How many illegal votes in the 2020 presidential election
Executive summary
The best, publicly available post‑election reviews and court rulings found no evidence of widespread or outcome‑changing illegal voting in the 2020 U.S. presidential election; methodical reviews flagged fewer than 475 potential instances of unlawful or questionable votes out of more than 160 million cast, and courts dismissed broad fraud claims as unsupported [1] [2] [3]. Independent statistical and legal analyses conclude that the anomalies promoted as proof of mass fraud do not withstand scrutiny and would not have altered the result [4] [5].
1. What the comprehensive counts actually found
A multi‑state, detailed review by reporters and local election officials — synthesized by the Associated Press and reported by PBS and other outlets — identified fewer than 475 potential instances of voter fraud across six key battleground states out of more than 25 million ballots in those states, a tally that investigators said would not have affected the presidential outcome [2] [1]; that AP‑centered figure is the closest public accounting to a consolidated estimate of “illegal” votes produced from local canvassing and post‑election audits [1].
2. What courts and official reviews concluded
Numerous court cases lodged to overturn or decertify results were dismissed or rejected on lack of evidence, jurisdictional grounds, or because claims rested on hearsay and speculative analysis rather than concrete proof; courts specifically found many fraud declarations to be based on anonymous witnesses, hearsay, and irrelevant comparisons to other elections [3] [5]. Officials including the Justice Department under Attorney General William Barr likewise reported no evidence to substantiate allegations of systemic vote‑counting irregularities that would change the outcome [6].
3. What statistical experts said about the claims
Peer‑reviewed and scholarly assessments of the statistical arguments used to allege mass fraud concluded those claims were not convincing: analyses found purported anomalies either mischaracterized, explainable by benign factors like demographics and voting method differences, or simply not anomalous when properly modeled, undermining assertions that millions of illegal votes were cast [4] [7].
4. The scale of confirmed, prosecutable fraud versus rumors
While isolated instances of illegal voting and fraud prosecutions did occur — and outlets such as Newsweek and other records document individual arrests and cases tied to the 2020 cycle — these were rare, geographically scattered, and overwhelmingly non‑systemic; databases and reporting capture some prosecutable incidents but not anything approaching the scale claimed by broad “stolen election” narratives [8] [9]. Media fact‑checks and election officials repeatedly exposed viral claims (for example, misattributed precinct data or misread registration rolls) as errors or misinterpretations rather than proof of massive illegality [10].
5. Limits of available reporting and remaining uncertainties
Public reporting and court records provide strong convergence that large‑scale illegal voting did not occur, but precise, single‑number certainty about “how many illegal votes” is constrained by variation in how jurisdictions investigate, classify, and publish post‑election irregularities; the most authoritative aggregated count available in public reporting is the AP’s compilation of fewer than 475 potential cases in contested states, and scholarly work and courts treated that as representative of the scale of confirmed or credibly alleged fraud [2] [1] [4]. Where claims fall outside documented evidence, reporting cannot and does not assert definitive counts beyond the documented investigations cited above [3] [5].