Alan Nathan is Professor Emeritus of Physics at the University of Illinois at Urbana-Champaign. After a long career doing things like measuring the electric and magnetic polarizabilities of the proton and studying the quark structure of nucleons, he now devotes his time and effort to the physics of baseball. He maintains an oft-visited website devoted to that subject:

In response to a question posed to me yesterday about late break on a fastball, here is my reply:

I'm not sure I really have much to say about the subject of late break. What I do have to say is summarized in the figure below (click to expand), showing Mariano Rivera throwing his cutter.

The solid curve is the actual trajectory, taken from PITCHf/x data. The dotted curve is a straight line projected from the release direction. The two curves look pretty nearly identical until about 20 feet from home plate, which is just about the point of no return for the batter. The ball deviates from the straight-line path at that point and ends up with about 0.5 feet (6") of movement, so the pitch that looks like it might hit the corner ends up 6" off the plate (and breaks the bat of a left-handed hitter, a rather common occurrence for Mo).

So, did the ball break late? No, it did not. The break is actually continuous, starting from when the ball is released. However, in this case the break was so little that the ball did not show any appreciable deviation from a straight line until the point of no return.

What does physics have to say about this? Physics says that the amount of break (assuming a constant force that causes the break) is proportional to the square of the flight time. Or in our case, proportional to the square of the distance from release. In the present case, the total deflection of the ball from 50 feet to 20 feet is about 35 percent of the total break from 50 feet to zero feet. The latter is 6", so the former is about 2". If you look very carefully at the trajectories in the figure, you should be able to see that there might be about a 2" difference between the solid and dotted curves at 20 feet. PITCHf/x tells us that essentially all pitches should have this behavior: Namely, the deviation from a straight line between 50 and 20 feet will be about 35 percent of the deviation between 50 and zero feet.

So, what is late break? In my view, late break actually means "not much break," meaning that the difference between no break and actual break is small enough at 20 feet that the batter can't perceive it. That is, 35 percent of the total break is too small to perceive. The Rivera pitch is an example of that.