Because the variance (or any other measure of spread) is a useful explanatory variable that is ignored and left on the floor. Switching from mean to median, wh

Because the variance (or any other measure of spread) is a useful explanatory variable that is ignored and left on the floor. Switching from mean to median, while more "robust" doesn't actually capture something new, just something old slightly better.