By yourNEWS Media Newsroom
Research published by political commentator Jerome Corsi in 2024 is drawing renewed attention following claims that the 2026 midterm election results followed a mathematical pattern consistent with an algorithm, prompting calls for additional examination of voter registration databases and paper ballots.
The renewed controversy centers on an analysis published by the social media account Truth Is Freedom, which examined 59 U.S. Senate and gubernatorial races and reported that Republican candidates received approximately 85% of President Donald Trump’s 2024 vote share across numerous states.
The researchers described the findings as an “85/15 Algorithm,” suggesting that approximately 15% of Trump’s previous vote share shifted toward Democratic candidates in 2026. A subsequent analysis reported an R-squared value of 0.87, indicating that the statistical model explained approximately 87% of the observed variation in the election results.
The findings were publicized by The Gateway Pundit and amplified by Trump, who shared the analysis on Truth Social following the Nov. 3 midterm elections.
Corsi argues that the statistical relationship resembles patterns he described in a 2024 American Thinker article examining research by Andrew Paquette, Ph.D., into voter identification numbers and registration databases.
In additional comments accompanying his earlier research, Corsi explained how a hypothetical election manipulation operation could use computer simulations to generate results favoring a predetermined candidate.
“In this article published in American Thinker, I explained how a political scientist could run a simulation of an election to achieve a desired outcome,” Corsi said.
“The political scientist could then throw away the actual votes in the election, run the simulation as the actual results, and the public would not know the difference.”
Corsi suggested that identifying a mathematical formula responsible for generating manipulated results could expose such an operation.
“The only way the election fraud could be determined with certainty is if some genius figured out tha algorithm the political scientist created to run the simulation,” he said.
He pointed to the reported R-squared value from the 2026 analysis as potentially significant, arguing that a strong mathematical relationship could be consistent with a predetermined voting pattern.
However, an R-squared value of 0.87 does not independently establish election fraud or demonstrate that an algorithm altered votes. The statistic measures how closely a particular model fits observed data, and strong correlations can arise from ordinary electoral trends, demographic similarities and differences in turnout between presidential and midterm elections.
Corsi’s earlier investigation focused on voter registration records rather than aggregate election results.
Paquette, whose work is documented through Gods Five Stones, reported identifying mathematical structures within voter identification systems in New York, New Jersey, Ohio, Wisconsin, Pennsylvania, Hawaii and Texas. He also examined records in Arizona and Georgia.
According to Corsi, those structures could potentially conceal fictitious voter registrations and allow unauthorized individuals to assign legitimate identification numbers to nonexistent voters.
Paquette’s research proposed a hypothetical scheme in which fraudulent registrations would be created, marked for later retrieval and used to request absentee ballots. After ballots were submitted, the associated records could allegedly be modified to conceal the activity.
In his analysis of New York voter records, Paquette estimated that approximately 338,000 registrations used in the 2020 election may have been improperly generated.
Those allegations have not been independently established as evidence that fictitious voters cast ballots or that election outcomes were manipulated.
Paquette’s work has also appeared in the Journal of Information Warfare, including research describing unusual patterns in voter identification number assignments. The existence of mathematical structures in database identifiers, however, does not by itself establish criminal manipulation.
Corsi maintains that sufficiently advanced statistical models could theoretically simulate election outcomes closely enough to resemble authentic voting patterns. He has also speculated that such operations would require access to election databases and potentially the involvement of individuals with sophisticated technical capabilities.
In a separate October 2024 essay, Corsi questioned whether intelligence agencies might eventually use computer-generated election simulations instead of actual voter results.
He connected those concerns to a 2017 World Economic Forum discussion in which founder Klaus Schwab considered whether increasingly advanced predictive technology could someday forecast election results.
“But since the next step could be to go into prescriptive mode, which means you do not even have to have elections anymore because you can already predict what, predict, and afterwards you can say, ‘Why do we need the elections?’ Because we know what the result will be. Can you imagine such a world?” Schwab said.
The World Economic Forum subsequently clarified that Schwab was describing a hypothetical technological development rather than proposing the elimination of democratic elections.
Corsi also cited Archbishop Carlo Maria Viganò’s writings on predictive programming, including his argument that fictional portrayals of future societies can influence public acceptance of political and technological changes.
Viganò wrote that such narratives involve “the use of the fictional literary genre as a tool for mass mental processing in order to make the population more willing to accept planned future events.”
The new 2026 analysis has nevertheless attracted scrutiny from researchers who question whether the apparent consistency of the reported voting pattern is unusual.
An independent examination published by Lead Stories found substantial differences between the proposed formula and actual state election results, challenging the suggestion that a uniform 85% multiplier accurately describes every race.
Other analysts identified comparable statistical patterns in earlier midterm elections, suggesting that changes in turnout and differences between presidential and congressional voting can produce similar relationships without manipulation.
The original Truth Is Freedom analysis also acknowledged that its findings would require examination of ballot-level evidence, including paper ballot counts, poll-book records and voter participation files, before conclusions about possible fraud could be established.
Those records would be essential to determining whether the reported statistical relationships reflect ordinary voting behavior or actual irregularities.
The distinction remains central to Corsi’s argument. While a computer could theoretically generate simulated election results, demonstrating that such a simulation replaced genuine votes would require evidence connecting the mathematical model to election systems, ballot records or unauthorized changes in certified results.
No such connection has been established by the reported 2026 R-squared value.
The debate has renewed interest in Corsi and Paquette’s earlier research while raising broader questions about election transparency, voter registration database security and independent verification of election results.
For now, the 85/15 analysis remains a disputed statistical interpretation rather than verified evidence of a manipulated election. Its underlying claims would require reproducible data, independent examination and direct ballot-level verification to establish whether any fraudulent activity occurred.
Original article
Comments