I agree that God could exist, and that my tendency to see things as products of unguided natural processes (rather than as deliberately created) is a result of the fact I would rather live in a naturalistic world than in a world where there's supernatural stuff going on that I can't understand. (And history reaffirms my feelings on this, because all kinds of things used to be attributed to God until science discovered the natural processes behind them). Conversely, I insist that what anchors most theists to theism is their preference to live in a world that has meaning and purpose, part of a narrative, where lives move towards a divine goal. This view may or may not be justifiable from the evidence.
But when it comes to believing that God is talking with you, that your prayer can cause someone to change their mind or recover from a disease, that a "religious experience" really means you're connected with the divine, then those things I feel can "be dismissed as the rantings of lunatics or the silliness of uneducated or even the misguided ideas of folks who should know better".
Without being picky, I would list the following questions for all of us to ponder out of Bernardo's well written response:
1. Does believing the world to be a naturalistic world result in more or less confusion about "stuff going on that we can't understand?" Does free will get clearer? How about origins? Purpose? Bad things happening to good people? Horrific diseases, death, and destruction?
2. Believers clearly prefer a universe with purpose and hope of justice, including heaven and hell. Does this mean atheists prefer a random universe where the sun might choose to come up in the West tomorrow or gravity might exert 45 pounds/sq' on earth?
3. How is it that those who don't hear from God, trust in their prayers to be answered, and feel connected in a personal way to the Creator of the universe set themselves up as clearer thinkers than the rest of us. I generally don't doubt the clarity of thought of those who see things differently. Merely their conclusions. And generally I will even regard their conclusions as having some percentage chance of being superior to mine.