Dogmatic belief =/= religion. If that were so, everything we believe would fall under religion, since we form a belief either without thinking about it or on the bases of an arbitrary judgment. You might argue that probability of truth-hood makes some beliefs superior to others. But probability is based on the assumption that the future will repeat the past--an assumption that always proves false, or time would stop--as well as faith in one's own experience as an accurate measure of unexplored realms--another assumption which proves false, otherwise "unexplored" would mean nothing.
Probability, in other words, exists only to the extent that things freeze in place and cease to change. But nothing ever ceases to move; it's only in our heads that things screech to a halt, and even there, you'll find a lot of movement.
So probability exists in your head and only as a projection of the past into the unknown. No such projection is bound to accuracy, for if it were, the unknown would not be the unknown. Nor is there any reason to believe that your projection will even resemble the future. If it doesn't, then you revise your probability calculations, that's all.
But you'll seek for a binding rule that predictions must have some truth in them; why else would we possess them? They can't be in vain, surely. You'll look to those who show a knack for using models or relying on their intuition to predict the future. But that only brings us back to point one: you've assumed that the past prescribes the future. A model can fail in an instant, and the greatest foresight can give way to blindness.
What this tells me is that probability is more like a stockpile of past experiences arranged into a subjective pattern that suddenly breaks off at the point where the future begins. It's more like a filing cabinet that arranges memories into a useful system than it is an oracle.
What this means, then, is that choosing to believe something will prove true rests on the plainly false assumption that your pattern will go unbroken. Your beliefs are not only uncertain; they're bound to prove false, for this is in the very nature of time. The only upshot is that you can nevertheless find satisfaction in what lies ahead; perhaps certain variations of the past's rhythm, for example. Either way, to believe is only to point to a thought of yours and on that basis make a doomed prediction.