleEmployee]

low limit...$3/6, $4/8. People NEVER fold.

[bustufools]

No I won't help you enhance your bot