Let’s be honest: when the Department of Homeland Security first announced that more than 15,000 noncitizens were potentially registered to vote in Nevada, it was the kind of headline designed to stop you cold. Fifteen thousand! That’s not a rounding error, that’s an alleged invasion of the voter rolls. But then came the revision. The Trump administration quietly walked that number down to fewer than 200 possible cases. That’s not a minor adjustment, that’s a complete collapse. Imagine being accused of something based on a statistic that shrinks by roughly 98 percent. You’d want an explanation. You’d want to know how the original number was so spectacularly wrong, and how the new number was calculated. And above all, you’d want to see the evidence. But instead, we’re left with a government agency that apparently cannot make up its mind, and worse, cannot explain how it arrived at either figure. The only thing we know for sure is that DHS is claiming something, local officials are saying they can’t verify it, and the public is being asked to take a huge leap of faith. That’s not how credible investigations work, and it’s absolutely not how democracy should work.
The most frustrating part is that DHS refuses to share the methodology behind its new, dramatically reduced number of 185 potential noncitizen voters. Nevada Secretary of State Cisco Aguilar, a Democrat, has been put in an impossible position. He’s told, essentially, “trust us, we found 185 suspicious people,” but he hasn’t been given the information needed to check whether those people are actually noncitizens or actually registered to vote. Based on the limited details that DHS did share, Aguilar has said that the supposed noncitizens identified by the administration are likely citizens. But he can’t be sure, and he shouldn’t have to be in the dark. The federal government is tasked with protecting election integrity, but it seems to be using that responsibility to create confusion rather than clarity. The idea that DHS would withhold its methods is bizarre. If you have a list of 185 people who you believe are noncitizen voters, the natural next step is to hand that list to the people responsible for voter rolls and say, “here you go, investigate these folks.” Instead, the administration is saying, “we’ve done the analysis, but we won’t show our work.” It’s like a student who claims a perfect score but refuses to turn in the test. There are two possible explanations: either the analysis is so sloppy that it would fall apart under scrutiny, or there is no real analysis at all. Neither option inspires confidence.
Nevada isn’t the only state caught in this maddening loop. Consider Georgia, where Republican Secretary of State Brad Raffensperger has been genuinely aggressive about trying to find noncitizens on the voter rolls. Before Donald Trump returned to office, Raffensperger’s office was already conducting its own review. The result? Fewer than a dozen possible noncitizens were found in the entire state. That’s not nothing, but it is remarkably rare. Then the Trump administration announced that it had evidence of more than 400 noncitizens who supposedly voted in Georgia’s 2020 presidential election. A number that is staggeringly higher than what the state’s own officials found. Raffensperger, who is not exactly a Democrat and not exactly an enemy of Trump, immediately asked for the detailed information so he could “detect the alleged 400 noncitizens who would have voted in Georgia in 2020.” He wanted to investigate, not just read a press release. And what did DHS give him? Essentially nothing. No methodology, no names, no supporting documentation. So here you have a Republican election official, someone who was already looking for this exact kind of fraud, asking the federal government for proof of its blockbuster claim and being stonewalled. That’s not a partisan attack. That’s a professional asking for evidence and being met with silence. It’s the same story repeated in state after state: the administration makes a huge claim about illegal voting, local authorities ask to see the underlying evidence, and the evidence never arrives.
Part of the problem appears to be the way DHS is constructing its numbers. Experts say the administration seems to be matching U.S. Census Bureau data to voter rolls in order to identify potential noncitizens. On its face, that might sound reasonable. But the census data lacks the sort of personally identifiable information needed to confirm that the person on the census list is the same individual on the voter list. There’s no social security number, no exact date of birth, no reliable unique identifier connecting the two datasets. Instead, the process likely relies on names, maybe addresses, and other imperfect comparisons. That opens the door for spectacular amounts of error. Two different people with the same name in the same state can be incorrectly linked. A naturalized citizen might be flagged as a noncitizen because of an old database mistake. A woman who changed her name after marriage could be counted twice. The possibilities for false positives are endless. And the real-world results confirm that. ProPublica’s analysis found that between January 2025 and May 2026, fewer than 150 noncitizens were referred for prosecution under this administration, and only 41 actual criminal cases were brought. Forty-one. Out of tens of thousands of supposed potential cases. If DHS had really identified 15,000 or even 16,000 noncitizens registered to vote, you would expect thousands of prosecutions. Instead, we get 41 cases, and even those are likely the most solid examples among a mountain of shaky accusations. That’s an anemic outcome for an administration that talks about widespread voter fraud as if it were an existential crisis. Prosecutors love to bring election fraud cases, especially when the political winds are behind them. If the evidence were real, they would be filing charges left and right.
Even if you take the administration’s estimates entirely at face value, the numbers tell a story that is far less dramatic than the rhetoric. In August, the administration claimed to have evidence that about 24,000 noncitizens voted in the 2020 election. Let’s assume that’s true. That would represent less than 0.02 percent of all votes cast in that election. That is not enough to change a single statewide result, much less decide the presidency. It’s a rounding error, not a conspiracy. And when you look at states that have cooperated with the administration, the story is even more striking. In Ohio, roughly 1,000 voters were suspected by local and federal authorities of potentially being noncitizens. After review, how many cases were referred for prosecution? Two. Two cases out of a thousand suspected voters. That’s a confirmation rate of 0.2 percent. It makes you wonder whether the initial “suspected” list was based on anything serious at all. Yes, there are isolated instances of noncitizens registering or voting. That is a crime, and it should be investigated and prosecuted. But there is a massive difference between a handful of illegal votes scattered across a nation of hundreds of millions of voters, and the kind of widespread fraud the administration is implying. The facts simply don’t support the idea that noncitizens are clogging up voter rolls or swinging elections. The existing studies, the state audits, the local officials, and even the federal prosecution numbers all point in the same direction: this is exceedingly rare.
So what is really going on? Nevada Secretary of State Aguilar may have said it best when he told NPR that these kinds of claims, made without explaining how DHS arrived at its determinations, “just cause chaos and fear ahead of a critical midterm election.” That is precisely the point. The chaos and fear are not accidental by-products of a clumsy investigation. They are the intended result. If the administration were genuinely interested in prosecuting noncitizen voters, it would share its evidence with state officials, cooperate with prosecutors, and allow the legal system to work. Instead, it is dropping dubious numbers into the public sphere, refusing to answer questions, and letting the resulting outrage spread online and on television. That isn’t election integrity. It’s election sabotage. It’s a deliberate effort to convince the American people that our elections are untrustworthy, that illegal immigrants are deciding elections, and that the system is so broken that something drastic must be done. The real threat to American democracy isn’t the small handful of noncitizens who might accidentally end up on voter rolls. It’s the deliberate, sustained effort to manufacture a crisis that doesn’t exist, undermine public confidence in free and fair elections, and justify radical changes that could make voting harder for everyone. When a government agency has to rely on inflated numbers, secret methodologies, and headlines instead of evidence, it’s not protecting democracy. It’s attacking it. And the saddest part is that the strategy is working, because once people are convinced that elections are rigged, they stop believing in the outcomes, and that is exactly what the administration wants.