The term "regression to the mean" suggests time series and agency. But it's just about what happens with two correlated variables, if the correlation isn't 1. I suggest the alternative name: "a stupid prediction," as in "Because the parents' height is a stupid prediction for the children, the children of tall parents are not as tall as their parents." Perhaps it would be less vulgar to use "naive."

That's just a suggestion. Perhaps someone can think of a better name, but I think putting the focus on prediction is the key to a better name.