What do you think about a premise that if one of the gods of religion exists, it is more likely to be one that has been fairly successful or at least has not allowed his following to die out, perhaps on the assumption that such a god would be fairly powerless and not much worth following anyway. I suppose you could counter that such a metric of success would depend greatly on the time period you looked at. Still, I think it might be reasonable to assert that a god who let all his followers die out is less likely to exist than a god who has inspired his followers to successfully spread his name to every country on earth for two thousand years running. If so, you might at least restrict Pascal's wager to a few "major" religions and perhaps proceed from there.
Were that the case, however, it seems strange that the same being would demand more sacrifices of some than of others, or place stricter guidelines on one group versus another. (Though perhaps God is a behavioral economist, after all, and he's simply structuring different incentives for different groups, i.e., creating the frameworks the different groups need most? But, to your point, that wouldn't account for why some groups and gods died out altogether).
"So...which religion was the right one?" someone asks.
"Mormonism," he replies. It turns out Mormonism was the correct answer. Sorry."
Given that there is no evidence that the Christian God is more likely to exist than, say, Zeus, we must provide equal statistical weight to Zeus as to J.C. And so on for all the other possibilities, ad infinitum.
But it doesn't stop there. Some religions' tenets conflict directly with the tenets of others. So it can be said that there are multiple sets of compatible religions that can be held at a given time. Most of these religions are polytheistic. By contrast, adopting a monotheistic religion limits you to just one chip on the cosmic roulette table.
Ergo, from a statistical standpoint, choosing a monotheistic religion (such as Christianity) for the Wager is taking the worst possible odds.