Horizon Problem

A simple demonstration of the horizon problem is that light from opposite sides of the CMB sky have only just reached us, so that light (or any signal) cannot have connected opposite sides of the CMB. So why are they so similar in properties?

The situation is rather worse, since the horizon size at the time of the CMB was only a degree or so, so regions separated by more than 1 degree were outside each other's horizon -- again, why are they so similar.

Figure from Whittle's Teaching Company Course, and upcoming undergraduate text.