Have you looked at Pekar's full model, as set out mostly in the supplementary materials? This isn't any standard molecular clock approach. It's a byzantine stack of plausible but somewhat arbitrary assumptions, ending in a simulated phylogenetic tree. The shape of that tree with one introduction doesn't match the shape of the actual tree constructed from the earliest real samples in Wuhan, so Pekar concludes there were two introductions. But I'm not aware that such an approach has ever made a successful prediction, and there's no circumstance in any field where I can imagine trusting a model of such complexity without validation. Their sensitivity analysis is meaningless, varying some irrelevant parameters but keeping what seems intuitively like the main determinant of that shape (the connectivity of their contact network) fixed.
You are correct that Alina Chan's thread doesn't address that aspect of Pekar's argument. Others have though, per the Twitter threads I linked. What do you think?