SNG math theory??? what point does loose early play become ok?

i've seen some posts on people's strategies for tournaments... i guess i made reference to SNG, which i think is one table... but could be multi-table.

my question is this:

has anyone done studies as to best strategy early in these tournaments??? .... i think everyone agrees as time goes on, steal some blinds, hopefully taking advantage of tight image to do so.

but i was wondering, early on it basically costs almost nothing (i.e. one small bet when you have 75 of them), that doesn't it make sense to loosen up a little and try to double up....

i have found it somewhat difficult to play with catch up when a bunch of people have many more chips than me. the internet player seems happy to risk doubling up and then they'll play pretty much any 40% winning situation... so i play 60% winning chances and i have a very good chance of being out of the tournament after one showdown.

seems much better if you can throw a small number of chips in early and see if you can double up.

has anyone done any math work as to how low the starting blinds have to be relative to starting stack for throwing in a few chips early to be worthwile..... i mean, there must surely come a point where you'd put in the minimum bet (i.e. have to put in 2 chips and you have 1 million).

hope people know what i'm talking about.

thanks in advance!!