Other questions that make the argument very little compelling, if at all, are: how do we define civilization? Do we all belong to the same 'civilisation': Sumerians, Etruscans, Egyptians, and so on?
It has been argued that civilisation defined by cities, writing, and monumental architecture, came about as a consequence of the end of a glaciation.
Civilisations in a narrower sense like, e.g., the Minoans, probably disappeared because of a volcanic eruption. Others, like the Egyptians, because of people from the seas invading lands that were essential to their trade; as to the Mayas, it's debatable, but climate change may have played a role. It could also be for internal reasons... Earthquakes, meteorites, you name it.
I concur with @MigL that there are too many unknowns.
Also, I know you pitched this topic for applied mathematics, but we should try to make these ideas falsifiable, because we're concerned with science. You cannot experiment with civilisations as a subject of study in real time.
(My emphasis)
This is funny, because --and I think it's happened before-- you've made a good argument, but it seems to me that it works against your idea. How can we surmise that probability distribution?
On other threads I've argued that the word 'random' in itself doesn't mean much. You would have to make a statistical hypothesis => probability distribution.
Equiprobability doesn't apply in general.