The fallacy is that if God is really omnipotent than espousing a belief that you don't really hold or only hold out of fear won't "fool" him, so you gain nothing.

I think his 'wager' comes with a bit more... He writes as well that belief is constituted by 'acts of belief'. That by following very formal rules of fate, praying, engaging in rituals etc, 'belief itself' will follow, or possibly that this is all there is to 'belief'? (Was a while since I read this...) The wager is a wager on belief, not a wager on God. (Or as well, in addition to a wager on God, one on belief?) It is not a matter of 'tricking God', but of assuming 'belief' which is already 'belief'?

I kind of like Pascal, in a funny sort of way. I used to quote him on belief when teaching a computational model in an intro comp sci class. It turns out that most of my idiot students when introduced to the model would freak out, going on about not understanding it, etc. "Explain, explain! Explain again!" they would shout. I would tell them, that the way one 'understands' such a model, is to blindly assume it, work in accordance with its rules in a step by step fashion, and through the process of enacting the formal parameters of 'understanding', 'understanding itself' will follow... I am not sure the students adequately appreciated the cleverness of Pascal. Ungrateful little bastards, they were!