In effect, Pascal’s wager states that while we can’t know with absolute certainty whether the christian god exists, a rational evaluation should lead to a belief. If having to choose between believing (in the christian god), or not believing, the reward for being correct, and the price for being wrong, tips the balance in favor of believing.
It says, if you believe and are correct, you will gain heaven, while the price for being wrong is nothing. On the other hand, if you don’t believe, it says you will gain nothing for being right, yet lose everything if you are wrong. So, belief results in a win/neutral, and non-belief in a neutral/lose position, tipping the balance clearly in favor of the “belief” position.
Why Pascal’s Wager is a fallacy:
a) Pascal’s Wager assumes that there are only two options.
b) Pascal’s Wager assumes the christian god doesn’t care whether someone actually believes, or simply goes through the motions.
c) Pascal’s Wager discounts the price paid for belief before death.
d) Pascal’s Wager vastly overestimates the odds for the reward and the risk of punishment.
Positing only two options is ridiculous. There are, of course, thousands of possibilities when it comes to gods. Based on the evidence available for these gods, it is not reasonable to assume one is more likely than any of the others. To increase the odds of a positive outcome of this wager, the believer would have to believe in, and worship, every possible god. Including the ones that haven’t been invented yet. Aside from the drain on the available time, it presents the problem that quite a few of these gods are pretty selfish. They frown upon believers believing in other gods. In some religions that is enough to not be eligible for the reward (making the belief position a lose/neutral one).
Also, just going through the motions and pretending to believe may fool your community, but it can’t fool an all-knowing god. It is very unlikely that anyone would gain the ultimate reward for simply faking belief (making the belief position a lose/neutral one).
The price paid for the belief position isn’t nothing. It involves going through the rituals, day after day, week after week. It may have severe side effects on physical and mental health. Sex life suffers, too.
In estimating whether the cost of any given action is worth it, an evaluation of risk versus reward is in order.
Risk is (simplistically) the chance that a negative event occurs, multiplied by the cost of that event. As an example, being hit by a meteorite carries a very high cost (probably death), but since the odds are extremely low, the risk associated with it is low. Similarly, the chance of getting rained on is pretty high, but the cost is very low, representing also a low risk. On the other hand the cost and chances of, and therefore the risk associated with, a traffic accident are high.
The choice whether to mitigate a risk depends on, among other things, the severity of the risk, the cost of the mitigation and the tolerance of that risk. In the above examples, the cost to mitigate each risk are; exorbitant, low and high, respectively. Methods to reduce or eliminate the risk of meteorite impacts are cost prohibitive and far exceed the risk. An umbrella and a check of the weather forecast effectively mitigate the risk of getting rained on, and is easily worth the cost. Car crashes, and their after-effects are mitigated to various degrees by expensive technology (from street surface technology to driver training, airbags and traction control). People bear those costs to their financial ability and tolerance for the risk.
A similar reasoning applies to reward. The choice whether to pursue a reward is guided by the perception value of the reward, the perception of the odds of gaining the reward and the cost to pursue it.
In the belief versus non-belief question, believers tend to irrationally overestimate both the reward for belief, and the risk associated with non-belief.