Yes, by a wide margin
In Philip Tetlock's research, summarized in 'Superforecasting,' the people who assigned precise probabilities and updated them as new evidence came in were consistently and substantially more accurate than those who made flat yes-or-no calls.
