James Carville floated this theory last night on the same network.

“I think there is a better chance Donald Trump does not run for re-election, than he is reelected,” Carville told MSNBC’s “The 11th Hour,” hosted by Brian Williams…

“Zero chance he’s gonna be reelected,” Carville said again, suggesting that perhaps someone close to Trump should break the bad news to the president.

“Jared, or somebody, is gonna have to sit down and have a real, like, talk, a real man-to-man, a come-to-Jesus as people used to say, I don’t know what is it,” Carville said. “That is, that which can’t continue will not.”

That thought has occurred to me too in sillier moments for a simple reason. Losing a national popularity contest would be intolerable to someone like Trump, who’s narcissistic even by the standards of politicians. Getting blown out by a margin that he couldn’t somewhat plausibly blame on cheating would be a fate worse than death. Carville and Scarborough are coming from the same place, believing that POTUS would do anything to avert that outcome once he believed it was assured, up to and including declaring the political equivalent of bankruptcy and walking away in the home stretch of the campaign.

No presidential nominee wants to play out the string and take their lumps on election night when it looks like they’re headed for sure defeat but all other modern nominees have been conventional figures who’d feel duty-bound to their parties to follow through. I don’t think Trump feels partisan loyalty. He owns the GOP; it owes him loyalty, not vice versa. If he were to conclude circa Labor Day that the only way to avoid a humiliation is to toss McConnell the keys and walk away, why wouldn’t he?

Or so the theory goes.

But that theory doesn’t mesh with the facts. He could have done the same thing in mid-October 2016 after the “Access Hollywood” tape dropped and women began sharing stories with newspapers about sexual misconduct he allegedly engaged in. The fallout from all of that got so bad at the time that some Republican politicians withdrew their endorsements and called on him to drop out. Hillary’s polling briefly climbed. A blowout seemed possible, if not likely. If he was unwilling to endure the possibility of a drubbing, that would have been his moment to quit.

Instead he fought on. Why wouldn’t he do that in the current circumstances too?

Another thing. Although he has bad polling news coming at him from all sides right now, including his internal polling, it’s quite possible that he just doesn’t believe the numbers. Any of them. Tell him in August that he’s down 12 points and he might very well reply that it’s because his voters aren’t answering the phone when pollsters call and that they’ll be there for him en masse on Election Day. Some forecasters claimed in the final weeks of the 2016 campaign that Clinton had a 99 percent chance of victory and we know how that turned out. Trump’s heard the “you can’t win” song before. He’ll never believe it this time.

…even if his own allies suddenly sound pretty grim about the state of things. Here’s Ed Rollins, chairman of a pro-Trump Super PAC:

“The message is weak or nonexistent. The rambling on about Biden and Pelosi or Clinton and Obama is old and tired. This needs to be about the future, not the past,” Rollins said.

“He needs to show empathy, which he hasn’t, and project strength by doing what Reagan, Thatcher and Churchill did with strong speeches. Not macho bullshit, but thoughtful solutions to serious problems. This is about the future, not what the Democrats did in the past. Make people be concerned with what [Democrats] can do with total control of the government, House, Senate and the White House.”…

Rollins described the campaign as “adrift” and argued that it needs to be more nimble in ad-making, resource allocation and in recruiting senior GOP officials to act as prominent surrogates for the president in battleground states.

“Let’s be honest about it: The president is behind today,” said Karl Rove today on Fox, stressing that it’s important for Trump to start talking about what he’d do with another four years in office. News broke this afternoon that the campaign just bought its first TV ad in Georgia, evidence that they’re playing defense in states that Trump won four years ago and which shouldn’t be competitive if the incumbent were running strong. These latest polls from YouGov are a bit better for him than some of the other garish polling this week in that they show tight races in several swing states…

Here's what our YouGov polls currently say about the horse-race in the closest states:

OH: Biden -5
NC: -4
TX: -3
IA: -2
FL: 1
AZ: 2
PA: 2
GA: 6
WI: 8
MI: 9

National: +7

MOEs range from 5-10 points. Among likely voters.

Expected EVs IF THE ELECTION WERE HELD TODAY: pic.twitter.com/E2Djf9Ifib

— G. Elliott Morris (@gelliottmorris) June 26, 2020

…but there are major footnotes. If Biden held all the Hillary states from 2016 and won only the Wisconsin, Michigan, and Georgia this time — the three states where he leads by six or better in YouGov’s survey — that alone would be enough for him to win the presidency. If he won Arizona, Pennsylvania, and Florida too, where he’s barely ahead, he’d be over 330 electoral votes. The probability model at the bottom of Morris’s tweet gives Trump something like a three or four percent chance of winning right now; FiveThirtyEight’s final forecast in 2016 gave him a 29 percent chance of winning. He really is in a hole at the moment.

But I don’t think he’ll stay there. He’ll find an unlikely ally in the media in the months ahead. True, they hate his guts and badly want Biden to win, but they’re also suckers for “comeback” storylines. Reporters are writers first and foremost, and writers would much rather write a suspense story (that ends with their guy winning, of course) than a boring slog whose conclusion is foregone. Trump’s also sure to get some good news from time to time between now and the election that’ll boost his polling, with economic growth the likeliest source. And on some level I just don’t believe that the modern American electorate can produce a landslide, at least not one that’s stable for months in advance. (Obama’s 2008 landslide only materialized in the final weeks thanks to the financial crisis.) Trump will still be the underdog down the stretch this year but I think he’ll be grappling with a five-point deficit instead of a 10-point one.