
DHS said Nevada’s rolls held up to 16,000 noncitizens—then could not produce a single verified name that Nevada agreed wasn’t a citizen.
Story Snapshot
- DHS floated a maximum of about 15,903 possible noncitizen registrants in Nevada.
- The agency later said only 185 were “confirmed” after manual review.
- Nevada officials said those 185 people were citizens, not noncitizens.
- The big number was called a “ceiling,” not a verified count.
The claim that opened a political firestorm
The Department of Homeland Security told Nevada in July that as many as 15,903 noncitizens might be on the state’s voter rolls, setting off alarms and headlines. The message focused on registration status, not proven illegal voting. Nevada asked for the names and evidence. Weeks later, reporters learned the agency described the 15,903 as a “ceiling,” meaning an upper bound from automated matches that needed manual checks before anyone could claim a hard count.
The number soon shrank. By mid-August, the government said 185 people had been manually confirmed as noncitizens on Nevada’s rolls, drawn from a fraction of the original list. That is a massive drop from the original claim. The agency also shared a separate subset of 6,218 “higher-confidence” matches for further review, which suggested a tiered process and uncertain data quality across the full set. The story now hinged on the quality of those 185 cases.
Nevada’s rebuttal flips the script
Nevada’s top lawyers and election officials reviewed the 185 records and said the people were citizens. Chief Deputy Attorney General Greg D. Ott put it in writing: “As for the 185 individuals, our records indicate that they were citizens,” and he added the government had provided “nothing of substance” to prove otherwise. The Secretary of State’s office echoed that the state’s records do not support the federal claim and that the agency still had not produced proof to change that view.
Competing claims hardened from there. The government said the 185 were the subset that had gone through manual review. Nevada said its documentation showed those same individuals were lawful citizens. One side cited internal checks; the other cited state files tied to the actual voter roll. Those positions cannot both be right. When facts clash, common sense says show the receipts. The government did not release a complete, line-by-line evidentiary list for the original total.
How the numbers fell apart
The “ceiling” framing admitted the big figure was preliminary. That matters. Database matching across voter files and immigration records is noisy. Names repeat, dates get mistyped, addresses change, and status updates lag. Even a tiny error rate at scale creates thousands of false matches. Officials said the 15,903 would need person-by-person confirmation. Then came the 185. When Nevada checked those, the state said each was a citizen. That undercut the only concrete slice on the table.
Critics saw a classic pattern: start with a dramatic top-line, walk it back, and stall on the underlying proof. Supporters countered that manual checks take time and that states should help resolve mismatches. Both points can be true in theory. But this dispute is not about theory. The test is whether any clear case in Nevada stands up. As of the latest public letters and reporting, Nevada says the answer for the 185 is no, and the agency has not proved otherwise.
What reliable election integrity should look like
Real election integrity demands tight claims, transparent evidence, and fast corrections when data is wrong. Start small. Validate names one by one. Publish the verification method. Share the documentation with the state that owns the voter file. If the federal side has ironclad proof of even a handful of true noncitizens on the rolls, it should release those records to the state and the public in a lawful, privacy-protected way. Precision beats spectacle. Voters deserve facts, not ceilings and walk-backs.
Sources:
reddit.com, theguardian.com, reviewjournal.com, newrepublic.com, thenevadaindependent.com
© restoreamericanglory.com 2026. All rights reserved.















