I know very little about counting votes, but what exactly is the point of a recount like this? It sounds like they're literally just repeating the process of counting the votes, in the same place, with the same people, and with the same precautions. If that's what they're doing, that won't uncover any bias or fraud. They'll get exactly the same result every time, within the margin of human error. Which explains why recounts never change anything.
Wouldn't it make more sense to start with a spot check, where they pick a few precincts that might be suspect, and then randomly pick a few more as a control, and then have everyone originally involved go through the same process they did the first time, except this time they bring in outsiders from both parties to repeat and thus validate every single step? I.e. no "observers" -- have every last vote handed from the original counter, to an outside member of the red party, and then to an outside member of the blue party. Tally up the results separately, and then compare them to the original results. That would allow them to vet exactly how the procedures differ from precinct to precinct, and by comparing the original vote count with the second closely watched count, and the two outside counts, they'd have a real chance of spotting unequal treatment or fraud. And if that unearths widespread problems, then the state should either do a full recount with a completely different crew under much more stringent conditions, or refuse to certify the vote.