The Sterlingov case Daubert ruling makes for an interesting read. We'll see what happens in the appeal. Its goal IIUC is to determine if evidence is admissible, not that it's conclusive. So the bar is a bit lower.

My initial impression after reading is that the defense tried to bite off more than it could chew, by trying to dismiss Chainalysis for all the things, rather than focus on the specific transactions that (allegedly) link him to the admin role.

It seems reasonable to expect that they can roughly trace flows between individual darknets and mixers.
https://storage.courtlistener.com/recap/gov.uscourts.dcd.232431/gov.uscourts.dcd.232431.259.0.pdf https://image.nostr.build/fa9d12758fb8cc39cf70caeab5ddac32aeaffa4af608ec33035011e739715005.jpg