All eyes are on Georgia today as voters head to the polls to vote in primaries across the state to select who will appear on the general election ballot in November. While today's election focuses on the state's future, Georgia remains embroiled in controversy and discrepancies surrounding the 2020 general election, which saw a scant 11,779 votes out of nearly 5 million separating Joe Biden and Donald Trump in the final tally for the 2020 presidential election. With an irregular recount and a "nothing to see here" audit affirming Joe Biden as the winner in 2020's critical race, Fulton County remains a focal point in the crucial swing state's ongoing election debacle. One area in need of attention is the county's machine count.
Indeed, significant deficiencies persist in the recount of machine votes in Fulton County. With a detailed report in Gov. Brian Kemp's possession for the past two weeks, the governor's office has been verifying the data outlining the shocking findings presented by Kevin Moncla and Mr. Joseph Rossi. As his office continues to review the information, below is a summary of the Moncla and Rossi's report sitting on Gov. Brian Kemp's desk.
Recount Deadline and Discrepancies
It is essential to point out that the deadline for the machine recount was Wednesday, Dec. 2, 2020, at midnight.
Following an Open Records Request for "email correspondence" between Fulton County Election Director Rick Barron and Ryan Macias of The Elections Group, the complainant received an email on Dec. 3, 2020, at 12:13 PM from Mr. Barron. The email, which included no subject, had an attachment named "Batches Loaded Report.xml" that contained two separate PDF files. Immediately upon viewing the files, it was determined that the Batches Loaded Report was 'the Fulton County's reported results for the Machine Recount."
Notably, a Batches Loaded Report is generated from the Dominion Election Management Server. The report shows the batches loaded into the system, along with the status of each one. In other words, the report contains official election results at the batch level. For context, the Batches Loaded Report for the Nov. 3, 2020 count exactly matched the official results. The report notes, "The votes for the Presidential contest totaled only 511,543, which is 16,382 less than that which was officially certified."
A closer look by Moncla and Rossi at the internal timestamp on the raw data of the email and attached report revealed that the document itself (Batches Loaded Report) contains the timestamp of Thursday, Dec. 3, 2020, at 12:06:29. The data confirmed the email timestamp to be Dec. 3, 2020, at 12:13:02 PM, which indicates the email was sent approximately 6 minutes after the Batches Loaded Report was saved. Moncla and Rossi note, "These separately sourced timestamps are reasonably synchronized, thereby removing the possibility that there was incorrect time set on any of the servers or workstations involved."
Still, as previously stated, the deadline for the machine recount was midnight on Dec. 2, 2020, which was nearly 12 hours before the email sent by Rick Barron. Moreover, a tweet posted at 11:52 PM on Dec. 2, 2020—8 minutes before the official deadline—declared, "Fulton County has completed the recount of the Nov. 3 Presidential Election. Results will be released by the Secretary of State's Office." Moncla and Rossi point out:
"Simply put, the recount results that the Fulton County Elections Director emailed to Ryan Macias a few minutes after noon on Thursday, Dec. 3, 2020 (the day after the recount deadline), showed only 511,543 votes. Significantly less than the official results of 527,925."
Fulton County Board of Registration and Elections meeting minutes during which the Recount results were officially certified.
What Caused the Ballot Count Difference?
To figure out what caused the difference between the Batches Loaded Report and the certified election results, the complainant submitted an open records request to Fulton County, explicitly asking for the Recount Batches Loaded Report. In response, Fulton County returned the Batches Loaded Report seen in the screenshot below, highlighting the report's relevant portion.
Screenshot of Batches Loaded Report presented by Fulton County
In looking at the two Batches Loaded Reports (the one from Rick Barron's Dec. 3 email, which is referred to as BLR 1; and the one referenced immediately above, which is referred to as BLR 2), it is clear that BLR 2's total includes 16,198 more votes than BLR 1.
The Fulton County Board of Registration and Elections (BRE) held a meeting on Dec. 4, 2020, where Mr. Barron provided a briefing on the recount results. Barron contends the results were uploaded on Wednesday, Nov. 3, at which time they discovered they were "short." On Thursday morning, the Secretary of State advised Fulton County to "reconcile" the results. That afternoon, they determined that a scanner had two sets of 62 batches that were named the same. Unsure of why this happened, the situation caused them to re-scan the 62 batches of ballots on different scanners. Assessing Barron's summary of the recount, Moncla and Rossi stress:
"There are several problems with what Mr. Barron's claims necessitated the "reconciliation" and the actions he said were taken.
First, Mr. Barron claims there was no way to know the recount results before they were uploaded to the Secretary of State. A point which [Mr. Wingate] was right to question as such a process defies logic. The same also defies fact as the Dominion Voting System manual defines several steps of verification before results can be submitted or "published."
Second, Mr. Barron attributes the cause for the results being "short" as two sets of 62 batches which were scanned on the same machine and both numbered the same, thereby requiring one set to be re-scanned."
Elaborating further on his explanation of why the results were "short," Barron stated:
"Once that occurs [two sets of batches scanned on the same machine], there are only 2 ways you can do it, either you have to append those batches at the end, or you can scan them on different scanners. So we used different scanners rather than appending them on the end."
Nonetheless, as pointed out by Moncla and Rossi, the Dominion manual provides instructions for the very situation described by Mr. Barron, as referenced below.
According to Barron's statement to the BRE, he chose to re-scan the outstanding batches instead of appending them to the end. However, tabulator records dispute that claim. Records show that multiple additional batches were added along with the 62 batches added during the reconciliation. In fact, these batches were "appended" to the end of not one but seven tabulators "in an obvious attempt to spread the 16,000+ votes amongst many machines."
Moncla and Rossi report that on the evening of Dec. 3, 2020, Barron added seven batches (see below) to the results of several tabulators. As an example, the batches added to the end of Tabulator 793 are also highlighted below.
Fulton County Board was Unaware of Size of Recount Discrepancy
A conversation towards the end of the Dec. 4, 2020 BRE meeting with board member Mr. Mark Wingate reveals that the Fulton County Board "was completely unaware" of the size of the discrepancy that triggered the reconciliation:
(59:14) Mr. Wingate:
"We have counted 3 different times and you have gotten 3 different results. When you did the hand recount you were at 742 less than what we certified on 11/13 and now we are recertifying 852 less than what was originally certified on 11/13. It's hard for people to understand buckets and why they become unreconcilable and they why you have to go back, like this time, to try and come up with some reconciliation because if you wouldn't have done that you would have been what, some 3 or 4 thousand votes even less?"
(1:00:16) Mr. Barron:
Mr. Barron disclosed during the meeting that they realized they were "short" after the recount was uploaded to the SoS, prompting the SoS to order Fulton County to conduct a "reconciliation." Barron then offers several excuses for why the submitted results were deficient and more excuses about his actions to remedy the deficiencies. Yet, Moncla and Rossi explain that every excuse is refuted in detail by the facts and record at hand. They proclaim:
"Votes were not added for the reasons Barron explained, derived from the source Barron described, nor distributed the way Barron stated.
No matter the reason, 16,198 votes were in fact added to the recount between the time of Rick Barron's email to Ryan Macias (not an employee or contractor of Fulton County) at noon on Dec. 3 and the time of the BRE meeting the following morning."
Missing Ballot Images: What Wasn't Included in the Recount
Moncla and Rossi uncovered through analysis by "WarTornTrump" that the official recount included 17,852 votes with missing ballot images. Recent filings in a pending lawsuit verify that Fulton County has confirmed it does not have the disappeared ballot images. There are also 3,125 duplicate ballot images included in the official recount results, as recorded by Seeforyourself. Together, these two additions total 20,977 ballots for which there are votes but no ballot images. Moncla and Rossi reiterate:
"There are 20,977 unsubstantiated votes, which were in fact counted and included in the official results for the recount."
Remarkably, Moncla and Rossi report that "original election-night [Nov. 3] count included 10 Advance Voting tabulator results for which we have no record." Specifically, this means "no poll open tapes, no daily status tapes, and poll closing tapes." An Open Records Request to Fulton County seeking the ten tabulator tapes resulted in a reply stating, "No such records."
Likewise, in an attempt to clarify if the records were missing or didn't exist, they sent two emails to Fulton County. Upon reply, they were told, "The records do not exist." The ten tabulator results with no backup total 20,713 ballots. Undoubtedly, the realization Fulton County has no record for the tabulators and no record of the 20,713 ballot images from the allegedly scanned ballots is significant.
Summarizing the unsettling circumstances in their detailed report sitting on Governor Kemp's desk for the past two weeks, Moncla and Rossi explain:
"It would be far more than incredible to believe the 20,713 ballot scanning/ballot image deficiency of the original count, and the 20,977 ballot image deficiency of the recount, are not one in the same."