🧵Woot! I finally debunked @GAballots categorising as a failure the counting of 909 Fulton ballots (documented on one batch sheet) in the 2020 Georgia hand count.
The batch failed according to his formula which checks for differences in tallies for Trump, Biden and # ballots.
Context:
As with HUNDREDS of batches David erroneously marked as fails, he failed to allow for write-ins and undetermined ballots; they are separated from the batch and sent to a Vote Review Panel to be adjudicated, often entered into the CSV separately.
docs.voting.works/arlo/full-manu…
10 batches (323-332 from scanner 1) were counted together and recorded on one audit batch sheet.
That's okay, because audit tables received CONTAINERS of batches of ballots.
Manual: "Pull the ballots out of the container and stack neatly on the table."
docs.voting.works/arlo/full-manu…
David marked this batch (sheet) as a Fail for 3 reasons:
- Trump tally didn't match the ballot image tally (1 less in RLA CSV)
- Biden tally didn't match the ballot image tally (5 less in RLA CSV)
- total number of ballots didn't match (893 in CSV, 909 ballot images) = 16 diff
However, the audit batch sheet shows that 16 ballots were sent to the Vote Review Panel:
10 write-ins
6 undetermined
That accounts for the 16 ballot difference between the ballot images and the total tallies entered into the CSV from this batch sheet.
I looked through the 909 ballot images and found the 10 write-ins.
(Cropped and labelled here for convenience.)
ballot images:
dropbox.com/sh/v0njezpupx8…
That left 6 unaccounted 'undetermined' ballots to find.
I went through the 909 (!) ballot images again and tried to think like a fussy auditor: I looked for marks that weren't basic marks in ovals.
I found several ballots where the oval was filled but also had crosses or lines.
Variations found:
1 Trump with a diff write-in crossed out
1 Trump oval with X
1 Trump oval with /
3 Biden oval with /
5 Biden ovals with X
It now seemed likely that the 5 Biden ballots with Xs through ovals but only 1 of the 3 Trump ballots were sent to VRP for adjudication.
To test my theory, I looked through Fulton's Vote Review Panel sheets to see whether 10 write-ins & 6 undetermined ballots were ever received.
Counters are meant to write the batch name on the envelope of ballots going to VRP. But there were no obvious candidates for scanner 1
However, page 71 in the PDF document of VRP sheets showed 2 "und" rows:
+ 10 ballots:
7 invalid write-ins
3 valid write-ins
+ 6 ballots:
1 vote for Trump
5 votes for Biden
What are the chances, eh?
I declare that these two 'und's are for Scanner 1 batches 323 - 332!
Hurrah!
Sources:
scanner 1 ballot images:
dropbox.com/sh/v0njezpupx8…
ballot images by filename:
ballots.youpeople.org/?wide=true
audit batch & VRP sheets:
sos.ga.gov/page/2020-gene… (Fulton)
Hey, @GAballots, this is what #transparency looks like.
Share this Scrolly Tale with your friends.
A Scrolly Tale is a new way to read Twitter threads with a more visually immersive experience.
Discover more beautiful Scrolly Tales like this.