Price cannot be predicted accurately for the simple reason that the actions of people, who ultimately control price, are not themselves predictable.
Most rules have exceptions, and here is the exception for this rule. Sometimes people's behavior CAN be predicted, once there is a trigger.
One would be pretty silly to "predict" that a fire is going to be started at such and such a time in a theater, but I WOULD predict that IF a fire started in a theater, most of the people there would get the &$&% out.
As for the market, predicting that XX stock will be XX price in XXX days/months is silly. No one knows what is going to happen in the near future.
But if a stock gaps up because of great earnings, and then does the unlikely thing and go immediately back down, it would not be foolish to ascertain that the stock is probably going to keep going down because the buyers are now "trapped" and going to want to get the &$&% out.
The best moves are fueled by emotion, not by logical "prediction." That is why the academics--and I am one myself--often get the market wrong when they say that the market is "efficient" and, therefore, no one can in the long run succeed in beating it.
The market is efficient most of the time, obviously. Perhaps it is "efficient" 99% of the time. But it is in that 1%, the sweet spot of the markets, that contain the opportunities to succeed.