The difficult part of Wasserman's math as well is that an increase in turnout in a larger county with over 1M is going to mean significantly more votes than it would a county of about 80k (like Rockingham). So, for instance, if Fairfax County is at 106% of 2017 and Rockingham at 117, without doing the math, that still means Fairfax is turning out more raw votes over the last election than Rockingham is.
The question is whether the same pattern in Rockingham holds in other deep red counties in Virginia, and that we really won't know until they count the votes. I suspect he thinks that's the case, there's reason to believe that could be the case, but still. We don't know.