No, it really does not work like that, for reasons that should be obvious.

Yes, credibility is one component of evaluating conclusions from evidence.