[alanbrown]

I think that one reason that moves might be -EV in a cash game but +EV in a tournament is because people can always reload in a cash game. So I think part of Snyder's POV is that there are many +EV situations that come about because of the increased FE of an AI in a tournament.

I believe a lot of what Snyder bases his theory on is FE and I'm a big fan of that theory. He talks about preying on the tighter players also but the largest part of his point is that you need to milk that FE for all it's worth.

I think in a lot of these discussions people presume (for the sake of the discussion) that they can accurately assess the range of their opponent and the way they'll play that range depending on your action. I think part of the reason why I agree with Snyder's principles is that people assume ranges that are too tight for villains and they assume villains will call looser than they actually will (villain will make bad decisions in actual fact because he'll misread your hand frequently). Therefore I believe Snyder's plan works primarily because of people underestimating FE and because they overestimate villains perceptions and skills.

I actually don't think Snyder advocates calling with 44, but he definitely advocates pushing with it in a wider set of circumstances than most would.

As for your point about variance. It's generally accepted that if your finishing position at SNGs were graphed they should make the shape of a smile. More 1's and 10's than any other number. That says to me that increased variance does indeed lead to increased tournament winnings.