Basically, without going into a lot of detail, subject matter experts are generally very good at narrative construction, but shockingly bad at predicting the outcomes of policy goals and decisions. The primary case study that spawned the field of research involved asking academics and assorted Federal employees - many with access to classified information - to make predictions about the near and further future of the Soviet Union. They were all wildly wrong. The people who made the best predictions are generally intelligent, educated people who aren't experts in the relevant subject and are generally characterized by open-mindedness. This has been borne out in a number of fields since.
Although, reading your original statement again, it was probably true. Trump supporters by and large are less likely to be well-educated, and I would guess the average Trump supporter is anything but open-minded.