Specifically in terms of being predictive you might be interested in reading Expert Political Judgment: How Good Is It? How Can We Know? by Philip E. Tetlock if you are not familiar with it already. He tried to empirically test experts political judgement and the main conclusion was pretty much everybody was bad. He now runs the good judgement project which trying to measure good forecasting skills to figure out who really is good at it. It was very eye opening to me at the time.
IIRC the broad conclusion was that "experts" are really specialized, whereas predicting political developments requires a broader awareness. Hedgehogs and foxes.
https://press.princeton.edu/books/hardcover/9780691178288/ex...