The serve speed on your screen is measured at the wrong end

Radar captures the ball near the racquet, and by the time the returner touches it a large share of that speed is gone.

A hundred and thirty on the board looks enormous. It is also a reading taken essentially at contact, and the ball the returner has to deal with is nothing like that fast. Air resistance strips speed continuously across the flight, the bounce takes a further large bite depending on the surface and the spin, and what arrives is a good deal slower than the headline figure.

So the number is honest about one thing and misleading about everything the viewer cares about. Two serves clocked identically can present completely differently. A flat one hit through thin air on a slick court keeps more of its speed and skids low. A heavy kick serve at the same radar reading has already spent energy on rotation, sheds more of it in flight, and then converts a chunk of what remains into a vertical kick off the bounce. Same figure, entirely different problem for the returner.

This is why I have very little patience for lists of the fastest serves ever recorded. They tell you about calibration and about the moment a player decided to swing out of their shoes on a dead point. They do not tell you who is hard to return.

The metric that would actually be interesting is the one broadcasters have the tracking data to produce and mostly do not. Give me the ball's speed as it crosses the returner's contact point, and give me the time from the server's contact to that moment. That second number is the whole game. It is the returner's available decision window, and it collapses when a big server is standing inside the baseline on a fast court against a returner who has chosen to receive close in.

There is a related distortion in how the second serve is discussed. Coverage frames it as the safe one, the one you take pace off. Many of the best second serves are not slower in any meaningful physical sense. They are the same racquet head speed redirected into spin, which buys net clearance and a violent bounce rather than raw arrival velocity. The radar reads it as timid. The returner does not experience it that way at all.

The gun measures the swing. The point is decided by what is left of it.