Assuming this is correct, O'Connor will need about 71% of all these provisionals in order to win (roughly 6,000 of the 8,400) and considering a solid # will likely be thrown out, probably closer to if not over 100%.
Remember: that number is not just provisional ballots but also unreturned absentee ballots that theoretically could have been postmarked in time to be valid. That means a significant chunk of the absentee ballots are likely not going to be returned and a significant chunk that are will be for Balderson (mailed ballots overall are historically more Republican than election day ballots, no idea about those postmarked in time but not received by election day). I haven't read how many are actually provisional ballots, which should be heavily O'Connor if valid (which also isn't a given).
So, yes, O'Connor has a theoretical path, but it's extremely unlikely he wins based on the outstanding ballots. What's probably more likely at this point (though still quite remote) is some sort of counting mistake that gets uncovered in the next couple days.