"there is no reason for X to be exactly described by a simple analytical function" is not what you want to read at the start of the chapter on measuring and modeling X