It's been a week since the election and, as of yesterday, 342,936 Arizona ballots remain uncounted. This includes 171,889 early ballots and 171,047 provisional ballots.
The senate race is slowly tightening, but not enough to suggest Dr. Carmona (D) could pull off the win. Flake (R) currently leads by around 80,000 votes. This 80,000 vote margin has remained consistent throughout the count, even though the percentages have changed.
As of this morning's update, Ron Barber (D) leads Martha McSally (T) by 512 votes. There have been a half a dozen updates since Saturday, and Barber has lead in all of them. On Monday, Cochise county reported 9,121 ballots remaining to be counted.
Cochise has consistently leaned toward McSally about 2-1 60/40. This means she will likely pull about 5,500 of those votes and Barber around 4600 3600.
Pima county has 4,000 early ballots and 27,000 provisional ballots.
Barber is pulling about 52% in Pima right now. So, assuming 70% of provisional ballots get counted, he should pick up around 11,900 to McSally's 10,900. This makes a projected total: Barber: 150,360
I've done quite a bit of rounding here, but this should be fairly close. This means Barber may just barely scrape out a win. The margin might be close enough, however, to trigger an automatic recount. My apologizes, I originally did my math wrong and added 1,000 vote to Barber's tally. With the corrected math, Barber may actually lose by less than 400. If there is less than 500 votes between them, that triggers an automatic recount. Sorry for that mistake. [UPDATE] This assumes all uncounted ballots in Pima county are part of CD-2. Pima county is not all in CD-2, so some, maybe even most, of these ballots may not have votes for this race. So this projection that I did really is not worth much. The most important thing is that the votes be counted.
It's best just to ignore any projections or assumptions I made about that count.
There is some very good news in here. The Chicago Tribune reports that, according to Arizona Secretary of State Spokeman Matt Roberts, if the count is not completed by the November 16th deadline it can still continue. Let's hope that is the case.
Only if you make the assumption, contradicted by all the data that have come in since the election, that the new EVs look like the average old votes, not like the old EVs, do you get such pessimistic results
Barber is consistently leading by 7% among Pima early voters.
He trails by only around 12% in Cochise early voters. The late-counted ones have looked just like the early-counted ones, but not at all like the polling place ballots. So continuing to apply the EV percents to the remaining EVs would leave Barber with about a 450 vote lead going into the provisionals.
What happens then with the provisionals? All lore, nationwide and in AZ, is that provisionals tilt strongly D, for obvious reasons. Even if you make the very pessimistic assumption that the provisionals then look like the EVs, Barber's lead then grows by hundreds when they're counted.