It's a well-known logical argument why you should believe in God, even if there's a strong chance that it might not be true. Simply put, the argument is that you should believe in God just because there's a chance that you might go to heaven and avoid hell.
Blaise Pascal, a philosopher and mathematician in the 17th century, first formally put the argument forth. He is considered the founder of probability and he made other significant contributions. There's also a programming language named after him.
Pascal's wager, in a nutshell, is this. No one knows for certain whether God exists. Maybe he does, maybe he doesn't. It's a gamble whether you believe in him or not. So let's treat it like a gamble, says Pascal, and look at the odds.
He described the payoff of this gamble like so. If you choose to believe in God, and you happen to be right, then the reward is infinity: eternal bliss in heaven. However, if you are wrong, then you lose nothing at all. On the other hand, if you choose not to believe in God, and you're right, you GAIN nothing (in either of the previous two cases, you just die and that's the end). But if you are wrong, your payoff is negative infinity: eternal suffering in hell.
Now here's the main thrust of the wager. Since the chance of God existing is unknown, but the payoff/punishment scheme is infinitely in favor of believing in God, just on the small chance that he might exist, you'd better believe. It's the only wager that makes sense.
Okay, that's Pascal's wager, now here are our reasons for not agreeing with it.
Reason 1: In the case where God does not exist, there really is a clear advantage to not believing. In other words, the payoff is not zero. For one thing, if you go through life believing a lie, that is a bad thing in itself. Besides that, there is more to being a believer than just saying "Okay, I believe now" and getting on with your life. Serious believers spend a lot of their time in church, and contribute a lot of money as well. There's a reason why some towns have very affluent looking buildings for churches, and why large and elaborate cathedrals are possible: they're funded by folks who donate 1/10th of their income throughout their lives to tithing. This is surely quite a waste if the object of worship isn't real. That's to say nothing of the persecution of other groups that's been instigated in the name of God throughout the ages.
Reason 2: Even if you buy into Pascal's wager and decide you should believe, that doesn't give any basis for choosing which religion to believe in. Fundamentalists often use the wager to prove that you should be a Fundamentalist, but of course, Pascal was Catholic and was using it to prove you should be a Catholic! This just highlights the whole problem of which religion is the right one. Since many Fundamentalists believe that Catholics are going to go to hell, Pascal's not much better off than an unbeliever. We don't know if the Jews are correct, or perhaps the Muslims, or if reincarnation is right... or worse, if there's a perverse God who only lets atheists into heaven! It's not impossible. For all we know, maybe God exists but he doesn't care at all whether people believe in him.
Reason 3: If you can accept Pascal's wager as a realistic reason to believe, that leads you to a point where you have no choice but to believe just about everything on the same grounds. Maybe if you don't own a complete library of Seinfeld episodes, you'll go to hell! Why not? You don't know. Maybe you have to send $10 a week to the Atheist Community of Austin for life. Hey, what's a measly ten bucks if it will save you from eternal hellfire? Or maybe God really likes nude mud wrestling and he will punish those who do not partake of His gift.
Does all this sound utterly silly to you? Good! That's probably because you know that you should only believe things that have some sort of clear evidence favoring them. You don't believe just any old preposterous claim about UFO's, pyramid shaped get-rich-quick schemes, or magic pixies just because somebody tells you they're true and because there's a chance you might be wrong. You have a brain—use it!