Yes. In some devices, the pad is split and then fastened together after the ball is inserted. This keeps the ball well-captured, and you can adjust the friction. It is expensive.
In most, the pad socket is slightly larger than the ball, and the edge of the pad is then deformed (essentially rolled) to hold the ball. There is usually not much holding the ball in, but you don't need much. This is fast, easy to automate, and cheap.
You could make one with a shrink fit, I suppose, but that would be both expensive and not as strong.
__________________
"Well, I've wrestled with reality for 35 years, Doctor, and I'm happy to state I finally won out over it." Elwood P. Dowd
and sometimes they heat the pad and super cool the ball, insert the ball into the socket and then let them set until the temperature equalizes between them. There is enough contraction and expansion between the two to grasp the ball.
__________________
Could be worse, it could be me. :)
I figure, the pad is put in an induction heater to about 250 C, just before things start changing in the metal molecules. The ball -40 C or so, easy to get with dry ice (frozen CO2) or liquid nitrogen. These temperatures are far enough apart to allow for enough expansion and reduction size to allow them, to be mated without too much push and keep together when at the same temperature.
__________________
Could be worse, it could be me. :)