[Nitilism]

You can't make fish last longer than they otherwise would and keep it hidden from a regular's database provided he gets a sufficient number of hands. It's impossible to hide the fact that his good hands are not holding, or that his KK hands are bumping into AA more often than they should. Countering this point, I do not how feasible it is to say the typical high volume player can pack in enough hands to get the sample size he needs.

MyTurn2Raise claims to have shown that on pokerstars, the best hand in preflop all-ins only collects 95% of its expected value, you can read about it in the zoo. No one is jumping up and down about it, which is weird, because if the the math is right, then wouldn't that be proof positive that stars is not on the level?

I add that if you are going to say one more site is more likely to be rigged than another, it would be the site that doesn't allow datamining over the one who does. The samples you need would have to be extremely large, to the point that probably only leatherass and/or that Harry Potter kid from Italy could get enough hands on their own to do any meaningful analysis.

But let's be realistic here--in order to hide your rigging so that it won't show up in the average high volume player's data, you'll be rigging your game at such infrequent intervals that I doubt it would be worth it at all from any perspective at all, considering there a bazillion other ways out there to bilk the players. For example, stars FPP satellites are a great example. Give the player rakeback, but then give the player means to ensure the rakeback stays in the system. And while we are on the subject of tournaments, don't those already do a great job of ensuring a few hundred more fish are going to be rolled for games they have no business playing in every week?