The accuracy rate of polygraphs is high. But this accuracy depends as much on the examiner as on the equipment. A technician who develops inappropriate questions and reads the results badly may reach a result of "inconclusive", even a result of "deceptive", when the person may be telling the truth. I think we should be wary of any evaluation process that depends on the evaluator, that cannot be replicated by another examiner every time.
Similarly with psychological examiners. Some people consider psychology and psychiatry "junk science" and frankly, I am almost there, almost ready to join those ranks. Two different practitioners can come to totally different conclusions about the same person.
Why do we take these evaluations so seriously? We need to put them in context with other evidence and not give them too much weight.