[Election Wizzard] US Census data released last week called into question the official vote tally from the 2020 election. As part of the Census, the government collects data on citizens who self-report as having voted in presidential elections. The collected data shows an unusual anomaly in the reported results.According to the Census, the recorded number of people voting in 2020 was tallied at 154,628,000. On the other hand, official results place the number of actual ballots cast slightly north of 158 million. That’s a discrepancy of nearly four million votes.
Speaking to pollster Richard Baris during an episode of "Inside the Numbers," lawyer Robert Barnes said historically, the Census tends to "pin on the nose" the recorded vote numbers with the actual results. In other words, often the two data sets reasonably match.
Barnes is right. For example, the bureau was nearly spot-on in 2008, slightly under-reporting that 131,100,000 voted, while the official results showed 131,300,000 ballots cast.
Of course, sometimes the Census has missed the mark. But for decades, in almost every case where the Census grossly botched the results, it was because the bureau over-recorded the number of those who voted.
Consider the following: In 1992, the Census over recorded the official results by slightly more than nine million. In 1996, the Census again over recorded the number of reported voters by roughly nine million. Similarly, the bureau recorded the number of those who voted in the 2004 election as 125 million, while official results placed the total at 122 million.
The same over-recording phenomenon occurred in 2012, with the Census over-reporting the number of voters by several million. In fact, even in 2016 where the Census was quite close, it still over-recorded the official election results.