In this 1964 clip from the BBC Horizon show, Arthur C. Clarke makes a fairly precise prediction, but one that is only half right. "We'll no longer commute in cities," he says, "we'll communicate instead." He also says, "I am perfectly serious when I suggest that you'll be able to call a man and not know where he is, whether he is in Tahiti or Bali or London." He got that part right, with cell phones everywhere, but on average we still do commute in cities.
However it is the preamble to his prediction, where he hedges his bets, that I think he is the most insightful. Clarke says that if you find a prediction reasonable, than it is probably wrong, because the future is not reasonable; it is fantastic! But if you could return from the future with the exact truth about what will happen, no one would believe you because the future is too fantastic! By fantastic he means issuing from the realm of fantasy and the imagination -- beyond what we expect.
This is the futurist's dilemma: Any believable prediction will be wrong. Any correct prediction will be unbelievable. Either way, a futurist can't win. He is either dismissed or wrong.
Except if he hits that razor's edge between the two realms, right on the cusp between plausibility and fantasy, where it is almost true in the improbable future. This is the sweet spot that science fiction authors aim for. Occasionally one hits it. Like Arthur C. Clarke.
Getting it right is very, very difficult. Most people, particularly most smart people, even most science fiction authors, will err on the side of not being fantastical enough. Because absolutely no one wants to be dismissed. What's the point of making a prediction if no one is listening? So 99% of future predictions will fall short of the necessary unreasonableness for a correct prediction.