> I doubt I'm following your metaphors/allusions, but the parallel may be that you're only hoping for any portfolio to have excess performance, and not the sum total.

I'm sorry but I don't know how else to explain it to you when you're ignoring such an obvious fact: if you're hoping for any portfolio to hit big, but they're all picked without any intelligence (i.e., without an oracle) then you will always be only as good as the worst portfolio. It's just such a patently obvious logical argument I don't know how else to state it so that it becomes clearer. I mean I could easily write out the formula for expected value of the max of N uniform iid random variables but you've literally been witness to it by watching the wall-clock time on your threads!

Let me put it without any metaphors/allusions: you simply don't understand search and/or SAT if you think that dividing a 2^N search space into N pieces will give you a speedup. It will not. It's basically the "geometric" definition of NP.