“Tweeted out a poll” An opinion poll about this = nada.
When handled per AZ law, a messed up ballot that can’t be fed into the counting machines, is given a serial number. In the presence of multiple observers, an election official is supposed to mark a clean ballot with the same votes as the spoiled one, and write the same serial number on the “duplicate” as is on the spoiled original. The original goes into one pile for storage, the duplicate goes for counting.
At the Senate hearing, the auditors described finding numerous “duplicate ballots” with the required serial numbers written on them. Reportedly many “spoiled ballots” had no numbers (or just a 2 digit number that didn’t fit the serial numbers’ multi-digit pattern). This thwarted the intent of the law, which was that originals could be matched to their replacements, allowing the accuracy of “duplication” to be checked.
They did not say, in the parts I heard, that there was a numerical shortage of spoiled ballots versus duplicate ballots. It would seem that auditors could easily isolate and total the raw number of total ballots in the “spoiled” stack and in the “duplicates” stack. Count the votes for each candidate on the spoiled originals and the same for the duplicates, then compare the two aggregate totals for discrepancies. I’d expect some very small variance, no matter how hard people and machines “try” to get it perfect. But more than a few off, and that signals a major flaw in the canvas.
Has anyone seen credible reports of the total number of spoiled originals versus duplicates? I’d be very interested in seeing those.
I haven’t seen any reports what you speak of. It’s all going to be revealed though when the audit is finished.