I think that's still a pretty good guess, though getting from Biden +75/80k to 100k depends on two kinds of hazy votes that may be worse (or better) for Biden than I've roughly penciled-in: the 10k late absentees and provisional ballots
Right now, Biden's up ~60k and there are another ~40k regular absentees. So getting into the 75-80k range before provisionals/post-election absentees is straightforward, even if we conservatively assume that a disproportionate share of remaining absentees will be rejected
From there, we've got
--10k late absentees--which I'd expect to be good (say 2-1) but maybe not amazing for Biden, since Dems disproportionately returned their ballots early.
-- ~70k uncounted provisionals, very disproportionately in Dem areas
The provisionals are the key to getting Biden up to 100k+, and they're not *amazing* for Biden. The prov results in Allegheny so far aren't as good for Biden as previously advertised by county officials. In general, provisionals are ~like the non-prov by county
That said, the remaining provisional ballots are disproportionately in Democratic areas and especially Philadelphia, so I'd expect Biden to pad his lead. At the moment, I think I'd bet against him getting to 100k out of it
• • •
Missing some Tweet in this thread? You can try to
force a refresh
A few random thoughts this morning on what Democrats should take from the election results, as the recriminations and so on begin
1) Democrats shouldn't blame themselves for *quite* so much!
This election was about Trump. He was the incumbent. It was a referendum on him, and everyone knew it. If the GOP did better, it's mainly because Trump was stronger than understood--not simply because Dems blew it
Or put differently: I'm not really sure there was much Democrats could do about what happened here. Trump completely dominated American life for the last four years, people knew what they thought of it, and it's hard to believe there were magic words to undo it
I've got a lot of questions about race calls in my 'interactions' today, so let's go through a quick update
First, we've uncalled NJ-07, which the AP called some time ago.
In general, we automatically accept AP calls (a few exceptions, like top presidential states--where we independently evaluate whether to accept). I think we've seen enough to say the NJ-07 call was premature
Next up is Arizona, where Trump has no realistic hope of mounting a comeback but where none of the news network decision desks have made a projection (excluding the erroneous AP/Fox call on Election Night). Why isn't this one called?
I have no doubt--based on great data in FL/NV, and fuzzier data elsewhere--that GOP beat our final turnout estimates, and perhaps by a lot
Whether that's only a modest or big part of the polling error is still an open question
As an aside, I do wonder whether Democrats will rethink whether mail/early voting is a great deal, given how badly they get clobbered on Election Day--the time when low propensity, election winning voters go to the polls.
What if 'GOTV efforts' are a lot less powerful in mobilizing turnout than, say, everyone around you going to vote. Election Day being a nonevent in your neighbrhood/community/peer group may be demobilizing
I haven't tweeted much about NC since Election Night, but with Cunningham conceding today let's take a look in at what's going on there
As far as I can tell, there are not nearly enough provisional and absentee ballots left for Ds to have a credible chance of making up their deficit.
Biden's down 75k votes; but there are 40k provisionals (won't all count) and maybe another 35k absentee ballots. Doesn't add up.
I do have pause about one thing, though, and I wonder whether it's a factor for any of the decision desks: some of the oddities in the results that I was tweeting about on Election Night
Let's look at the change in turnout *so far,* based on number of votes cast compared to 2016 turnout.
Let's start nationwide, at a high level: areas in purple have counted *fewer votes* than 2016. Most of those areas still just have lots of votes left to count
As you can see, there's a ton of vote left to count in Chicago, New York state, and parts of MD/CA. I'd guess these ballots generally break Democratic, padding Biden's national lead.
One thing you might have overlooked: Philly.
For all the complaints about the count there... it's still the only place in PA that's counted fewer ballots than 2016
Before we go into what went wrong, let's just call a spade a spade here: this was a bad polling error. It's comparable to 2016 in size, but pollsters don't have the excuses they did last time.
This year's polls would have been *way* worse than 2016 with a 2016 methodology
There are really two halves of polling: the quality of the sample you get, and the adjustments you take to improve the representativeness of your sample.
Since 2016, pollsters got better at the adjustments, but the underlying sample got worse