As far as C, you can't be first in queue on the major ECNs, but there's still plenty of room to trade it... I've had no more than a handful of negative days in C this whole year. As far as KO/PEP, his BP won't even let him take 100 shares of both of them, AND while yes, he can maintain some neutrality being long KO and short pep, their strong correlation to the S&Ps will mess him up. Also he's limited to a 3 cent stop - pure noise.
Thanks for the thoughts everyone. Yes, KO/PEP, is not an option. These stocks are too expensive. It has to be a group of inexpensive stocks, $4 to $5. What I really want to be able to do is essentially "black box" a group of stocks. In other words, treat the box as if it were one stock. In this way, it really doesn't matter what each individual stock does so long as together they converge such that the weighted sum of their prices bounces around within a few cents . In this way, I can take my loss on a few stocks so long as I take a gain on the others, and take a gain on the "black box" as a whole. Having said that, do they need to be convergent, or can stocks be selected randomly to create the box? The algos will have me largely bettered on individual stocks, but can they predict and better my random selection? One stock goes up, the other goes down, one goes sideways, together they add up to a sum that can be traded as an individual stock.