Faith is usually improper methodology, imo. It doesn't provide a good mechanism for fact-checking or removing errors. The result of it is that people in different places all believe different things, and entirely different views of metaphysics are believed.
Contrasting this with science, or believing that which has sufficient evidence, knowledge is more certain and uniform, and growing. For instance, different cultures don't all have various opinions on what the speed of light is, but they do all have different opinions on what god is. Weeks ago when scientists announced they discovered neutrinos traveling at superluminal velocity, a portion of the worldwide scientific community jumped at the data and came up with explanations, such as one of the more likely scenarios of improper measurement with the GPS satellite. Proper methodology and fact-checking can increase knowledge and rule out errors.
If there aren't enough facts for something, then probabilities should be assessed, or the evidence needs to be analyzed and understood to still be a question rather than an answer. Weak varieties of evidence can and should lead people in the direction that they imply, but faith doesn't add anything to that scenario.
I think their argument is "If I'm correct, then infinite gain. If I'm wrong, then no loss."
However, they're not considering many actual and potential losses -- such as any time, effort, and money invested into the faith-based belief; or the possibilities they're failing to consider where some other faith is actually the "correct" one and that by believing in the "wrong" faith they've offended some other actual god(s) and will therefore suffer for it (rather than having "no loss.")
They may think "If I'm right, infinite gain. If I'm wrong, no loss." But it's actually something more like, "If I'm right, infinite gain. If I'm wrong, anything between no loss and infinite loss." Sounds like very poor odds to me -- and the height of irrationality to ignore critical thought processes when such stakes are the case.
It's interesting how widespread Pascal's Wager is, either by name or just by agreement.
Actually, the analysis is "if I'm right - gain, if I'm wrong - still
gain." The only way to lose is if one of few flavors of religion that insists they are the only way that is right, really is.
If I'm right, I've lived my life in a manner aligned with God's will. If I'm wrong, I've lived a good life and done things that build up myself, my family, my immediate community, and my world.
Some of this presupposes that you would not do these good things if not for your belief, which I find unlikely.
Religion influences the ethics of culture, and culture influences the ethics of religion, because religion is intertwined with culture and an aspect of it, but considering there are ethical and not so ethical varieties of all groups of people, I don't see how faith is a win/win scenario at all.