As it turns out, the ultimate expression of the infamous “take President Trump seriously not literally” mantra is a lawsuit filed on his behalf by Texas Attorney General Ken Paxton.

Paxton (R), who could use a favor from Trump, is suing Georgia, Michigan, Pennsylvania and Wisconsin on behalf of the residents of his state. His argument, in essence, is that the electoral systems in those states, which tabulated victories for President-elect Joe Biden, burdened Texans because [mumbles for a while] and therefore those results can’t be trusted. And, therefore, whaddya know, Trump probably gets a second term in office.

With Trump pressing Republicans to join Paxton’s fight, many have, often arguing that the claims should be taken seriously, if not literally. Sure, the result is throwing out the vote in a number of states, but we’re just asking questions! Is that so wrong?

Taking Paxton’s lawsuit literally, on the other hand, is nearly impossible, given how many dubious assertions it makes. We’ve been over this to some extent before, evaluating the arguments about alleged improprieties, claims for which there’s no credible evidence. Experts are generally of the opinion that Texas has no basis to sue, lacks the evidence necessary to sustain its claims, relies on faulty rhetoric and analysis and won’t have its lawsuit heard by the court. Otherwise, it’s swell.

Georgia, Michigan, Pennsylvania and Wisconsin on Dec. 10 urged the Supreme Court to reject a lawsuit filed by Texas to overturn the results of the election. (Reuters)

The states being sued similarly went over the lawsuit, poking more holes in it than a quarantined needlepointer. So Texas offered up a rebuttal, seeking to argue that its bad arguments, demonstrated as bad, are in fact good. In so doing, the state largely succeeds in making additional bad arguments.

Consider its claim that Wayne County, Mich., home of Detroit, had 174,384 suspicious absentee ballots, which, its original suit suggested, “likely resulted from the phenomenon of Wayne County election workers running the same ballots through a tabulator multiple times.” Michigan responded by saying that the rationale for those ballots being identified as suspicious, that they were “counted without a registration number,” doesn’t actually refer to any validation method the state uses. The state was “at a loss” for what Texas was talking about.

To which Texas replied by suggesting that Michigan being “at a loss” to explain what happened meant that it had no answer for its claims. It’s like me saying that you stole a glurblebarg and you replying that you have no idea what a glurblebarg is — and my then trumpeting the fact that you have “no idea” how to defend yourself against my charge. Not how it works.

As The Washington Post’s Dave Weigel points out, though, the idea that Wayne County whipped up nearly 175,000 suspicious votes is itself ridiculous.

Consider the vote margins in each county in the four states in both 2016 and 2020. If we compare the margins against one another we see that, almost uniformly, the margins in 2020 shifted in Biden’s direction relative to how Hillary Clinton matched up against Trump in 2016. (The circles below are scaled to the number of votes cast in each county.)

If we remove 175,000 votes from Wayne County’s total and assume those votes followed the same distribution as the county overall, its circle suddenly shifts dramatically to the right; that is, its margin in 2020 moves dramatically toward Trump. In so doing, it drops below the diagonal line, one of the only counties to do so.

Do we really think that the county surrounding Detroit not only shifted hard toward Trump but did so while nearly every other county in these four swing states went the other direction? There we go again, taking Texas’s claims literally.

That this allegation is centered on Detroit is intentional. In its new response, Texas summarizes its argument neatly, if perhaps not intentionally.

Were the Supreme Court not to address its spurious claims, the state argues, it would incentivize “further lawlessness and … drive honest voters from the polls: why should anyone vote if a few urban centers will manufacture an unlawful and insuperable vote margin?”

There you go. It’s those nasty urban Democrats who stole the election from Trump. Fie on them!

Again, though, this is nonsense. If we look at the results in 2016 and 2020 through the lens of urban density (using data from the MIT Election Lab, Edison Research and the Centers for Disease Control and Prevention), we can see how the vote in 2020 compared with 2016 and evaluate whether it was a shift in cities that made Biden successful where, four years ago, Clinton wasn’t.

In short, that’s not what happened. If we take all of the counties of each type in each state, tally the vote totals in 2016 and 2020 and compare the margin of victory for Trump or Biden relative to the 2016 outcome, we see that the biggest margin shifts in all four states were in suburban counties. We also see that, even in relatively rural counties, the outcome shifted to Biden, matching the scatter graph above.

We can look at this solely through the lens of votes cast, too. Below, the dotted outlines indicate the actual vote margin for Trump or Clinton in 2016, the solid circles the margin for Trump or Biden. In large cities, the margins for Biden were similar to those for Clinton. But in suburbs, Biden expanded his margins (see Georgia and Pennsylvania) and/or Trump’s margins narrowed (Michigan and Pennsylvania).

A lot more people voted in 2020 than in 2016 and, in less populated areas, Trump’s margins grew a bit. But in suburban areas, Biden’s grew a lot.

That’s why he won, not because of some imaginary manufacturing of votes in Philadelphia. The numbers are clear.

But, then, Texas argues that the numbers can’t be trusted. In its original lawsuit, it cited the analysis of a man named Charles J. Cicchetti. Cicchetti argued that the odds of Biden winning all four of these states were 1 in 1 quadrillion, a claim he made based on the assumption that there was no significant difference in the votes counted before 3 a.m. on Nov. 4 and those counted after. The implication was that such a late shift for Biden was nigh impossible.

This has been debunked six ways from Sunday. The votes counted later were obviously different from those counted earlier: They were absentee ballots that were only counted beginning on Election Day itself in most states (generally because Republican-led legislatures declined to allow counting to begin earlier). Since those ballots demonstrably favored Biden by a wide margin, suggesting that they should have trended the same way as the votes cast on the day of the election is, to put it gently, idiotic.

Texas’s response?

“Cicchetti did take into account the possibility that votes were not randomly drawn in the later time period,” its response reads, “but, as stated in his original Declaration, he is not aware of any data that would support such an assertion.”

Cicchetti himself weighed in stating that his “analysis did not attempt to speculate about why there was a difference” between early and late votes — “though I expressly noted possible unverified anecdotal explanations.”

Those unverified, anecdotal explanations?

“I am aware of anecdotal statements from election night that some Democrat strongholds were yet to be tabulated,” Cicchetti’s first analysis stated. “There was also some speculation that the yet-to-be counted ballots were likely absentee mail-in ballots.”

This is a bit like saying that it will be sunny forever because, while there are some anecdotes floating around about the sun “setting,” we should instead assume that the sun’s existing presence in the sky is its natural, default state.

What can one say to this besides, “Come on, man”? There’s literal video of Milwaukee County, Wis., officials submitting the county’s votes after the mail ballots there were counted — a report that couldn’t be made until the vote-counting was done. The late votes were unquestionably from more-populous counties with more mail ballots to count and unquestionably largely mail ballots. Shrugging at this as unknowable is simply embarrassing.

It is, however, nice to have Paxton on the record in response to questions about the lawsuit. If an initial assessment of the suit suggested that it was shaky, the new response makes its status clear: It has already collapsed.

Not literally, mind you. At least, not yet.