I have to believe Fermi would have noticed this pattern, but there seems to be little comment on it either in the paper, or in the linked article. Perhaps the reason for it is so obvious that he thought it not worth publishing. I just wish I knew what it was.
This is an unrelated but nice quote:
> "In the compound pendulum problem, one has a transfer of energy from one degree of freedom to another and back again, and not a continually increasing sharing of energy between the two. What is perhaps surprising in our problem is that this kind of behaviour still appears in systems with, say, 16 or more degrees of freedom."
Also worth noting that the graph isn't perfectly symmetric but this is presumably just due to accumulation of numerical errors.