There isn't just one ERV. There are thousands of them, in humans. Your infection-to-fixation-to-presrvation scenario must be played out thousands of times to produce the pattern you see. Not possible.
And you can't say that your data 'exactly matches the phylogenetic tree'. The 'tree' doesn't exist. The 'forks' are all imaginary and all of the bushy leaves haven't been sequenced.
You are way over-extended.
Once every 5,000 years for 5,000,000 years is 1,000 infections getting fixed in the genome. Not obviously impossible. What's "preservation"?
And you can't say that your data 'exactly matches the phylogenetic tree'. The 'tree' doesn't exist.
Huh?
The 'forks' are all imaginary and all of the bushy leaves haven't been sequenced.
OK, I should have said that the tree of presence/absence of genetic markers is always a subtree of the already-known phylogenetic tree.
But the trees produced by the different genetic markers are always consistent with each other and with the phylogenetic tree.