Nate Silver

Nate Silver @NateSilver538

@thehowie @conorsen I didn't make this index and wasn't aware of how it was constructed but yeah, a population-weighted index would be a lot better! I'm not sure I buy though that Sweden is stricter than the US overall (population-weighted), especially given the lack of mask-wearing there. — PolitiTweet.org

Nate Silver @NateSilver538

@thehowie @conorsen I'm sure it's not a perfect metric, but the Oxford index says Sweden is quite a bit less strict than the US overall FWIW. https://t.co/Opa0xE93Qo https://t.co/PW7ZbXvpWG — PolitiTweet.org

Nate Silver @NateSilver538

(Sorry, Nebraska 2 should be above the tipping point, not below it, but you get the point.) — PolitiTweet.org

Nate Silver @NateSilver538

Overall, this phase of the campaign encompassing the conventions and their aftermath has not been very successful for Trump. He's gained a bit in national polls and quite a bit in Florida, but Biden's lead in most other swing states has been steady or increasing. https://t.co/fFswai9SjY — PolitiTweet.org

Nate Silver @NateSilver538

One question is whether Georgia could conceivably be bluer than Florida and North Carolina, either now or in the near future. We have them pretty tightly bunched together now, with Trump +0.2 in GA, Biden +1.0 in NC and Biden +1.8 in FL. https://t.co/VYrvEGxHXu https://t.co/mgPbqeWa3w — PolitiTweet.org

Nate Silver @NateSilver538

@mattyglesias @conorsen I think different people in Sweden probably thought different things about what their strategy actually was. — PolitiTweet.org

Nate Silver @NateSilver538

@conorsen Yeah, I mostly agree. What seems pretty clear across a wide range of countries is that it's hard to sustain support for lockdowns if cases are low and/or declining. — PolitiTweet.org

Nate Silver @NateSilver538

@conorsen Yeah, although you could probably substitute "Western Hemisphere" for "US" in that sentence. — PolitiTweet.org

Nate Silver @NateSilver538

RT @markos: Here you go: https://t.co/WLG3VToT0l https://t.co/DVQnhV9wg7 — PolitiTweet.org

Nate Silver @NateSilver538

Going to confess to being a little bit annoyed that we had all these high-quality pollsters sweep through Georgia and almost nobody bothered to poll the potential runoff matchups. Would be very curious about Loeffler-Warnock and Collins-Warnock head-to-heads. — PolitiTweet.org

Nate Silver @NateSilver538

tbh I think the more interesting result here is in the GA-Senate special, where it's clear that Lieberman's support has collapsed and Warnock is likely to advance to the runoff against one of the Republicans. They also show Ossoff-Perdue very tight. https://t.co/HEBDAFv4Wp — PolitiTweet.org

Nate Silver @NateSilver538

Quinnipiac has Biden +3 in Georgia. It's their first poll of the state, but it's worth keeping in mind that they've generally shown strong numbers for Biden in other states. https://t.co/XLjXy9a4Ft — PolitiTweet.org

Nate Silver @NateSilver538

RT @laurabronner: The candidates' net favorability ratings are somewhat uneven: Biden is at +3, while Trump is at -23. While their "very… — PolitiTweet.org

Nate Silver @NateSilver538

RT @laurabronner: However, there are more who think both candidates are about equally likely to win. Our panel poll with @Ipsos, with @ele… — PolitiTweet.org

Nate Silver @NateSilver538

RT @laurabronner: Because we like probabilities, rather than asking who people were going to vote for, we asked *how likely* they were to v… — PolitiTweet.org

Nate Silver @NateSilver538

I can't believe I'm tweeting about a North Dakota poll, but Biden losing it by "only" 19 when Clinton lost it by 36 is more evidence that his biggest gains relative to Clinton are with white northerners. — PolitiTweet.org

Nate Silver @NateSilver538

RT @tressiemcphd: I recently diatribed on the difference between journalistic storytelling and sociological storytelling and this is case i… — PolitiTweet.org

Nate Silver @NateSilver538

RT @FiveThirtyEight: TBD. https://t.co/WL5P4lmDV3 — PolitiTweet.org

Nate Silver @NateSilver538

Aside from the debates, the issues that are likely to dominate the final month of the campaign (see below) are not great for Trump. He probably needs to raise voter doubts about Biden and hope for a lesser-of-two-evils election like 2016. https://t.co/yTCyaSF1OB — PolitiTweet.org

Nate Silver @NateSilver538

Of that 20-25% of the time that Trump comes back (I can't believe I have to say this but note: our probabilities don't account for the chance that the election would be stolen) I'd think the debates are a pretty important part of the story. They may be his best chance. — PolitiTweet.org

Nate Silver @NateSilver538

Trump still has decent comeback chances (20-25%) but some of the recent chatter about the election coming down to the Supreme Court seems to imply that we're headed for a photo finish, which is not (necessarily) true. Biden has a solid lead right now. https://t.co/ajG88SznSA https://t.co/E6C8crnlDr — PolitiTweet.org

Nate Silver @NateSilver538

It's not a huge hedge, but it matters *some*. If e.g. Biden gains 4 points in our Florida polling average after the debate, the model would treat that as more like a 3 point gain for forecasting purposes. Biden would get full credit if he were able to sustain it after ~2 weeks. — PolitiTweet.org

Nate Silver @NateSilver538

Something to keep in mind: although debates can produce polling swings, those swings are sometimes temporary. In fact, our *forecast* will hedge against changes in our *polling averages*, which will dampen shifts in the candidates' win probabilities a bit. https://t.co/nVtlpQscbB https://t.co/Ad2VdFTWhP — PolitiTweet.org

Nate Silver @NateSilver538

We're doing pre/post-debate polls with @Ipsos this year! Here is the pre-debate version. We'll follow up with this same group of voters after the debate. — PolitiTweet.org

Nate Silver @NateSilver538

In fact, "mediocre and uneven—though not poor" is probably how you can describe both Trump and Biden's debate performances **as they were judged by voters in real time**. — PolitiTweet.org

Nate Silver @NateSilver538

And the primary debates? Generally Trump did very well on the unscripted days where he completely dominated coverage. The debates were a chance for his rivals to even the score and often resulted in Trump losing a point or two in polls. He was mediocre and uneven—though not poor. — PolitiTweet.org

Nate Silver @NateSilver538

Clinton actually led by only 1.4 points in national polls entering the first debate, eventually got that lead back up to 6 points before it started falling again. There were other things going on in that period ("Access Hollywood") but still.... https://t.co/sJBHxVdNSS https://t.co/DjKR2HRjuk — PolitiTweet.org

Nate Silver @NateSilver538

Worth reading but it's worth remembering that Trump was judged the loser of the debates by polls and actually lost quite a bit of ground to Clinton in head-to-head polls following the debates. — PolitiTweet.org

Nate Silver @NateSilver538

Our Pennsylvania polling average now shows Biden +5.5, which really reduces the split with Wisconsin (B +6.8) and Michigan (B +6.9) as well as with national polls (B + 7.0). https://t.co/wKbW02AGO5 https://t.co/9zacNqYBdU — PolitiTweet.org

Nate Silver @NateSilver538

...there were questions before as to why Pennsylvania was polling as a fairly tight race even though Biden was polling well in demographically similar states such as MI/MN/WI and also even leading in some OH polls recently. Now PA looks more in line with the rest of those states. — PolitiTweet.org