An interesting explainer from Steve Kornacki. Watch it and then we’ll do some math to see how daunting the task is for Trump. Note that Kornacki’s not saying that he expects a Trump comeback, he’s saying that so long as a Trump comeback is theoretically possible the networks will want to hold off on calling Pennsylvania for Biden. After all, this isn’t any ol’ state. This is the big one, the one that makes Biden the next president even if all of the remaining uncalled states break the other way. Gotta be sure before you call PA. (Reminder: Decision Desk HQ says it’s sure.) Watch, then read on.

WATCH: @SteveKornacki details the outstanding ballots that remain to be counted in Pennsylvania.#TrackingKornacki #MSNBC2020 pic.twitter.com/epjmpGxRLh

— MSNBC (@MSNBC) November 6, 2020

Biden has a lead of about 13,000 votes per the count on the screen behind Kornacki. There are about 100,000 mail ballots remaining and mail ballots lean heavily blue to the tune of 75/25 or so. That’s a net of 50,000 votes for Biden, bringing his lead up to 63,000 votes. There are also 100,000 provisional ballots outstanding and those ballots don’t lean heavily blue, per Kornacki. Provisional ballots from red counties are red. Trump could cut into Biden’s lead via provisionals, potentially quite a lot depending on how they lean. How red-leaning would they need to be to erase Biden’s 63,000-vote advantage?

Answer: 82/18 or thereabouts. Trump would have to utterly crush it among provisionals, even beyond the Democratic partisan lean in mail ballots to make up the gap.

*But* note what Kornacki says about some mail ballots potentially being thrown out due to procedural irregularities, like failing to enclose a ballot in an inner “secrecy” envelope before enclosing in an outer mailing envelope. Let’s say, hypothetically, 10,000 votes for Biden end up in the trash for those reasons. What margin does Trump need among provisional ballots in that case to erase a 53,000-vote deficit?

Answer: 77/23. Not quite as steep as in the first hypothetical but … really steep. Which, I assume, is why DDHQ called the race. Higher-profile media outlets realize that Trump is highly, highly unlikely to win provisionals at that clip but none of them want to get the all-important call on Pennsylvania wrong. You’ve gotta be absolutely sure Trump is out of gas before you declare PA, and thus the entire election, over.

And if you watch the clip closely, you’ll realize that Kornacki never claims that provisional ballots are trending overwhelmingly red. He says the ones from red counties are trending red. There may be *some* tilt towards Trump among the pool of 100K but he needs a 80/20 tilt or something of that order. That’s why Nate Silver is skeptical that provisionals will save him:

So, I am open-minded but not super persuaded by this. There are a handful of counties to have counted provisional ballots so far and those ballots indeed went for Trump, but they came from counties where the rest of the vote was *even stronger* for Trump.https://t.co/DXMdQJyfS5 https://t.co/h3gyCwCeNK

— Nate Silver (@NateSilver538) November 6, 2020

Also, about 40K of the provisionals are in Philly County, which is a considerably higher proportion than Philly's share of the vote overall.https://t.co/Jx5M55Y4gz

— Nate Silver (@NateSilver538) November 6, 2020

Note his point in the first tweet. In red counties thus far, provisionals are actually slightly bluer relative to the Election Day vote. That’s not the stuff of which 80/20 margins are made. And logically, it seems you’d have a fair number of Democrats using provisional ballots this year for the simple reason that mail ballots were overwhelmingly favored by Dems. Shouldn’t that mean a hefty number of Biden votes among the remaining provisionals? Seems that way:

Furthermore, the universe of requested-but-unreturned mail ballots, which is where most of these provisionals originated from, was D +17 by party registration. Not as D as completed mail ballots, but obviously still quite D.https://t.co/sTPHw4KZsl

— Nate Silver (@NateSilver538) November 6, 2020

Trump’s big-picture problem right now is that he’s behind in multiple states, not just Pennsylvania, which means he’s playing whack-a-mole in hoping for a comeback. As soon as he pins his hopes on one state, a new batch of blue-leaning mail ballots in another trickle in and make him a longer shot to win there too. Obviously a big burst of Trump votes among those that are left in PA would be exciting, but PA doesn’t matter if Arizona and Nevada go blue — and they’re on track to do so, as noted earlier. Other analysts are feeling more confident in Arizona this afternoon as well:

Seeing no path to victory for Donald Trump, two prominent Phoenix polling firms said Friday they expect Democrat Joe Biden to win Arizona.https://t.co/8LJA5UVozw

— KTAR News 92.3 (@KTAR923) November 6, 2020

Biden could end up losing Pennsylvania and Georgia and still (barely) get to 270 with AZ and NV. That’s what winning Michigan and Wisconsin did for him.

Exit quotation from Byron York, lamenting the problem Trump created for himself by steering Republican voters away from mail ballots: “After all this is over and if the president loses, he and certainly a lot of Republicans, are going to regret their strategy of denigrating mail-in voting so strongly… [H]e fostered distrust in mail-in voting and urged his voters to have a big red wave on Election Day. Meanwhile, Democrats, day after day after day, had bagged hundreds of thousands and then millions of votes already in the bag by the time Election Day came around.”

Update: Via Silver, this local election analyst expects the remaining provisional ballots to trend somewhat blue, not red, which would foreclose any path for Trump.

Lara Putnam, a University of Pittsburgh professor who studies the electoral landscape in Western Pennsylvania, said she expects the provisional ballots to “less Biden-heavy than mail-ins, but still net Biden more votes than Trump.”

— Julian Routh (@julianrouth) November 6, 2020

Some Democrats on social media are getting restless with the networks for not calling the race yet given how unlikely a Trump comeback is now, and fearing that every minute of uncertainty is making it easier for Trump to claim that something untoward is going on.