You're playing with words. Predictions (Metalwanker's word, not mine) can of course include statistical prodictions basded on probabilities. Which is precisely what the Insitute for Fiscal Studies and god-knows-how-many other institutions, analysts and experts have done.
No, the problem is not with the word "prediction" or it's suitabiliy or not. The problem is that the concensus position is that in all likelihood, Brexit will be very bad for our economy, for jobs, for inflation, prices, peoples' wealth. The Brexiters who are idealogically sold on leaving, can't stomach this likelihood so we have all this *shite* - and it is shite - trying to pretend the predictions, forecasts, calculations and in some cases cast iron certainties, are wrong.
I'd have a modicum of respect if they actually said, "you know what, I think we may well be worse off for a number of years, but in the end it will be worth it". I'd fundamentally disagree with that assessment, but at least it would be an honourable position.