According to millions of Americans, President Donald Trump was right when he said that the 2020 presidential elections in the US had been rigged. We evaluate the most prominent statistical claims made by Trump and his allies as proof of election tampering, such as claims that Dominion voting machines switched votes from Trump to Biden, erroneously towering numbers in the Democratic buildup, and the allegedly puzzling failure of Biden to win “bellwether counties.”
In order to evaluate these assertions, we combine statistical analysis with original data analysis. The uprightness seen in the 2020 voting and more general issues surrounding election security and administration are topics that we hope our analysis will help advance in public discourse. Donald Trump refused to accept the results of the US presidential election in 2020, claiming unprecedented and widespread voting fraud. In an effort to shed uncertainty on the result, Trump’s supporters used a number of statistical reasons.
We review the most notable statistical assertions and come to the conclusion that none of them are even close to being true. The general reasoning behind these assertions is that certain elements of the pragmatic 2020 voting outcome would be improbable or impossible if the election had been fairly conducted. In every instance, we discover that the allegedly abnormal fact is either false or not abnormal.
In public declarations and legal actions after the 2020 US elections, President Donald Trump and several Republicans contested Biden’s victory. Even if election administration may have been faulty, the 2020 election was notable in several aspects (such as the relatively high turnout and mail-in voting rates). However, these statistical analyses show little evidence to back Donald Trump’s assertion that the election was rigged.
Trump advocates argued on the basis of a statistical analysis that there was a “one-in-a-quadrillion” chance that Joe Biden legitimately won the election. This claim comes from an expert report submitted as part of Texas Attorney General Ken Paxton’s lawsuit against the Commonwealth of Pennsylvania. In that report (12), Paxton claims that the expert, Charles Cicchetti, calculated a one-in-a-quadrillion chance of Biden winning; Cicchetti concludes his report by arguing that “In my opinion, the outcome of Biden winning… is so statistically improbable that it is not possible to dismiss fraud and biassed changes in the ways ballots were processed, validated, and tabulated.”
Cicchetti’s claim that Biden’s victory was “statistically improbable” is based on a gravely erroneous use of unacceptable premise-implication testing. The likelihood that Biden will win is never calculated by Cicchetti. Instead, he investigates the null hypothesis that the projected vote totals in specific states for Joe Biden in 2020 and for Hillary Clinton in 2016 were equal.
However, if the goal is to determine if Biden won honestly, then it is irrelevant if it was Biden or if Clinton also had the same level of projected support. It is illogical to assume that any variation in support among candidates is proof of election fraud because there are a variety of reasons why support can fluctuate between candidates.
Specifically, Cicchetti tests the hypothesis that the anticipated number of Democratic votes (for example, in Arizona) was the same for Joe Biden in 2020 as it was for Hillary Clinton in 2016. He does this by treating the figure of the Democratic election count as more than one specific algebraic expression circulated as a random variable.
In our opinion, the most notable statistical allegations of election fraud in 2020 have been thoroughly studied. Despite the variety of assertions, our conclusion holds true for each one: Whatever is claimed to be aberrant information concerning the vote consequence is not considered to be the truth or is not abnormal. If the alleged fact could be verified, it would hardly always be sufficient to prove that Joe Biden had been chosen illegally.
For instance, a little advantage for Biden in counties utilising Dominion machines could be attributed to chance, to variables not taken into consideration in statistical models, or even to pro-Trump fraud carried out by means of extra selection machines. In fact, when correctly measured or presented in the proper context, the apparently abnormal qualities we evaluate look normal.
