A pitch traveling 90 mph as it approaches the front of home plate is traveling more than 40 thousand millimeters per second. How well can a person standing 2-3 feet in back of home plate estimate when the ball first enters the strike zone?
And I'm not convinced it can be done perfectly on a baseball field with electronic means.
I agree on both the difficulty of the person estimating and the technical challenges with the automated technology.
And you've stumbled upon one of the issues with the electronic means. The strike zone in the rulebook is defined as a 3D shape and if any part of the ball touches any part of that area, then it's a strike. The automated systems use a variety of methods, but they're all in 2D. The official one that MLB uses measures the ball as it is across the middle of the plate. So if a ball is knee high at the front of the plate, but 1mm below that in the middle, the automated zone considers that a ball, but by rule it should be a strike (same thing with a high pitch that drops just as it gets to the back of the plate... looks high to everyone, but might be a strike by the book).
As an umpire with proper training (and I have officiated a few pitchers who threw 90+, although not a regular basis), I can tell you that I CAN tell if any part of the ball was over the width of the plate at any point while it went past. Judging the height as it goes through there, however, is much more difficult, and we're left with having to use some clues to determine that (basically where the catcher's glove was when he caught it... for example - if we think it may or may not be too low and we see a catcher raise his glove, we assume he thought it was low and call it a ball - if he leaves his glove where it is, and it's at/above the knees then it's not too low and it's a strike).
