If you pull down the other bet once you win $7, then it is true that you will on average lose 1.03% of the $12 you are risking.
However, you can't make two 1.52% house edge bets and magically reduce the vig to 1.03%. If you leave both bets up until both bets are decided, then not surprisingly, the house edge is still 1.52% on the $12 at risk.
Let's look at what happens. There are three possible outcomes:
If you hit a 7 before either 6 or 8, you lose $12 (p = 6/16).
If you hit a 6, then a 7, you win $1. (p = 5/16 * 6/11)
If you hit an 8, then a 7, you win $1. (p = 5/16 * 6/11)
If you hit both a 6 and an 8, you win $14. (p = 10/16 * 5/11)
Add it all up and you get an average loss of $0.18, which is of course 1.52% of the $12 risked.
ChefJJ said:
Nobody was talking about taking the other bet down after a win, FLASH...except for you.
If you want that 1.03% to be correct, you'll need to pull down the other half your bet once you win.
Once you decide that you might not leave a bet up until it is decided, all sorts of strange things can happen. For example, if we just agree to place the 6 only for $6 and only leave it up for one roll, should we really gain any solace in the knowledge that we on average will lose only 0.463% of the amount risked? Hey, sounds like the best bet at the table when we look at it like that!
I'll stick with knowing that place bets on the 6 or 8 will cost me 1.52% of the amount risked, no matter what combination I play them in. To look at what happens if I pull the bets down before they are decided doesn't really tell me much.