Election Truth Alliance

Allegheny County (Analysis of Vote Share by Vote Count, Continued)

While not an entirely linear visual representation, mail-in voting data for Allegheny county does eventually appear to average out around the 60% vote share mark for Harris and the 40% vote share mark for Trump.

Clustering in precincts where fewer ballots were cast spans a larger range of vote share percentages than in Philadelphia, ranging from around 60–95% for Harris and 5–40% for Trump. As Allegheny encompasses both Pittsburgh as well as some smaller surrounding semi-rural communities, this broader range of vote share could conceptually be caused by geographic differences.

However, in Election Day data, there is a more pronounced shift toward a higher vote share for Trump and a lower vote share for Harris. This trend appears to become visible in precincts with around 250 votes.

The result is a very heavy upward skew towards 60% vote share for Trump and downward to 40% vote share for Harris in precincts where more votes were cast. The shift is more visible when the candidates’ charts are broken out separately.

This Election Day data does not align with expected voting patterns. Essentially, one candidate benefited, unexpectedly and disproportionately, in precincts where more votes were cast.

Notably, the ETA has documented striking similarities to the patterns seen in Pennsylvania in our analysis of Clark County, Nevada.

Erie County

Lastly, we examine vote share by vote count in Erie County.

Erie County is a much smaller community than Philadelphia or Pittsburgh, meaning there are far fewer data points. As Erie is similar in population size to San Mateo, these two sets of charts should conceptually be easier to compare with each other.

In mail-in data for Erie, a largely linear pattern is observed when represented visually, along with expected human variation in voting patterns.

The shift in pattern appears to become visible in precincts where over around 400 votes were counted.

What Could Cause This Pattern?

Hypothesis 1: Votes From Two Different Populations (But Only On Election Day)

The divergence of results shown in Philadelphia in particular is reminiscent of what might occur if the votes cast on Election Day originated from two distinct populations. The complicating factor is that these trends are so pronounced on Election Day only and not in Mail-In Voting.

What different populations could these be?

Urban and Semi-Urban? This is more likely in Allegheny, but less likely in largely urban Philadelphia where the trend is most pronounced.
Degree of Partisanship/Political Fervor? Conceptually, voters with more ardent or zealous political beliefs could have been mobilized to vote on Election Day – potentially resulting in two voting populations: a more moderate population and a more zealous population. This does not explain why the trend is limited to precincts where more votes were cast.
Different Election Machinery by Precinct? Conceptually, a non-malicious explanation could be unintended failures or complications with certain types of voting equipment that is specific to higher-volume precincts.

A next step in our analysis is intended to be plotting precincts using a tool that can provide additional information about each precinct point (i.e. location, machinery used, etc.) This will allow both our team and the broader public to explore the data more easily and freely.

Hypothesis 2: Different Voting Types = Different Voting Patterns

This is unlikely given what we see in other county voting data that reflect expected voting patterns. In an unmanipulated election, we would expect to see a different Democratic and Republican vote share on Election Day compared to Mail-In data – but represented somewhat consistently across precincts, regardless of how many votes were cast within a given precinct.

Hypothesis 3: “More Republicans Just Showed Up on Election Day” (But Disproportionately in Larger Precincts)

As mentioned in the previous section, this is unlikely a satisfactory explanation unless there is a reason why this trend would be true only in precincts with larger numbers of votes cast and not in precincts with smaller numbers of votes were cast and counted.

Hypothesis 4: Vote Manipulation Targeting Precincts With More Votes

Conceptually, if bad actors intended to manipulate the outcome of an election, precincts where more votes were cast would offer the most “bang for their buck”. Targeting more vote-rich areas and not targeting areas with fewer votes would be one option for reducing risk of being detected.

An alternative but similar scenario could be compromised voting equipment (such as tabulators) only being ‘triggered’ to manipulate votes after a certain threshold of votes were cast or counted. As many state audits process batches of votes in batches of about 200-280 votes, this kind of functionality could reduce the risk of compromised machinery being identified through pre-election machine testing or post-election audits.

3. Turnout Analysis

(Does Unusually Large Turnout Benefit One Candidate Disproportionately?)

There is also a notable difference in Philadelphia, Allegheny, and Erie county voting data patterns when comparing Mail-In Voting and Election Day turnout relative to:

The number of votes received per candidate (shown in histograms) and/or
Vote share received by candidate (shown in scatterplots).

In this report, “turnout” specifically means the percentage of voters who cast a ballot relative to the number of registered voters. In Pennsylvania, precinct-level voter registration data is available – so we have calculated turnout percentage by precinct.

Why Analyze Turnout?

To some extent, different voting patterns in different vote types might be expected: there are differences in the populations who tend to vote using these different methods. As mentioned previously, in the U.S., particularly since the 2020 COVID pandemic, Democrats are broadly more likely to vote by mail and Republicans are more likely to vote in-person on Election Day. (Source: Michigan Institute of Technology)

However, patterns present in Philadelphia, Allegheny, and Erie county election data displays a trend that has been observed in other elections internationally where vote manipulation is suspected: one candidate or party receiving a disproportionately high number of votes in areas that report unusually high turnout.

Multi-year studies have shown that in Western democracies, the relationship between turnout (how many voters go to the polls) and vote share (the percentage of votes a particular candidate receives) is surprisingly predictable and stable. While there is natural variation in these two factors (turnout and vote share), when represented visually, the overall shape of those two factors is “remarkably stable” across elections. (Sources: Borghesi and Bouchaud 2010, Klimek et al 2012, Behrens 2023)

In particular, high recorded voter turnout correlating with disproportionately large numbers of votes (or share of the vote) for one candidate/party can be an indicator that the candidate/party may be benefiting from vote manipulation.

Moreover, the ‘shape’ created by forms of manipulation such as ballot-stuffing can cause the shape of the represented data to change in specific ways.

Per Klimek et al 2012:

A different strategy for detecting signals of election fraud is to look at the distribution of vote and turnout numbers [...]. This strategy has been extensively used for the Russian presidential and Duma elections over the last 20 years. These works focus on the task of detecting two mechanisms, the stuffing of ballot boxes and the reporting of contrived numbers. It has been noted that these mechanisms are able to produce different features of vote and turnout distributions than those features observed in fair elections. [...]

Here, it was also observed that ballot stuffing not only changes the shape of vote and turnout distributions but also induces a high correlation between them. Unusually high vote counts tend to co-occur with unusually high turnout numbers. (emphasis added)

Put simply: if a candidate receives a high percentage of votes cast, but disproportionately receives those votes in places where there was very high voter turnout, it may be an indicator that some of those votes were artificially inflated.

This next section compares turnout relative to number of votes received by Mail-In Voting and Election Day voting data in all three counties through the use of histograms.

Histograms and Global Comparisons

One challenge with scatterplots is that it can be difficult to see how the “occurrences” (the coloured circles that represent precincts) are distributed among the two candidates. Particularly if there are a large number of data points (in our case, precincts), the actual volume can become challenging to discern.

Normal data that is not manipulated and follows natural variability often forms a bell curve, or ‘normal distribution.’

When data forms a bell curve, most ‘values’ (the things that we’re measuring) fall close to the average (the ‘middle’ of the bell curve). The further away from the average a value is, the less likely it is. When representing data visually, whether or not the data you are representing ‘fits’ a bell curve can be an indicator of whether something unusual or unexpected happened.

Evidence of manipulated election data in other countries has been identified through this approach. One known indicator of unfair elections is an anomalous deviation from normal distribution, wherein one candidate receives a disproportionate number of votes in areas where turnout is high. Referred to as a ‘Russian tail’, such a spike may indicate election result falsification, particularly if only one candidate benefits.

Source: Sergei Shpilkin, Elections Statistics Roundtable 2017

(title cards and markup added)

Note that in the bottom right-hand corner of the charts displaying Russian election results, there is a ‘bump’ coloured in with pale pink hatching. The hatching highlights the votes that Shpilkin deems were likely the product of ballot-stuffing efforts, increasing the turnout in the areas where ballots were stuffed while only benefiting one candidate.

As line graphs are often used to represent temporary fluctuations (i.e. market fluctuations) rather than finalized data, our team has opted to use histograms in our analysis.

Histograms are another way to represent data visually. Histograms merge data into a simpler view than scatterplots, allowing us to show a ‘higher level’ impression of certain trends. While histograms should also display a ‘bell curve’

In the charts below:

The y axis is the number of votes cast;
The x axis is the percentage of voter turnout;
The coloured bars (called “bins”) represent precinct-level results, showing how many votes each candidate received at different turnout levels.
- Each bin represents a range of 2.5% voter turnout per bin.
- The colors in the charts below indicate:
  - Votes for Harris ( Blue)
  - Votes for Trump (Red)
  - Overlap to Show Scale (Purple)

Philadelphia

Philadelphia is an overwhelmingly Democratic city, having voted for the Democratic Presidential candidate by 81.44% in 2020, 82.53% in 2016, and 85.2% in 2012. As such, the majority of precincts voting mostly Democratic is in line with expectations.

When ‘stacked’ on top of each other, we can see the ‘shape’ of both candidates’ votes received in relation to how many registered voters cast a ballot.

With Mail-In votes, the results for both candidates align with normal distribution – they largely follow a pair of bell curves. While the Democratic candidate received more votes in Philadelphia than the Republican candidate, the respective histograms ‘peak’ at about the same level of turnout (around 15%-25%).

Neither candidate appears to benefit disproportionately from areas with unusually high turnout.

In the Election Day results for Philadelphia, there is a noticeable shift. In the charts showing the candidates’ respective Philadelphia Election Day results below, observe what happens to the Trump distribution as reported by precincts that measured high levels of voter turnout:

At the 58% turnout threshold, there is a spike in the number of votes for Trump.

This is the only time that Trump exceeds Harris’s number of votes received relative to turnout. The dashed line shows the point at which the number of votes received by Trump exceeds the number of votes received by Harris – which only happens at high turnout level, above the 58% turnout mark.

The ETA has also represented and shared this data in line graph form. For comparison with the Shpilkin turnout analysis above, here is Philadelphia Mail-In and Election data election result data represented in line graph format:

Once again, greater numbers of Trump votes appear to only overtake the number of Harris votes at high turnout levels, in this case at about 52% turnout as shown by the dashed line. It is unclear why increased enthusiasm for Trump would not manifest on Election Day in precincts where there were fewer votes cast and counted.

Ultimately, in Erie County Trump won 50.05% of the vote (68,866 votes total) compared to Harris’s 49.03% (67,456 votes total). This aligned with state and national election outcomes, meaning Erie County appeared once again to have made good on its bellwether status.

Turnout Trends – Total Votes, All Types:

In Pennsylvania election results data, the following turnout trends are present:

When examined as a whole, areas that received over about 70% voter turnout receive an disproportionately high volume of votes for Trump
Because U.S. Presidential elections are largely a race between two candidates, there is a corresponding decrease in votes for Harris in areas above 70% voter turnout.
This pattern is present with varying degrees of severity in numerous counties.
This trend is driven by Election Day data, not Mail-In Voting data.

For the purpose of visualization, we can easily see the difference between the degree of Trump vote share relative to Harris vote share in precincts that recorded below or above 70% turnout:

Note:

As part of our turnout analysis, the ETA also developed two sets of turnout-related scatterplots:

Scatterplots showing turnout in relation to vote share percentage (a format considered more easily comparable with election results in other places); and,
Scatterplots showing turnout in relation to the number of votes cast (to provide an alternate way to depict the same data shown by the histograms).

To prevent this report from becoming too long for our web provider to accommodate, these two sets of scatterplots will be available for viewing shortly once we migrate to our new web provider.

What could cause these patterns?

Hypothesis 1: “More Republicans Just Showed Up on Election Day” (But Disproportionately in Larger Precincts)

As mentioned above, this hypothesis may be initially appealing in that it appears, at first, to align with the expectation that Democrats are broadly more likely to vote by mail and Republicans are more likely to vote in-person on Election Day.

However:
- Increased Republican turnout on Election Day is not unexpected, nor is it concerning. It is where that turnout occurs (in all precincts, or just certain ones?) and who benefits (one candidate disproportionately, or both?) that is concerning.
- The trends we see in Pennsylvania on Election Day are similar to the trend we see in Clark County, Nevada in Early Voting data.
- The United States does not exist in a vacuum. Other countries also have political parties or candidates that receive higher or lower turnout. There are academically rigorous models and methodologies developed over 30+ years dedicated to assessing manipulated versus unmanipulated elections. Certain precincts recording high voter turnout that benefits only one candidate has, in multiple countries on multiple occasions, aligned with places where there is substantive evidence of manipulated elections.

Independent election observer, data analyst, and forensic expert Roman Udot describes the change in Armenian election results pre- and post- Velvet Revolution:

“I gave lectures [in Armenia] on how rigged elections look. The authorities used to claim, ‘You don’t understand, that’s just how people in the villages vote — high turnout, and they love the government.’ Then the ‘Velvet Revolution’ happened, and the election commissions didn't follow Pashinyan’s orders, so he came to power without fraud. As a result, the [Russian Tail in the election result data] immediately disappeared, like a dog’s tail falling off. Now, there are no more unusually high turnouts in the mountain regions. Before, they were just fabricating the numbers, reporting inflated figures to please their superiors, and taking advantage of the fact that no one went there to monitor.”

Hypothesis 2: Geographic Differences Causing Turnout Difference?

There may be some geographic connection between precincts where more votes are cast that could correlate with a higher degree of Trump voter turnout.

Precincts with more votes appear more likely to correlate to urban precincts where more people live. Our assumption would be that higher turnout in these areas would benefit both candidates rather than just one.

Additional analysis of which precincts are affected by these trends is our next intended step in order to better understand the geographic extent of where they occurred.

Hypothesis 3: Vote Manipulation, Such as Electronic Ballot-Stuffing, Resulting in Inflated Turnout in Only Some Precincts

Vote manipulation can include adding ballots, removing/destroying ballots, and/or ‘flipping’ votes from one candidate to another.
- One known weakness of turnout analysis in election forensics is that while it is a useful tool for detecting the addition (and sometimes removal) of votes, it is less useful for identifying ‘flipped votes’. (This is because changing a vote to be for one candidate rather than another has no effect on the underlying voter turnout, just on the results of the votes cast.)

In principle, electronic ballot-stuffing is very similar to physical ballot-stuffing. The difference is that, instead of a physical ballot box and handfuls of extra ballots, bad actors may instead use a hacked “ePoll book” and/or a compromised vote-counting tabulator.

The trends in Philadelphia, Allegheny, and Erie Election Day turnout analysis is reminiscent of trends in election results from Russia in 2000. There is considerable contemporaneous reporting, eyewitness testimony, and subsequent analysis indicating both ballot-stuffing and “doctoring of official results” were factors in that election outcome.

Summary

A combination of factors raise election integrity concerns in Pennsylvania. This includes, but is not limited to:
- Large-scale acquisition of names and street addresses of registered Pennsylvania voters occurred just before the 2024 U.S. Presidential election. This is concerning in light of severe election security breaches between 2020-2024, as such information (names and addresses of registered voters) could – in the wrong hands, with the right access – be used to falsify votes.
- Disruptions occurred across Pennsylvania on Election Day, potentially granting bad actors an opportunity to interfere with vulnerable election infrastructure. This includes bomb threats and machine failures, resulting in deviation from normal voting procedures.
- Comments made by Trump about “vote-counting computers” during a pre-inauguration rally in Washington, D.C. on January 19, 2025 – specifically, a reference to Elon Musk’s familiarity with such computer in relation to “winning Pennsylvania like in like a landslide”.
Our analysis of Philadelphia, Allegheny, and Erie election results indicates patterns consistent with vote manipulation. These patterns are present in Election Day result data, but not in Mail-In Voting data, using multiple analytical approaches. This includes:
- Drop-Off Analysis. We compared Presidential election results to Senate election results relative to historic Pennsylvania voting trends. While the drop-off rate for Mail-In ballots is as expected, the drop-off difference for Election Day results is far more pronounced between the Democratic versus Republican candidates. One potential cause for a difference in drop-off rates is failure to interfere with multiple races to equal extents.
- Vote Share by Vote Count Analysis. We measured the share of the vote received by each Presidential candidate relative to the number of votes cast at a given precinct. In doing so we identified a disproportionate increase in the Republican candidate’s share of the vote in precincts where more votes were cast and counted. This is similar to what we saw in tabulator data from Clark County, Nevada, where machines that processed higher numbers of votes showed a visible skew in favor of the Republican candidate.
- Turnout Analysis. One candidate benefitting from unusually high turnout has been credibly associated with election fraud in other countries. In all three counties, we observed that the Republican presidential candidate received disproportionately more votes in precincts with unusually high turnout. A similar uptick in Trump votes in lower-turnout precincts is not present. Unusually high voter turnout may be cause for further scrutiny, in particular in relation to whether that turnout may have been artificially inflated through electronic ballot-stuffing.

Hand Audits of Paper Voting Records Recommended. We are calling on state and local election officials in Pennsylvania to advocate for – and, for those with the authority to do so, undertake – full hand audits of paper voting records from the 2024 U.S. Presidential election in Pennsylvania.
- The goal of this exercise would be to ensure that paper ballot records match the reported vote totals, and that results as reported were not manipulated – electronically or otherwise.
- If nothing else, future elections must be safeguarded. The potential harm that could come from leaving an election unverified is far greater than the work it takes to conduct a hand-recount and verify the results.
- We are also launching a Call For Audits Toolkit to support regular people in reaching out to officials and representatives and add their voice to the call.

Remaining Questions

Location of Precincts Can any of the trends identified in this report be explained, in full or in part, by the geographic location of the precincts that display concerning trends? To explore this further, ETA is currently exploring options for displaying data for ourselves and others in a way that allows for easier identification of specific precincts that display concerning trends.
Trends Present in Other Pennsylvania Counties? This analysis has been limited to just three of Pennsylvania’s sixty-seven counties. The extent to which these trends may also be present in other counties requires additional analysis and exploration. Limitations associated with the data published by certain counties may be a limiting factor.
When Was PA’s Last Entirely Hand-Counted Election? In attempting to examine historic trends in these three counties, our team acquired records of the total votes cast for past elections. We encountered two challenges in comparing pre-2020 data to 2024 data: a) lack of comparable “vote type by precinct” data, and b) lack of clarity around what an appropriate “baseline” year may be, i.e. how long has Pennsylvania been using electronic voting infrastructure, and when was the last time an election in Pennsylvania was cast and counted entirely by hand? If you have local information or contacts that may help answer this, please reach out to us directly at board@electiontruthalliance.org.

Our Next Steps

Truth matters. Confidence in the integrity of our elections is essential for a successful democracy. Here are some of our immediate next steps:

Call for Audits

We urge voters and election officials in the state of Pennsylvania to seek additional hand audits of paper voting records from the November 5th, 2024 General Election. We believe there are too many statistical anomalies in this election to not investigate further. If the cost of a hand recounted audit and use of taxpayer funds to pay for those costs is a barrier, our organization is willing to enter into discussions about a cost-sharing agreement. We have sent a summary of our findings to state and local officials, and will be sharing that letter publicly in the coming days.

Replicate and Corroborate

We are seeking others to recreate our analysis, review our work, and interrogate our findings. We believe you will reach the same conclusions about these statistical anomalies, but it’s important for third parties to review this work thoroughly.

How Can You Support the Election Truth Alliance?

Share Our Findings: Please consider sharing our findings with local media outlets, politicians, and election security experts in Pennsylvania. Spreading the word about these issues is crucial to our election security. Our "Call for Audits Toolkit" is currently under development and will be posted to our website in the coming days.
Sign Our Petition: Add your voice to others urging for Pennsylvania officials to undertake a hand audit of paper voting records in Pennsylvania.
Volunteer: The Election Truth Alliance needs help from a wide variety of skill sets. If you think you can help, please fill out our Volunteer Sign-Up form on our website.

Donate To Support Our Work: We are a non-partisan, non-profit organization that doesn’t accept money from politicians or political action committees. You can help our work by donating at the ‘Donate’ tab on our website ElectionTruthAlliance.org

Advocate For Transparency: Contact your local officials and request that all ballot data be made public. This information is crucial in our effort to #VerifyTheVote
Sign Our Petition: Add your voice to the 17,000 people who have called for hand audits of paper ballot records in Pennsylvania. You don't have to be from Pennsylvania to sign!
Join Our Mailing List: Brand new! We will be sending out a newsletter to provide updates. If you're interested in receiving these, please join our mailing list!

Pennsylvanians, Contact Your Officials: If you are a voter in the state of Pennsylvania, contact your local Representatives and County Officials to seek answers to these questions, discuss our findings, and to express how serious of an issue this is.

Accreditation

Election Truth Alliance (April 2025)
Three Counties in Pennsylvania

Development and Drafting Leads:

Lilli M. & William B.

Featured Charts:

William B. & Thomas EK

Nathan T.

Chart Formatting:

Tasha M. & Lilli M.

Exploratory Analysis:
PT, Tasha M., Joshua C.

Data Processing:
Laila S.

Data Analysis:
William B., Thomas EK, Lilli M.

Copy Editing:

Kim H.

Advisory:

Nate WT, Ben H.

And many, many others who provided review, support, and input.

Posted: April 7, 2025

Last Updated: May 9, 2025

Other Expert Analysis

eforensics Analysis of Three Pennsylvania Counties

in the 2024 Presidential Election

Walter R. Mebane, Jr. (May 6, 2023)

Dr. Walter R. Mebane, Jr. is a leading U.S. expert in election forensics and detecting election fraud. He is a professor of political science and statistics at the University of Michigan.
Summary of Working Report (drafted by ETA):
- Dr. Mebane's analysis using the 'eforensics' method identifies approximately 29,000 potentially fraudulent or distorted votes across three Pennsylvania counties (Philadelphia, Allegheny, and Erie). This estimate represents approximately 24% of the statewide margin of victory in Pennsylvania.
- There are 64 other counties in Pennsylvania that Professor Mebane and the Election Truth Alliance have not yet analyzed.
- Across the three counties, these concerns appear predominantly linked to election-day voting.
- Dr. Mebane's analysis suggests that Philadelphia County appears to demonstrate the most concerning indicators among the three counties analyzed.
A brief overview of Dr. Walter R. Mebane, Jr.'s election forensics writings and a plain language characterization of his eforensics model of estimating fraudulent votes —can be found here.

Download PDF

Sources Used In ETA's Analysis

Source Data

Other Sources

United States Census Bureau. State Population Totals and Components of Change: 2020-2024. https://www.census.gov/data/tables/time-series/demo/popest/2020s-state-total.html (Archived)

Statistica. Number of electoral votes from Pennsylvania designated to each party's candidate in U.S. presidential elections from 1789 to 2020. https://www.statista.com/statistics/1130755/pennsylvania-electoral-votes-since-1789/ (Archived)

AdImpact. Pennsylvania Saw a Record Breaking $1.2B in Election Ads.

https://adimpact.com/blogs/blog/in-review-24-pennsylvania-saw-a-record-breaking-1-2b-in-election-ads

Verified Voting. Pennsylvania. https://verifiedvoting.org/verifier/#mode/navigate/map/makeEquip/mapType/normal/year/2026/state/42 (Archived)

World Population Review. Pennsylvania. https://worldpopulationreview.com/states/pennsylvania (Archived)

Commonwealth of Pennsylvania. Election Reporting Center. https://www.electionreturns.pa.gov/ReportCenter/Reports (Archived)

Philadelphia City Commissioners. Resources & Data | Archived Data Sets. https://vote.phila.gov/resources-data/past-election-results/archived-data-sets/ (Archived)

City of Philadelphia. Voter Election Registration and Turnout. https://opendataphilly.org/datasets/voter-turnout/ (Archived)

Clarity Elections. Allegheny County, Pennsylvania. 2024 PRESIDENTIAL ELECTION November 5, 2024 | Official Results. https://results.enr.clarityelections.com/PA/Allegheny/122424/web.345435/#/reporting (Archived)

Erie County, Pennsylvania. Election Results. https://eriecountypa.gov/departments/elections-voting/election-results/ (Archived)

Commonwealth of Pennsylvania. Open Data Portal. Population Estimates Statewide & County Current (Census) data.PA.gov https://data.pa.gov/Census-Economic/Population-Estimates-Statewide-County-Current-Cens/hv5f-e4e3/about_data (Archived)

Commonwealth of Pennsylvania Department of State. 2024 Voter Registration Statistics (Archived)

World Atlas. The Great Lakes Ranked By Size. https://www.worldatlas.com/lakes/the-great-lakes-ranked-by-size.html (Archived)

America Political Action Committee. https://theamericapac.org/ (Archived)

AP News. Musk offers voters $1 million a day to sign PAC petition backing the Constitution. Is that legal? https://apnews.com/article/musk-1-million-giveaway-trump-voters-petition-b4e48acbfe04fde735e60b1911ad0197 (Archived)

Forbes. Elon Musk’s PAC Is Paying $47 For Each Solicited Petition Signature From A Swing State Voter—Here’s Why It’s Controversial. https://www.forbes.com/sites/maryroeloffs/2024/10/07/elon-musks-pac-is-paying-47-for-each-solicited-petition-signature-from-a-swing-state-voter-heres-why-its-controversial/ (Archived)

PBS. Musk’s PAC claims $1 million ‘winners’ not chosen by chance. https://www.pbs.org/newshour/politics/musks-pac-claims-1-million-winners-not-chosen-by-chance (Archived)

Election Law Blog. Elon Musk Veers Into Clearly Illegal Vote Buying, Offering $1 Million Per Day Lottery Prize Only to Registered Voters. https://electionlawblog.org/?p=146397 (Archived)

CNN. Elon Musk’s daily $1 million giveaway to voters can continue, Pennsylvania judge rules. https://www.cnn.com/2024/11/04/politics/elon-musk-1-million-giveaway/index.html (Archived)

America PAC FAQ - https://petition.theamericapac.org/faq (Archived)

C-SPAN. User Clip: Trump Talking About Elon Musk Knowing About Voting Computers. https://www.c-span.org/clip/public-affairs-event/user-clip-trump-talking-about-elon-musk-knowing-about-voting-computers/5150057 (Archived)

U.S. Senate Select Committee on Intelligence Russian Interference in the 2016 U.S. Hearings: Elections Expert Testimony by J. Alex Halderman, Professor of Computer Science, University of Michigan. (2017) https://www.intelligence.senate.gov/sites/default/files/documents/os-ahalderman-062117.pdf (Archived)

Election Truth Alliance. Election Truth Alliance YouTube Channel. "Vote-Counting Computers" - Donald Trump Thanks Elon Musk for Pennsylvania Win #VerifyTheVote. https://youtube.com/shorts/DbHqZjNTu_0?si=wUUnnAOIRdBgDhmo (Archived)

Election Truth Alliance. 2024 U.S. Presidential Analysis: Clark County, Nevada. https://img1.wsimg.com/blobby/go/9087f51c-d3bd-4002-9943-79706c6e82a3/Election%20Truth%20Alliance_Clark%20County%20N-a246330.pdf (Archived)

Federal Bureau of Investigations Press Office. FBI Statement on Bomb Threats to Polling Locations.https://www.fbi.gov/news/press-releases/fbi-statement-on-bomb-threats-to-polling-locations#:~:text=The%20FBI%20is%20aware%20of,among%20the%20FBI's%20highest%20priorities. (Archived)

USAToday Network. Hoax bomb threats sent to at least 32 Pa. counties on Election Day, police say. https://www.goerie.com/story/news/politics/elections/state/2024/11/07/pa-bomb-threats-election-day/76109630007/ (Archived)

Cambria County. Cambria County done counting election day ballots, about 30k more ballots than expected. https://wjactv.com/news/local/cambria-county-done-counting-election-day-ballots-about-30k-more-ballots-than-expected (Archived)

The Tribune Democrat. Cambria commissioners say printing error caused ballot problems, new protocols announced. https://www.tribdem.com/news/cambria-commissioners-say-printing-error-caused-ballot-problems-new-protocols-announced/article_02fc75fc-df2c-11ef-80ce-5b30918b3bc1.html (Archived)

NBC Philadelphia. More time to vote in one Pa. county due to issues with voting machines - https://www.nbcphiladelphia.com/decision-2024/more-time-vote-one-pa-county/4019337/ (Archived)

Altoona Mirror. Cambria duplicates ballots for accuracy County officials said hand counting was ‘extremely slow’. https://www.altoonamirror.com/news/local-news/2024/11/cambria-duplicates-ballots-for-accuracy/ (Archived)

PA.gov All County Voting System Report https://www.pa.gov/content/dam/copapwp-pagov/en/dos/resources/voting-and-elections/voting-systems/malfunction-reports/combined%20voting%20system%20malfunction%20report%20final_redacted.pdf (Archived)

Commonwealth of Pennsylvania. 2024 All County Voting System Report. https://www.pa.gov/content/dam/copapwp-pagov/en/dos/resources/voting-and-elections/voting-systems/malfunction-reports/combined%20voting%20system%20malfunction%20report%20final_redacted.pdf (Archived)

Election Facts PA. How Elections Work in Pennsylvania. https://electionfactspa.com/ (Archived)

Pennsylvania Department of State. Guidance Concerning Civilian Absentee and Mail-in Ballot Procedures. https://www.pa.gov/content/dam/copapwp-pagov/en/dos/resources/voting-and-elections/directives-and-guidance/2024-guidance-civilian-absentee-mail-in-ballot-procedures-v3.1.pdf (Archived)

Democracy Docket. News Litigation Voting Pennsylvania Up to 17,000 Pennsylvanians Who Haven’t Received Mail-in Ballots Permitted to Vote Early in Person. https://www.democracydocket.com/news-alerts/tens-of-thousands-of-voters-in-pennsylvania-county-never-received-their-mail-in-ballots/ (Archived)

NBC News. How Pennsylvania's mail ballot rules will lead to thousands of provisional ballots on Election Day. https://www.nbcnews.com/politics/2024-election/pennsylvanias-mail-ballot-rules-will-lead-thousands-provisional-ballot-rcna178746 (Archived)

Commonwealth of Pennsylvania. Find your Local Polling Place to Vote in Person. https://www.pa.gov/services/vote/find-your-local-polling-place.html (Archived)

Commonwealth of Pennsylvania. Voting System Demos. https://www.pa.gov/agencies/vote/voter-support/voting-system-demos.html#sortCriteria=%40copapwptitle%20ascending%2C%40title%20ascending (Archived)

Commonwealth of Pennsylvania. 2024 Presidential Election (Official Returns) Statewide. https://www.electionreturns.pa.gov/General/OfficeResultsVM?officeId=1&districtId=1&ElectionID=105&ElectionType=G&IsActive=1&isRetention=0 (Archived)

SMARTelections. 2024 General Presidential Election Multi-State Drop-Off Comparison. Comparison of the Presidential Vote with the Next Down-Ballot Race by State. https://smartelections.us/dropoff (Archived)

Christian Borghesi, Jean-Philippe Bouchaud (2010). Spatial correlations in vote statistics: a diffusive field model for decision-making- https://arxiv.org/abs/1003.2807 (Archived)

Peter Klimek, Yuri Yegorov, Rudolf Hanel, Stefan Thurner. Statistical detection of systematic election irregularities (2012). https://pubmed.ncbi.nlm.nih.gov/23010929/ (Archived)

Lion Behrens (2023). Statistical Detection of Systematic Election Irregularities. https://madoc.bib.uni-mannheim.de/66208/1/BL_Dissertation_20231212.pdf (Archived)

FreeSpeechForPeople Letter. “Computer Scientists: Breaches of Voting System Software Warrant Recounts to Ensure Election Verification.” Letter dated November 13, 2024. https://freespeechforpeople.org/computer-scientists-breaches-of-voting-system-software-warrant-recounts-to-ensure-election-verification/ (Archived)

Free Speech For People. Letter to VP Harris. https://freespeechforpeople.org/wp-content/uploads/2024/11/letter-to-vp-harris-111324-1.pdf (Archived)

The Center for Politics. The 2024 Senate Undervote: Not High By Historical Standards.

https://centerforpolitics.org/crystalball/the-2024-senate-undervote-not-high-by-historical-standards/?__cf_chl_tk=Vj5TLzpvDQD44oXGeyWetykBnrEjIrxGM90ZfdNnz74-1743637175-1.0.1.1-e8irotbfaP91iNrzMxvrXvstmVDvh97AIrWKZf1rXcw (Archived)

The Insider. Georgian parliamentary election voting charts indicate large-scale fraud, experts say. https://theins.ru/en/news/275735 (Archived)

UPI Archives. Report: Russian election falsified (2000).

https://www.upi.com/Archives/2000/09/11/Report-Russian-election-falsified/2936968644800/

Electoral Graphics/Sergei Shpilkin. Election Statistics Roundtable (2017). https://www.electoral.graphics/en-us/Home/Articles/sergei-shpilkin-statistical-analysis-of-elections (Archived)

Posted: April 7, 2025

Last Updated: May 9, 2025