THE ANTRIM REPORT

INVESTIGATION & SYNOPSIS

October 20, 2021

Written By:
Matt Braynard, Executive Director
Ian Camacho, Director of Research
The LAA Research Group
Table of Contents
Summary ............................................................................................................................................................... 3
Chronology of Events ........................................................................................................................................... 4
Summary of the Vote Numbers Over Time ....................................................................................................... 6
Parties to the Dispute .......................................................................................................................................... 6
The Allied Security Operations Group ............................................................................................................ 6
DePerno Law..................................................................................................................................................... 6
Michigan Secretary of State Jocelyn Benson ................................................................................................. 7
Dominion Voting Systems ............................................................................................................................... 7
Dr. J. Alex Halderman....................................................................................................................................... 7
The Allegations ..................................................................................................................................................... 8
The Tabulator Wireless Modem Quote Claim ............................................................................................... 8
The Tabulator Wireless Chip Claim ............................................................................................................... 10
The Taiwan and Germany IP Address Claim ................................................................................................ 11
The Internal Email from Dominion Claim ..................................................................................................... 12
The International Server Claim ..................................................................................................................... 12
The International Subsidiary Claim ............................................................................................................... 12
The Smartmatic Connection Claim ............................................................................................................... 13
The Wiped Ballot Images Claim .................................................................................................................... 13
The Removed Vote Adjudication Logs Claim ............................................................................................... 13
The Failed Attempt to “Zero Out” Claim ...................................................................................................... 14
The 90-Day Safe Harbor Act Violation Claim ............................................................................................... 14
The Reprogramming the Ballots Claim ......................................................................................................... 14
The Exceeding the Maximum Allowable Error Rate Claim ......................................................................... 14
The Vote Switches / Software Glitches Claim .............................................................................................. 15
The Ranked Choice Voting Claim .................................................................................................................. 15
The SQL / EAC Claim....................................................................................................................................... 15
The Security Vulnerability Claims ................................................................................................................. 16
Contractual Violations ................................................................................................................................... 17
Decertify Dominion Because of Inaccuracies............................................................................................... 19
Final Assessment of the ASOG Report & Antrim Events ................................................................................. 22
An Open Source Solution ................................................................................................................................... 23
Acknowledgements ............................................................................................................................................ 24
Get Involved! ...................................................................................................................................................... 24

The Antrim Report Page 2 of 24

Look Ahead America
Summary
On the night of the 2020 General Election, Antrim County went to Joe Biden. This drew
immediate scrutiny given that then-candidate Donald Trump had won handily four years
earlier. The following day, a corrected count showed that Dominion machines had erroneously
underreported Trump and that he had won Antrim County by nearly 6,000 votes.

The Allied Security Operations Group (ASOG), on behalf of DePerno Law, performed an
independent investigation in Antrim County.

ASOG claimed that Dominion Voting Systems (DVS) machines in Antrim County intentionally
switched votes from Trump to Biden, transmitted the results wirelessly, had an exceptionally
high error rate that exceeded certification limits, and illegally wiped the ballot images in
violation of state law. Democratic Michigan Secretary of State Jocelyn Benson, the bipartisan
Michigan Senate Oversight Committee, Dominion Voting Systems, and University of Michigan
Professor of Computer Science and Engineering Dr. J. Alex Halderman – who performed an
independent analysis – concluded that ASOG’s claims had no factual basis for several reasons.

Our analysis of all reports and data showed that, for the most part, the defendants correctly
rebutted ASOG’s claims. No evidence suggests that Antrim County’s DVS machines had
fraudulent features in their designs, were online, or suffered from wiped ballot images.

Vulnerabilities in the system did exist, however, although mostly due to human error and
negligence. Due to missing poll tapes, one local race could not correctly audit or accurately
account for 72 ballots. In another race, a 3-ballot discrepancy changed the outcome of the race,
with machine error as a conspicuous suspect. That race also had prior scan records erased
without explanation. These instances alone should decertify DVS machines.

Furthermore, our research uncovered evidence that Dominion violated its state contract
because the tabulators and memory cards failed to store write-in ballot images as specified in
contract point 1.1.A.23, they violated the state agreement. These points along with the
tabulation errors provide sufficient cause to decertify the DVS machines.

As nobody could view the source code due to DVS products being proprietary, “black box
voting” machines, nobody outside Dominion knows for sure what it contains. All of these issues
could be resolved through the adoption of open source voting machines, with requirements to
archive ballot images and ballot definitions.

The Antrim Report Page 3 of 24

Look Ahead America
Chronology of Events
January 2017 – Former Michigan Secretary of State (SOS) Ruth Johnson, a Republican,
announces the purchase of Dominion Voting Systems (DVS), Election Software & Services
(ES&S), and HartInterCivic for state elections.1

March 1, 2017 – Dominion Voting Systems signs a 10-year contract with the Michigan
Department of Technology, Management, and Budget.2

January 19, 2019 – The U.S. Election Commission certifies Dominion Voting Systems’ Democracy
Suite 5.5-A (modification) election management system.3

November 5, 2019 – Antrim County, Michigan has its first two elections using DVS’ Democracy
Suite in Warner Township and for Boyne Falls Public Schools.4

August 25, 2020 – DVS employee Cheryl Homes sends an email summarizing the efficacy of the
election process following the primaries. The email describes issues related to the transmission
of vote totals via wireless modems, and it states that DVS turned off image saving.

November 4, 2020 – Unofficial results posted by the Antrim County Clerk show that Joe Biden
got over 7,700 votes, or 3,000 more than Donald Trump. Antrim County voted 62% in favor of
President Trump in 2016.

November 5, 2020 – Media outlets publish the county’s first attempt to rectify the initial vote
discrepancy, which shows Trump with 9,783 votes and Biden with 7,289 votes.

November 6, 2020 – The Allied Security Operations Group (ASOG), on behalf of DePerno Law,
on behalf of client Bill Bailey, begins its independent investigation after the re-tabulation.

November 7, 2020 – Secretary of State Jocelyn Benson, a Democrat, releases a statement:

“After discovering the error in reporting the unofficial results, the clerk worked diligently to
report correct unofficial results by reviewing the printed totals tape on each tabulator and
hand-entering the results for each race, for each precinct in the county.” 5

November 16, 2020 – Michigan state’s Official Vote report releases.

November 21, 2020 – Per the ASOG report, someone unsuccessfully attempts to zero out the
election results on the same day as the canvass/certification. The bipartisan Michigan Senate
Oversight Committee (MSOC) also begins its own investigation on this day.

1
https://www.michigan.gov/documents/sos/SecJohnsonAnnounce_549600_7.pdf
2
https://www.michigan.gov/documents/localgov/7700117_555468_7.pdf
3
https://www.eac.gov/voting-equipment/democracy-suite-55-modification
4
https://www.michigan.gov/documents/sos/Aug_6_2019_Elec_Jurisdictions_657811_7.pdf
5
https://www.michigan.gov/documents/sos/Antrim_Fact_Check_707197_7.pdf
The Antrim Report Page 4 of 24
Look Ahead America
December 6, 2020 – Forensics acquisition of Antrim County election equipment.

December 13, 2020 – Russell Ramsland, Jr. publishes the ASOG Report findings and allegations. 6

December 14, 2020 – Dominion issues a public relations piece regarding various election claims
titled “Setting the Record Straight: Facts & Rumors.”7

December 15, 2020 – Dominion Voting Systems CEO John Poulos goes before Michigan’s Senate
Oversight Committee and gives testimony under oath.

December 17, 2020 – The State Bureau of Elections conducts the Antrim County recount of the
presidential race only. The final count shows that President Trump won the county.

March 26, 2021 – Professor of Computer Science and Engineering at the University of Michigan,
Dr. J. Alex Halderman offers his analysis on behalf of the Michigan SOS, countering ASOG.8

April 9, 2021 – DePerno Law files a follow-up claim. 9

June 23, 2021 – MSOC issues a report which concludes that although the election had many
severe weaknesses, there were no significant acts of fraud.10

June 24, 2021 – Dominion issues their “Fact Check” of the ASOG report.

September 15, 2021 – Dominion revises “Setting the Record Straight: Facts & Rumors.”11

6
https://www.depernolaw.com/uploads/2/7/0/2/27029178/antrim_michigan_forensics_report_[121320]_v2_[red
acted].pdf
7
https://www.dominionvoting.com/election2020-setting-the-record-straight/
8
https://ftt-uploads.s3.amazonaws.com/wp-content/uploads/2021/03/29140718/Antrim.pdf
9
https://www.depernolaw.com/uploads/2/7/0/2/27029178/ex_5-10.pdf
10
https://misenategopcdn.s3.us-east-
1.amazonaws.com/99/doccuments/20210623/SMPO_2020ElectionReport_2.pdf
11
https://www.dominionvoting.com/election2020-setting-the-record-straight/
The Antrim Report Page 5 of 24
Look Ahead America
Summary of the Vote Numbers Over Time 12

Parties to the Dispute

The Allied Security Operations Group

The Allied Security Operations Group (ASOG) alleged that Dominion Voting Systems (DVS)
intentionally and purposefully had inherent errors built into its design schematics in order to
create systemic fraud and influence election results. ASOG claimed that the system
intentionally generated an enormously high number of ballot errors in the electronic ballots in
order to transfer them to adjudication with no oversight, transparency, or audit trail. ASOG
claimed that this process led to election fraud as a result of machine and/or software error, not
human error. The ASOG Report concluded that because 48 other counties in Michigan used
these same machines and software, their alleged finds cast doubt on Michigan’s elections.

DePerno Law
DePerno Law claimed that the machines used in Antrim County used wireless modems and
transmitted the results online and overseas in violation of state and federal laws. The firm
primarily relied on The ASOG Report by Russell Ramsland, Jr., who headed the group. Ramsland
claimed that an unauthorized user attempted to zero out election results on November 21,
2020, which demonstrated intent to deliberately remove the log files. He also cited a DVS
training manual mentioning the digital adjudication feature, which they claimed proved that
the Antrim machines had digital adjudication that allowed county workers to change votes.
DePerno Law also cited an internal email from DVS employees following the 2020 primary
election discussing wireless transmission and turning off the ballot image feature as evidence.

12
https://misenategopcdn.s3.us-east-
1.amazonaws.com/99/doccuments/20210623/SMPO_2020ElectionReport_2.pdf
The Antrim Report Page 6 of 24
Look Ahead America
Michigan Secretary of State Jocelyn Benson

Michigan Secretary of State Jocelyn Benson first explained the difference between the voting
machines and the election management systems (EMS): “Antrim County uses the Dominion
Voting Systems election management system and voting machines (tabulators), which count
hand-marked paper ballots. Counties use election management systems to program tabulators
and also to report unofficial election results.” She explained that the anomalous tabulation
event was due to “human error” and isolated to Antrim County, and did not affect other states,
counties, or overall election results: “When the software was reprogrammed, the County also
had to update the software on all of the media drives that are placed in tabulators to ensure
tabulators communicate properly with the election management system [….] However, because
the county did not update the media drives for the tabulators that did not have changes to
races, those tabulators did not communicate properly with the County’s central election
management system software when the county combined and reported unofficial results.”13

Dominion Voting Systems

Dominion Voting Systems responded with a series of “Fact Checks” meant to debunk the ASOG
Report in its entirety on June 24, 2021.14 Often arguing from authority and using circular
reasoning (such as citing the statement from Michigan’s Secretary of State, which cited the DVS
site as a debunking source, which in turn cited her statement), it disputed ASOG’s claims as
unfounded and baseless. It largely relied on Dr. J. Alex Halderman’s analysis for its response.

Dr. J. Alex Halderman

University of Michigan Professor of Computer Science and Engineering Dr. J. Alex Halderman
provided the most objective review. Dr. Halderman concluded that human errors resulted in
inaccurate vote totals in his investigation. He explained that the corrected ballots did not affect
the presidential candidate’s ballot position, but that changes affected everything downstream;
everything had a 1+ in terms of assignment, so that votes cast for Trump went to Biden.

13
https://www.michigan.gov/documents/sos/Antrim_Fact_Check_707197_7.pdf
14
https://www.dominionvoting.com/election-2020-setting-the-record-straight-antrim-county-michigan/
The Antrim Report Page 7 of 24
Look Ahead America
The Allegations
Many allegations and rebuttals regarding the conduct and accuracy of the Antrim elections
came out of these events. We review each of them individually below.

The Tabulator Wireless Modem Quote Claim

In its April 9, 2021 filing, DePerno Law persisted in its claims that the machines used in Antrim
County used wireless modems. “Two versions of Michigan voting systems both Dominion and
ESS [sic] have been found to have utilized wireless technology. The Dominion Voting Systems
proposal for Antrim County shows a quote for wireless transmission capabilities.” 15 The
Michigan Senate Oversight Committee stated, “The cited proof of [wireless] modems is from a
quote for purchasing received by the county from Dominion, not an actual purchase receipt or
physical sighting of any [wireless] modems.”16

15
https://www.michigan.gov/documents/sos/Dominion_quotes_554556_7.pdf
16
https://misenategopcdn.s3.us-east-
1.amazonaws.com/99/doccuments/20210623/SMPO_2020ElectionReport_2.pdf
The Antrim Report Page 8 of 24
Look Ahead America
According to Halderman, the Democracy Suite had ImageCast X (ICX) ballot-marking devices
(BMDs), which had wireless capabilities:

Antrim County adopted the Democracy Suite system produced by

Dominion Voting Systems, consisting of ImageCast Precinct (ICP) ballot
scanners, ImageCast X (ICX) ballot-marking devices (BMDs), and the
Democracy Suite election management system (EMS). As deployed in
Antrim, the EMS consists of a single desktop PC running specialized
software for election preparation and results aggregation and reporting.
Antrim uses Democracy Suite version 5.5, which is federally certified by
the U.S. Election Assistance Commission.17

DePerno Law relied on cybersecurity expert James Penrose's report for evidence of wireless
transmission. His report showed a Dominion ICX schematic, stating that it had the ability to use
an external wireless modem and wired connectivity.

The ICX machine may also utilize an external wireless for

communications modem with the central listener server for Dominion
Democracy Suite. See the previously listed proposal from Dominion to
Antrim County. The manual for the ICX also shows an Ethernet port for
wired connectivity [emphasis added].18

Penrose's evidence referred to a proposal and not a purchase order, however, which even his
own report acknowledged. Furthermore, the proposal document cited had no signature and
therefore no authorization, as seen in the blank, unsigned line above "Authorizing Signature,
Title."19 Some Michigan counties had signed and authorized proposals (with different details
and criteria), which, when compared to Antrim County's unsigned, unauthorized proposal,
indicated that not all counties left theirs blank and so this was not a uniform collection of
county proposals all without signatures.20 Furthermore, Michigan only uses ICXs as BMDs; in
that configuration they do not record votes at all, so they never have any results to transmit.

This assumption that an unsigned, unauthorized proposal equaled a purchase order explains
why Ben Cotton, founder of CyFIR, LLC, wrote that he had seen purchase documents:

I examined the purchase documents produced by Antrim County with

respect to the purchase of the Dominion Voting system and note that the
following system components listed on the purchase documents were
not produced:

(a) ImageCast Listener Express Server

(b) ImageCast Express Firewall

17
https://ftt-uploads.s3.amazonaws.com/wp-content/uploads/2021/03/29140718/Antrim.pdf
18
https://www.depernolaw.com/uploads/2/7/0/2/27029178/ex_5-10.pdf (Exhibit 6, Figure 4)
19
https://www.depernolaw.com/uploads/2/7/0/2/27029178/ex_5-10.pdf (Exhibit 6, Figure 1)
20
https://www.michigan.gov/documents/sos/Dominion_quotes_554556_7.pdf
The Antrim Report Page 9 of 24
Look Ahead America
 (c) EMS Express Managed Switch
(d) ICP Wireless Modems (17)
(e) Image Cast Communications Manager Server
(f) ImageCast Listener Express RAS (remote access server) System
(g) ImageCast USB Modems (5)21

Again, these items appeared on the unauthorized, unsigned proposal under "Results
Transmission" but not on any purchase documents provided by DePerno Law. This fact would
also confirm Halderman's analysis and other report's findings.

One should ask why if DePerno Law had a purchase order that showed Antrim's purchase of all
these items, then why did the firm not include this key document in their evidence instead of
relying on an unsigned purchase proposal as its proof? Why have neither Antrim County
workers, Dominion workers, voters, nor poll watchers come forward to report wired or
external, wireless transmission equipment on any machines in Antrim County? Even though the
ICX had wireless capability, it does not mean that anything went online. While it could pose a
potential vulnerability, it does not in and of itself prove that any exploits occurred.

The Tabulator Wireless Chip Claim

The ASOG Report introduced an image in evidence of a wireless chip, suggesting fraud in
Michigan.22 It came from ES&S, however, a company not used in Antrim County. The MSOC
Report countered in a footnote that Antrim County DVS machines had no modems in them:
“ES&S and Hart InterCivic tabulators have internal modems, but not Dominion. However, they
are not turned on until the polls are closed and tabulation has concluded.”

According to the contract between DVS and Michigan, however, these machines did contain
internal dial-up modems: “ImageCast Precinct: The ImageCast Precinct tabulators are equipped
with internal dial-up modems, and can also transmit results via an external cellular modem.” 23
Bois Blanc Township confirmed that Antrim machines had internal modems: “While the State of
Michigan requires all tabulation systems to be capable of transmitting election results from the
precincts via modems, Antrim County does not use modems.”24 Dr. J. Alex Halderman also
confirmed that Antrim lacked external wireless, cellular modems: “Antrim County did not
purchase and does not use the Dominion wireless results transmission functionality.” NBC News
had confirmed that Dominion put them in tabulators and scanners: “The three largest voting
manufacturing companies — Election Systems & Software, Dominion Voting Systems and Hart
InterCivic — have acknowledged they all put modems in some of their tabulators and scanners.
The reason? So that unofficial election results can more quickly be relayed to the public.” 25 In

21
https://www.depernolaw.com/uploads/2/7/0/2/27029178/ex_5-10.pdf (Exhibit 8)
22
https://www.depernolaw.com/uploads/2/7/0/2/27029178/ex_5-10.pdf (Exhibit 6, Figure 5)
23
https://www.michigan.gov/documents/localgov/7700117_555468_7.pdf (p. 112)
24
http://www.boisblanctownship.org/downloads/to_satisfy_the_numerous_questions_about_the_dominion_tabu
lators_used_in_our_township_voting_process.pdf
25
https://www.nbcnews.com/politics/elections/online-vulnerable-experts-find-nearly-three-dozen-u-s-voting-
n1112436
The Antrim Report Page 10 of 24
Look Ahead America
other words, the machines could transmit online with their internal dial-up modems as
required by law, but did not use them.

Although ASOG erred on the external modem claim, the MSOC and DVS also misled by claiming
these machines had no wireless modems. In actuality, they had no external, wireless modems,
and they had internal, dial-up modems.

The Taiwan and Germany IP Address Claim

DePerno cited a report by Penrose which stated he found two IPs, one in Taiwan and the other
in Germany. “The first IP address was: 120.125.201.101. This IP address is registered to Ministry
of Education Computer Center located in Taipei, Taiwan. The second IP address was:
62.146.7.95. This IP address is registered to EDV-BV GmbH QSC Subkunde located in Nurenberg,
Germany.”26 Additionally, DePerno Law also cited Ben Cotton, founder of CyFIR, LLC, who
provided an image of IP addresses located in an unallocated space. Indeed, the image shows a
loopback address (127.0.0.1) and default gateway (192.168.0.1). 27 It also shows the address
120.125.201.101 that Penrose cited, but not 62.146.7.79 and 62.146.210.52, nor 62.146.7.95
that Penrose cited, nor the private File Transfer Protocol network that Cotton mentioned.

Even assuming that this IP information proves true, Cotton's report also noted that “the
unallocated nature of the artifact precludes the exact definition of the date and time that this
data was created.” Without time lines on IP connections, one cannot confirm that they were
connected on Election Day. One has no idea what happened with those machines from their
time of manufacture in Taiwan to their set up in Antrim County. Given that these IP addresses
come from Taiwan, the manufacturer could have connected the machines as a quality control
measure, although this remains equally as speculative as DePerno’s claims.

Halderman confirmed that Cotton’s screenshot shows a fragment of a real piece of malware,
but clearly a Windows executable (it references to ntdll.dll, for instance). He noted that seems
strange, since he states that he found it in unallocated space on the ICX, which runs Android.
Halderman also confirmed that a stock Windows installation from around the time Michigan
purchased the equipment contains a very similar fragment of the same malware, but it's part of
the Windows Defender antivirus signature database (used for identifying malware), and not any
kind of actual infection. Apparently the ICX originally had Windows loaded – perhaps at the
factory for testing – then Dominion installed Android without completely erasing the storage.
That would explain how these addresses ended up in unallocated space.28 In any case, these IP
addresses seem to indicate one of the reasons why Michigan demanded that the machines be
built in the USA in its contract.

26
https://www.depernolaw.com/uploads/2/7/0/2/27029178/ex_5-10.pdf
27
https://www.depernolaw.com/uploads/2/7/0/2/27029178/ex_5-10.pdf (Exhibit 8, Figure 1)
28
Email with Dr. J. Alex Halderman, 19 October 2021
The Antrim Report Page 11 of 24
Look Ahead America
The Internal Email from Dominion Claim

DePerno law cited an email from DVS employees as evidence that the machines went online
and wiped ballot images. 29 It stated “Dominion representatives also confirmed issues with
wireless transmission of vote totals and even went as far as disabling the saving of ballot
images without explicit authorization.” The MSOC responded, “The context in the email, to
make electronically transmitting the results after the election with the attachable modem
function better, makes the instruction to turn off transmitting the image a reasonable
instruction when coupled with there being no law in Michigan to save the images.” Further
examination of this email reveals that it appeared to be internal, and unrelated to Antrim
County. Second, clearly the email discussed 3G, wireless signals, and thus the attachable
modem referred to an external, wireless modem, which Antrim County did not have. Finally, it
came from an election before the 2020 General Election and would not have been relevant
anyway, even if what ASOG had claimed was true.

The International Server Claim

Having shown that these machines in Antrim were not online, the claims that Dominion servers
were owned and operated from Canada, Spain, Serbia, and Germany would not have affected
Antrim either way. Even so we reviewed the claim. MSOC explained international connections:

Mr. John Polous, Dominion CEO denied the existence of Dominion servers in
Spain and Germany, emphasizing that ballots remain local, are counted locally,
and are not moved over state lines, let alone overseas […] County clerks then
report these unofficial results both locally and to the secretary of state. The
secretary of state releases the unofficial results to media and their own page.
Clarity, a Spanish based company, also takes in these unofficial results from the
county or the state. This company, which is based in Spain and has servers in
Europe, makes the unofficial results available to multiple users, especially media
subscribers who utilize the unofficial results in their election night
prognostications. Scytl and others are companies that provide similar services.
All of these activities, especially due to media inquiries, constitute a significant
explanation for much of the cyber activity across the country and the globe on
election night.30

Indeed, the overseas servers did not belong to Dominion but rather to Scytl. One must prove
that Dominion owned Scytl or held a majority share in it, and ASOG did not.

The International Subsidiary Claim

The report explained that DVS is a Canadian owned company with global subsidiaries, owned by
Staple Street Capital, a New York hedge fund, which is in turn owned by UBS Securities LLC, of

29
https://www.depernolaw.com/uploads/2/7/0/2/27029178/ex_5-10.pdf (Exhibit 6, Figure 4)
30
https://misenategopcdn.s3.us-east-
1.amazonaws.com/99/doccuments/20210623/SMPO_2020ElectionReport_2.pdf (p. 20-22)
The Antrim Report Page 12 of 24
Look Ahead America
which 3 of their 7 board members are Chinese nationals. Dominion refuted this as impossible
with the oversights and full disclosure, with which they must comply. Although Dominion’s
main headquarters are in Toronto, Canada, it incorporated in Denver, Colorado, making it legal.
Furthermore, the report wrongly attributed US-based UBS Securities LLC’s purchase of Staple
Street Capital in October 2020 to the Chinese subsidiary, UBS Securities Co., Ltd. Additionally,
nothing suggested that UBS Securities LLC would remain the eventual buyer. The report showed
no direct ties between Dominion and the Chinese Communist Party, and could not connect the
dots with any certainty. Instead it relied on speculation, which has no legal weight.

The Smartmatic Connection Claim

ASOG claimed that Dominion either bought or leased (depending on its argument) Smartmatic
software in order to connect it to Venezuela. It cited a former Venezuelan professor of
mathematics and alumni of University of Michigan Gustavo Delfino who claimed that software
in the Smartmatic electronic voting machines got incorporated into Dominion Voting Systems
during the 2004 Venezuela national referendum. Dominion responded that they are two
separate companies that make electronic voting systems, and do not use or license Smartmatic
software and/or its software from Sequoia. Smartmatic also refuted such claims. Indeed, no
evidence presented has shown a connection between these companies, or that the Michigan
voting system software used the Venezuela voting system software.

The Wiped Ballot Images Claim

DePerno Law insisted that much of the purported fraud occurred due to a high rate of digital
adjudication of votes in Antrim County. Dominion claimed that Antrim County opted out of the
adjudication service. Dr. Halderman confirmed this fact and so the claim would not apply:
“Dominion’s adjudication system stores the ballot image together with the scanner’s
interpretation of the votes and a log of any changes made by the system operator. Adjudication
is an optional feature of the Dominion system and was not used in Antrim County.” Antrim
County could not delete or disable any ballot images, and the media drives failed to load ballot
images, precisely because the county chose not to purchase this feature. DePerno’s citing a DVS
training manual only applied to machines with the feature, therefore, was irrelevant. While
Halderman noted that a ballot image saving option should be mandatory for future elections,
DePerno’s evidence showed that ballot images were unavailable, not that deletions occurred.

The Removed Vote Adjudication Logs Claim

The report stated that the computer system showed vote adjudication logs for prior years but
missing adjudication log entries for the 2020 election cycle. The report’s authors believed that
this showed evidence that manual removal of the 2020 election cycle records occurred. Due to
purportedly missing files and adjudication logs, it had not yet determined where bulk
adjudication occurred or who was responsible.

The Antrim Report Page 13 of 24

Look Ahead America
The Failed Attempt to “Zero Out” Claim

Russell Ramsland, Jr. also claimed that an unauthorized user attempted to zero out election
results on November 21, 2020, which demonstrated intent to deliberately remove the log files.
Halderman showed that Ramsland did not understand how programming code functions:

The only evidence he offers for this assertion is an EMS log file entry that reads
“EmsLogger - There is no permission to {0}”, which he claims “is direct proof of
an attempt to tamper with evidence”. This is absurd and misleading. A
programmer would immediately recognize that {0} is merely a placeholder, in
this case one that was intended to be replaced with a description of the
attempted action. It has nothing to do with “zeroing” election results.

In many programming languages, anything indexed begins with the index 0 instead of 1,
contrary to what non-programmers might expect; non-programmers expect the first value to
always be “value #1” not “value #0”. In any case, this evidence did not indicate voter fraud.

The 90-Day Safe Harbor Act Violation Claim

The ASOG Report had claimed that the Antrim County election workers violated the “Help
America Vote Act, Safe Harbor” (HAVA) 90-day period. MSOC explained that no such definition
exists in HAVA. Indeed, no such statement exists in this document.31 No such statement appears
in any Michigan state law or document either. Therefore, it is a baseless claim.

The Reprogramming the Ballots Claim

The ASOG Report argued that new ballot programming occurred on Dominion ImageCast
Precinct Cards both on 10/23/2020 and on 11/05/2020. These system changes affected how
the machines read and tabulated ballots. The report concluded that the only reason to change
the software after the election was to obfuscate evidence of fraud and/or to correct program
errors that would decertify the election. Halderman demonstrated that the October changes
resulted from the local election update, and the change in November from this correction.

The Exceeding the Maximum Allowable Error Rate Claim

ASOG forensic researchers allegedly observed an error rate of 68.05% in Antrim County using
the tabulation log. They asserted that it recorded 10,667 of 15,676 individual events (68.05%)
as errors. They added that Central Lake Township purportedly had 1,222 ballots reversed out of
1,491 total ballots cast (an 81.96% rejection rate) that led to the sending of these ballots to
adjudication for a decision by election personnel without oversight. Halderman explained that
the “error rate” resulted from standard operating system error messages for basic computerized
machine operations. For example: error logs generated during start up or shut down sequences,
or issues such as not properly inserting a ballot. These do not represent large batches of human

31
https://www.eac.gov/sites/default/files/eac_assets/1/6/HAVA41.PDF
The Antrim Report Page 14 of 24
Look Ahead America
adjudicated ballots. While Halderman showed that the ASOG Report failed to produce the
claimed 68.05% error rate, he neglected to mention that the report misstated its source claim
in three ways: (1) the Federal Election Commission never cited an allowable 1 in 250,000 ballots
error rate, (2) 1 in 250,000 ballots would equal 0.0004% not 0.0008%, (3) the report later cited
its source as the National Conference of State Legislatures’ (NCSL) error rate of 1 in 125,000
(0.0008%), correcting its earlier misstatement.32 These failures to substantiate mathematical
claims, coupled with an incorrect citation and calculation, undermined the report’s credibility.

The Vote Switches / Software Glitches Claim

Dominion argued that no votes were “switched” by “software glitches” in Antrim County, but
rather by user-error. The Michigan SOS stated this occurred as well. Our research showed that
the issue stemmed from ballot issues, inadequate staff training, and a failure to update
software. The ASOG Report stated that “the vote flip occurred because of machine error built
into the voting software designed to create error.” Yet, Dominion’s source code remains
inaccessible and thus ASOG cannot make such a determination.

The Ranked Choice Voting Claim

The ASOG Report claimed that the ranked choice voting feature present in DVS allowed a bad
actor to assign fractional vote counts to specified candidates. Though the report claimed “In the
logs we identified that the RCV or Ranked Choice Voting Algorithm was enabled” it only showed
a Dominion training manual referencing RCV. DVS’ “fact check” and CEO John Poulos countered
under oath before MSOC that Michigan does not use RCV, nor support fractional or "weighted"
voting. Although Antrim did not have RCV, Eastpointe, Michigan does use RCV.33 34 While RCV
remains illegal in US presidential elections, it has legal use in some state and local elections.
Dominion markets itself to any state or municipal organization willing to purchase their EMS.
Therefore, RCV features in DVS would reflect laws established in some areas. However, this does
not support any claim that Antrim County machines had or used this feature.

The SQL / EAC Claim

ASOG claimed that Dominion systems had Microsoft (MS) Structured Query Language (SQL)
software. It referenced MS SQL as a tool that used but provided no additional reference as to
whether Dominion installed the program or if the program came installed already. Halderman
stated the system already had MS SQL Server Management Studio installed with no need for
additional passwords to view and edit election databases. DePerno later cited Cyber Ninjas’
finds that they identified an installation, and that it was not a component of Election Assistance

32
https://www.eac.gov/sites/default/files/eac_assets/1/28/VVSG.1.1.VOL.1.FINAL1.pdf
33
https://www.justice.gov/opa/pr/justice-department-reaches-agreement-city-eastpointe-michigan-under-voting-
rights-act
34
https://www.fairvote.org/_eastpointe_michigan_to_become_first_in_state_to_implement_ranked_choice_votin
g
The Antrim Report Page 15 of 24
Look Ahead America
Commission (EAC) Dominion Certification Configuration, which meant that the actual voting
system used was not in compliance with its certification. 35


While the EAC serves to ensure HAVA compliance, however, HAVA specifically says that EAC
certification criteria and guidelines serve as exactly that: guidelines. In fact, the certification
states upfront “The voting system identified on this certificate has been evaluated at an
accredited voting system testing laboratory for conformance to the Voluntary Voting System
Guidelines Version 1.0 (VVSG 1.0). […] This certificate is not an endorsement of the product by
any agency of the U.S. Government and no warranty of the product is either expressed or
implied.”36 It acts as a set of voluntary voting guidelines. The certification also states:

An EAC certification is an official recognition that a voting system (in a

specific configuration or configurations) has been tested to and has met
an identified set of Federal voting system standards.

An EAC certification is not:

• An endorsement of a Manufacturer, voting system, or any of the
system’s components.
• A Federal warranty of the voting system or any of its components.
• A determination that a voting system, when fielded, will be operated in
a manner that meets all HAVA requirements.
• A substitute for State or local certification and testing.
• A determination that the system is ready for use in an election.
• A determination that any particular component of a certified system is
itself certified for use outside the certified configuration.

The EAC guidelines act as a best practice rather than as a requirement. Unless a specific
Michigan state law was violated in relation to the EAC, no legal enforcement exists in terms of
EAC guidance. In good faith, Michigan should decertify, but, legally speaking, the state has no
grounds for doing so when going by EAC/HAVA “guidelines”. While MS SQL poses problems due
to malicious code possibilities and many hacker tools pertaining to SQL, citing noncompliance
with EAC certification has no legal weight. 


The Security Vulnerability Claims

The ASOG investigators discovered a number of security issues. They found an election program
thumb drive outside of a security vault specifically used to secure data memory cards and other
election-purposed thumb drives. ASOG investigators discovered election data package files on
an unsecured desktop computer, apparently a component of the EMS. In addition, the desktop
hard drive containing these data packages had no encryption, with software and security
updates/patches over two years out of date. The users operating this computer primarily
logged in with “Super User” privileges. This credential allows the user to have significant access

35
https://www.depernolaw.com/uploads/2/7/0/2/27029178/ex_5-10.pdf (Exhibit 7)
36
https://www.eac.gov/sites/default/files/voting_system/files/DSuite55_CertConf_Scope%28FINAL%29.pdf
The Antrim Report Page 16 of 24
Look Ahead America
to all aspects of the data and operating system present on the computer. Also, multiple users
shared login credentials (likely “Super User” credentials), making full audits of election data
activity significantly more difficult. Halderman confirmed all these claims and commented on
Antrim’s lack of election security practices. He did not conclude that it led to fraud or that any
exploits had occurred, but that such negligence posed serious security vulnerabilities. He
disagreed that machine decertification should occur on these grounds, however, noting that
“missing software updates are frequently an unfortunate consequence of the federal
certification process, under which voting system vendors must obtain EAC approval for any
changes to election system software, including Windows updates.” He provided many
suggestions for the county and state to follow to ensure that these issues never recurred, and
even MSOC acknowledged these security issues. Indeed, this election was not secure at all.

Contractual Violations

A more significant reason to decertify these machines is that they violated the state contract
and state regulations, specifically item 1.1.A.23. 37

The Contractor shall describe in detail all aspects of the write-in vote and
adjudication process. The tabulator shall allow for the voter to cast a write-in
vote by marking the target area and writing the candidate name of their choice
in a provided area. The tabulator shall store an image of the write-in vote,
which can be separated out (as a group) for later determination and
adjudication of valid write-in votes. [emphasis added]

Dominion agreed to this contractual item and even made the comment:

The ImageCast Precinct allows for the voter to cast a write-in vote by marking
the target area and writing the candidate name of their choice in a provided
area. The tabulators will divert ballots with write-in votes to the write-in bin for
easy retrieval after the polls have closed. Additionally, all tabulators save the
ballot image which includes the write-in name. Ballot images can be sorted by
all conditions including write-in votes, allowing election officials to separate
write-in votes as a group for later determination and adjudication. The Results
Tally & Reporting module allows the manual entry of qualified write-in
candidates and the associated vote totals which appear on the results reports
[…] The ImageCast X allows the voter to enter a write-in name which is printed
on the Verifiable Choice Summary Ballot. The image of the write-in name is
saved by the ImageCast Precinct tabulator. [emphasis added]”

When Halderman focused on the re-zeroing issue (yellow highlighted), he showed an image
titled “Figure 3: Memory Cards Mistakenly Re-zeroed” (red square). It does not seem that the
machines stored write-in images, though ASOG and others who support the wiped image log
theory might believe this. In reality, this line would indicate something being reset like a
37
https://www.michigan.gov/documents/sos/071B7700117_Dominion_Exhibit_2_to_Sch_A_Tech_Req_555357_7.
pdf 1.1.A.23: “Voting System HARDWARE Technical Requirements - Ballot Counter / Tabulator - Write-in Votes”
The Antrim Report Page 17 of 24
Look Ahead America
partition, but it alone cannot confirm or deny whether the files were previously blank or had
previously stored data. The presence of this line likely indicates that write-in ballot images were
enabled, but does not indicate whether any images were actually stored. In sum, one cannot
determine from this data alone.

Halderman’s report confirmed that the machines complied with providing the ability to select a
write-in option, write-in a candidate, and separate write-in ballots, as their contract required:

Mr. Ramsland makes several further mistakes in interpreting the election system
logs. He states that the scanner log shows that “Divert Options” were selected
and claims that this means “all write-in ballots were sent for ‘adjudication’ by a
poll worker or election official to process the ballot based on voter ‘intent’.
Adjudication files allow a computer operator to decide to whom to award those
votes (or to trash them).” In reality, the divert option simply means that when a
voter has selected a write-in, the scanner directs the physical ballot into a
separate compartment within the ballot box. This makes it more convenient
for a worker to later read the name that was written in. All voter selections,
including the presence of a write-in, are processed normally by the scanner and
reflected on the poll tape. The setting has nothing to do with electronic
adjudication. [emphasis added]

Nevertheless, Halderman confirmed that these tabulators not only disabled the adjudication
software, but did not even install it:

[A]djudication functionality was not enabled at all in Antrim County during the
November 2020 election. The adjudication software application is an optional
component of Democracy Suite. Antrim did not purchase it, and my examination
of the EMS shows that it was not installed. There are no adjudication logs for the
simple reason that adjudication was not used.

One must ask whether image storage for regular ballots was different from that of write-in
ballots, which may or may not have required adjudication? If they were not distinct and used
the same software that was required to take and store images of write-in ballots, then the
software configuration purchased was non-compliant for certification. If these two categories
of ballots were distinct, then the choice of Antrim County not to purchase the adjudication

The Antrim Report Page 18 of 24

Look Ahead America
feature says nothing as to whether the tabulators were in compliance with contract
requirements specified in 1.1.A.23. Halderman answered this question in the next sentence:

Moreover, the tabulators were not configured to store ballot images—a

necessary precondition for adjudication—and my inspection of the memory
cards confirms that no ballot images are present. This means that it would have
been impossible to use the adjudication feature even if the software were
somehow installed after the election [emphasis added].

Additionally, Cyber Ninjas even confirmed that not only were no ballot images stored (and
therefore not wiped), but that without ballot images no write-in candidates were handled:

None of the Compact Flash drives appeared to hold ballot images, and no ballot
images had otherwise been imported into the EMS. Ballot images are a critical
artifact and are essential for any type of system audit to determine how an
electronic voting machine interpreted results and where it might be
malfunctioning. Vendor training clearly state [sic] that ballot images should be
imported into the EMS immediately following the election, but this was never
done, and the images don’t even seem to be present. Without ballot images its
[sic] near impossible to match up and see the origin of where errors might be
happening.

It is unclear how write-in candidates could have been properly handled without
ballot images available for review.38

Regarding the write-in candidates mentioned, ballots with write-in images get physically
separated by the scanner into a different compartment for manual review. Nevertheless, one
must conclude that Dominion Voting Systems violated contract item 1.1.A.23 as it failed to
include images of write-in ballots as mandated by contract and law. These analyses remain
consistent with that evidence, and this supports enough reason to decertify Dominion.

Decertify Dominion Because of Inaccuracies

Halderman’s analysis gave yet another reason for decertification:

[M]y analysis shows that both the poll tapes and the final results are inaccurate
with respect to a few down-ballot contests in the three precincts where ballot
designs changed, as a result of the county’s failure to appropriately distinguish
between ballots voted using the old and new ballot designs. The largest
discrepancy is in Central Lake Village, where the poll tape is inaccurate for the
Central Lake Schools school board contest and State Proposal 20-1. I estimate
that approximately 72 ballots are affected. Small numbers of votes may also
have been affected in Mancelona Village for the Village Trustee race and in

38
https://www.depernolaw.com/uploads/2/7/0/2/27029178/ex_5-10.pdf (Exhibit 7)
The Antrim Report Page 19 of 24
Look Ahead America
 Warner Township for the Boyne Falls Public Schools Sinking Fund Millage
Proposal. The circumstances that caused these discrepancies did not affect any
other contests and do not affect enough votes to change the outcome of any
contest.

While the presidential election did not get impacted, this discrepancy affected down ballot
issues, such as the legalization of marijuana (state proposal 20-1), and the Central Lake Schools
school board contest. Although the outcomes of both contests did not change with this
discrepancy, 72 of 16,044 ballots cast produced an error rate of 0.45%, a number significantly
higher than the allowable election error rate as established by the National Conference of State
Legislatures’ guidelines of 1 in 125,000 ballots (0.0008%). Furthermore, it directly affected the
outcome of a local race for legalizing a marijuana retailer.

In the presidential election, Halderman confirmed a higher percentage than 0.0008%. He wrote,
“In December, the state conducted a county-wide hand count of the presidential contest that
agreed with the final reported results to within 0.08%. In January, the state conducted a state-
wide risk-limiting audit of the presidential contest that affirmed the reported outcome. These
findings and my analysis all support the conclusion that there are no significant errors in Antrim
County’s final presidential results.”39

While Halderman showed why ASOG incorrectly claimed that misfed ballots equated to errors,
the professor glossed over the 13-ballot difference found in the presidential hand recount as
well as a 72-ballot difference in down-ballot contests. A variance of 13 of the 16,044 ballots cast
in the presidential contest (0.081%) and 72 of the 16,044 ballots cast in the local contests
(0.45%), exceeded the allowable ballot error rate of 1 in 125,000 cast (0.0008%). True enough,
neither discrepancy affected the overall outcome of the presidential race, nor state or local
contests. Yet Halderman mentioned that a key item for a forensic audit, the poll tapes, had
several inaccuracies in three precincts. Nevertheless, Halderman dismissed it: “Since State
Proposal 20-1 was decided by a margin of more than 3 million votes, the error could not have
affected the outcome.” By this logic, no matter how large the discrepancy found in Antrim
County's 16,044 votes, no amount of errors found in Antrim could overturn the 3 million votes
already cast by Michiganders.

Most notably, he explored a discrepancy that directly affected the outcome of a marijuana
initiative, which failed the first time when scanned, but passed on the second scan.
Additionally, he noted someone overwrote the records from the first pass to do the second.
(Why nobody made a copy he did not explain.) Halderman assigned the blame to human error,
but could not entirely avoid the possibility of it having stemmed from faulty machines:

Beyond the discrepancies noted above, the poll tapes from the two Central Lake
scans show a large number of smaller differences. Many contests have one or
two fewer votes in the second scan. One of these differences potentially affected
the outcome of a contest. As shown in Figure 6, in the first scan, the Central Lake

39
https://ftt-uploads.s3.amazonaws.com/wp-content/uploads/2021/03/29140718/Antrim.pdf
The Antrim Report Page 20 of 24
Look Ahead America
 Village Marihuana Retailer Initiative was tied (and thus defeated), but in the
second scan, which became the final result, it passed by a single vote. The data I
examined suggests that three ballots that were included in the first scan were
omitted when the ballots were scanned again, either correctly or due to human
error. The scanner log from election day, as recorded in the EMS database,
shows that 1494 ballots were scanned. Yet the log from the memory card shows
that only 1491 ballots were scanned on November 6. The memory card from the
first scan was overwritten to prepare it for the second scan, so the original digital
records of the ballots are not available. […] The data supports the conclusion
that these three ballots fully account for the residual differences between the
poll tapes from the two scans. There are multiple possibilities for why they were
not included in the second scan. While it is possible that they were ruled invalid
due to some defect and properly excluded, it is also possible that elections
staff simply did not scan them the second time, due to human error. If these
ballots are valid, it is likely that the final reported outcome of the Central Lake
Village Marihuana Retailer Initiative is incorrect and that the true result is a tie,
as shown on the election day poll tape [emphasis added].

The Antrim Report Page 21 of 24

Look Ahead America
Final Assessment of the ASOG Report & Antrim Events
The ASOG Report demonstrated many possibilities where election security could have been
compromised, but failed to prove any actual compromise or intent of compromise from
Dominion Voting Systems. It seems likely that ASOG allowed their findings in previous forensic
investigations of election irregularities to influence their conclusions in the Antrim County
General Election forensic report. Although the Michigan Senate Oversight Committee claimed
that ASOG intentionally misrepresented its findings, it seems more likely that unconscious
confirmation bias seeped into the report.
The Antrim Report Page 22 of 24
Look Ahead America
The ASOG Report exposed insufficient oversight for election system procedures in Antrim
County. By its nature, the design of Dominion Voting Systems has additional points of potential
failure in the chain of election events. While human error demonstrably had the largest
contributing factor in the reported election irregularities, seemingly unintentional design flaws
in Dominion Voting Systems (as well as the lack of publicly available documentation and
disclosure of its design) increased that potential for human error. That no parties could review
the source code made the claims difficult to prove, but likewise difficult to disprove.

Nevertheless, the ASOG Report revealed a pattern of negligence by Antrim County in the 2020
General Election. Upon examining Halderman’s and Cyber Ninjas’ evidence, our research team
confirmed that the Dominion Voting Systems machines violated contract item 1.1.A.23. At
minimum, the 72-ballot difference in a local race raised issues of reliability with inaccurate
paper trails. Halderman did not rule out machine error as a possibility for the 3-ballot
discrepancy in the local marijuana initiative, which exceeded the NSCL error rate and directly
changed the election outcome. This lack of certainty with the discrepancies appears
emblematic of DVS machines’ unreliability. For these contract violations and variances,
Michigan should decertify Dominion Voting Systems machines.

An Open Source Solution

Open source voting would have rendered these unaddressed issues irrelevant by allowing any
interested party to inspect the source code and restore transparency to the voting process. In
short, it would have removed all lingering doubts. We recommend that counties investigate our
link to more information on open source voting: https://lookaheadamerica.org/blackbox/.

The Antrim Report Page 23 of 24

Look Ahead America
Acknowledgements
The assistance of @Russ, @Scottie, @H20colorartist, @n_nix_, @Tron_Legacy, @CherylT,
@Tolkien’s Apprentice, @Missy, and Tom Lewis in the Research Group helped us to compile,
analyze, proofread, and edit this report.

Get Involved!
If you wish to help the Research Group, then please volunteer at
https://lookaheadamerica.org/volunteer/.

If you liked our work and wish to support it, then please make a tax-deductible donation at
https://lookaheadamerica.org/donate/.

The Antrim Report Page 24 of 24

Look Ahead America