Jarmo: The way I see it, the visualisation bit was just the easier way to show what was going through his mind. e.g. "I'll open with a kick taking "a" out of the picture, "b" will no doubt strike with that baton so I'll dodge and split his jaw with a straight punch, "c" has a gun in his pocket so I'll have to be ready for that as well.. GO!"
That's remotely plausible to me. I don't see it as he'd be actually visualizing every fold of clothes or anything, despite what the movie shows. Obviously it's done very flashy, showy kind of way to go with the visual style.
The whole concept actually comes from samurai legends tradition, where two opponents just face each other down, and
know how the fight's going to play out.
The way I see it, some predictions are more "micro" in style and others are more "macro".
Without making a precise prediction about the specifics, you can make a broad prediction about an outcome by knowing that a set of smaller less predictable "micro" outcomes will converge torwards that more generic macro outcome.
An individual fighting move, in all it's precision is not one of those macro outcomes, because there are just too many permutations (in fact, a near infinite number).
If you're really clever (Sherlock Holmes clever), you'll make a near totally accurate prediction of the next move and a fairly accurate prediction of the next 2-3 moves maybe, but here, he was making a near totally accurate prediction of an entire fight sequence totalling over 10 moves (sometimes closer to 20 in group fights).
They say that battle is a chaotic, unpredictable thing, for a reason.
He's super-human in his ability to fight and while I was a lot younger (try older kid to young teenager) when I watched the old movies, I remember enough to know that old movies Sherlock didn't fight like that. He was a lot more "good, but human" in his fighting prowess.
EDIT: Ignore the poster above. I'm on my father's computer atm and it should have clicked that he was the one who was logged on.