Hi edgeman, welcome to WW, sorry to hear of your renewed misfortune.
I never recovered, so I wouldn't be able to contribute to the project you propose, but it would be very interesting to hear the results if you do get a discussion going.
Can you share the general nature of the changes you made?
It seems like you guys must be riding right along the edge of some sort of threshold. If we can figure out what that is, it might indeed be useful.
As for me, traffic is dead flat, no recovery evident. Changes made since last iteration were mostly text-content-oriented. I fixed lots of duplicate descriptions/captions, fixed a setup that was resulting in fragmented sentences in some captions, that sort of thing. I also made some further changes to user experience -- but I have to admit, for the vast majority of pages my user behavior metrics have't changed noticeably. I'm leaning more and more toward the theory that user behavior is at least part of this if not most of it.
It would be interesting to see if those who recovered noticed any changes in engagement metrics during the time of recovery or the time of re-pandalization.