Despite the fact there is zero evidence—none, nada, zilch—that widespread illegal voting had any sort of serious impact on the presidential election, Donald Trump went on a tweetstorm Sunday asserting that millions of people voted illegally.

So to defend their boss to perplexed reporters, Trump’s spokesmen cited a years-old study—and that study’s own authors say it doesn’t prove the existence of voter fraud.

Trump’s penchant for floating bombshell assertions with no evidentiary support seems boundless. In Sunday tweets, for instance, he wrote that there was “serious voter fraud” in California, New Hampshire, and Virginia (all states he lost). So, during the conference call, NPR’s Tamara Keith brought up Trump’s charges on the Trump transition team’s daily conference call on Monday morning. What evidence, she asked, is there to support his claims? Jason Miller, a spokesman for the transition team, responded by criticizing reporters.

“It’s been ridiculous that so much oxygen’s been given to the recount effort where there’s absolutely no chance of any election results changing,” he said.

But, of course, having a low view of Jill Stein’s efforts does not prove that voter fraud is a thing. So Miller then pointed to two pieces of evidence: a 2014 Washington Post article discussing a report in the journal Electoral Studies which argued that enough non-citizens vote that they could “change the outcome of close races.”

“That Washington Post story is important,” Miller said.

It’s also highly suspect, according to the Washington Post.

Editors added a note to the top of the piece after publication citing three rebuttals of the study that it itself published, as well as a peer-reviewed article that also ran in Electoral Studies and criticized the report’s findings. Needless to say, plenty of reporters and editors shared Miller’s view that the Post’s story (which is more than two years old) was important and worth covering. It also should be needless to say that an article from October of 2014 could not prove the existence of voter fraud in New Hampshire in 2016, since time travel is not a thing.

But the Washington Post’s story wasn’t the only report Miller marshalled to make his case; he also cited a report that Pew Charitable Trusts published in February 2012 titled “Inaccurate, Costly, and Inefficient: Evidence That America’s Voter Registration System Needs an Upgrade.”

The report found that about 1.8 million dead people were registered to vote at the time, and about 2.75 million people were registered in more than one state. These findings are obviously troubling.

But the it shows the Trump team is also conflating two different problems: The fact that voter rolls may need to be updated doesn’t mean there’s widespread, election-altering fraud. And just because the families of dead people haven’t alerted authorities that their names should be removed from voter rolls doesn’t mean any illegal voting happened. And just because people are registered to vote in two states at the same time––which often happens when people move from one state to another, register in their new home state, and forget to de-register where they used to live—doesn’t mean all 2.75 million of those people surreptitiously crossed state lines on Election Day to vote twice against Donald Trump.

That’s silly.

Even the report’s own authors agree. David Becker, the primary author, tweeted on Monday afternoon that it isn’t reasonable to use its findings to suggest voter fraud is a major problem.

“We found millions of out of date registration records due to people moving or dying, but found no evidence that voter fraud resulted,” he wrote.

Trump has taken significant heat from fact-checkers for suggesting otherwise. FactCheck.org called Trump’s Sunday assertions “bogus.” PolitiFact gave him a “Pants On Fire” rating for claims he made of “large-scale voter fraud” in the final weeks of the presidential campaign—claims based in part on his misinterpretation of the Pew study. And a report from CNN’s “Reality Check Team” noted that the report proffered zero evidence of people voting multiple times.

Perhaps Trump has some additional evidence up his sleeve that would suggest his loss in California was due to insidious shenanigans. But reporters haven’t been able to ask him about it. The mogul—who once lambasted Hillary Clinton for avoiding reporters—hasn’t had a press conference in exactly four months.