Pascal's Wager is one of those annoying bits of illogic that gets stepped on repeatedly but refuses to die.
The wager states, as its base principle, that either there is a god, or there is not.
It then reasons that, if there is no god, it makes no difference whether you worship or not. If there is a god, it goes on, then it definitely does matter. Worship and you win, don't worship and you lose.
It concludes that you might as well worship because if you're right, you win, if you're wrong, you lose nothing.
The problem is that the wager is flawed in all three parts.
In the base principle, it fails to recognise that there have been thousands of gods worshipped by humans over the millennia. Also, it fails to consider that it's not limited to just the possibilities we've thought of. The choice is not between god and no god, it's between no god and an infinite number of gods - worshipped in an infinite number of ways. So how does one choose?
This leads us into the problem with the reasoning: How do you choose who - and how - to worship? You'd be in just as much trouble if you worship the wrong god - or, the right god in the wrong way - as if you'd never worshipped any god. In fact, atheism looks like the best choice - you may not have worshipped the right god, but at least you're sure you're not worshipping the wrong one.
The conclusion, though, is where the biggest problem comes in. Pascal's Wager was formulated to support Christianity, a religion, remember, that believes its god can see into your thoughts. Surely it's obvious that such a god would know if you really believe or if you're faking it thanks to a persuasive, if flawed, argument. I could go to church, I could pray, I could kneel and bow my head, I could eat the little crackers or play with the beads, but I cannot make myself believe. I do not believe and accepting the wager would be self-defeating, because I would be simply pretending to believe.
And I, poor little atheist that I am, don't think any god would be too happy with that.