Nate Silver

Nate Silver @NateSilver538

A lot of high-information news consumers seem to think the evidence for immunity to COVID is iffy, but while there are uncertainties the actual news has been pretty good as research has evolved. https://t.co/Y8hntBLjUB — PolitiTweet.org

Nate Silver @NateSilver538

@jodyavirgan As a friend pointed out, it's also true that in some ways there isn't a lot of news right now in the rapid-fire ways people are used to. There's only one story, it's all terrible, and most developments in the story only shade it toward degrees of marginally more or less terrible. — PolitiTweet.org

Nate Silver @NateSilver538

@CarlBialik Yeah, my priors (especially in this context) are that shame works among the "in" group but can backfire among the "out" group, and tends to work well in the short run but less well in the long run. — PolitiTweet.org

Nate Silver @NateSilver538

Psychologically, I wonder if highlighting examples of noncompliance with social distancing⁠—e.g. a photo of people at a crowded beach/bar/etc. without masks on⁠—actually reduces that behavior vs. pointing out that the vast majority of people are doing a good job of distancing. — PolitiTweet.org

Nate Silver @NateSilver538

A lot of states are increasing testing and setting up contact-tracing programs. A lot of Americans are wearing masks. Consensus seems to be shifting toward allowing human challenge trials on COVID-19 vaccines. More understanding now that indoors is more dangerous than outdoors. — PolitiTweet.org

Nate Silver @NateSilver538

Incremental progress doesn't make for good headlines, especially when it's slower than it should be, but a lot of the things suggested in this article are happening to some degree or another. https://t.co/Shbos1xxK5 — PolitiTweet.org

Nate Silver @NateSilver538

RT @CohoKelly: @NateSilver538 @COVID19Tracking 7day rolling averages for all metrics: Deaths Today: 1.4k 1 wk ago: 1.8k 2 wk ago: 1.8k… — PolitiTweet.org

Nate Silver @NateSilver538

This is a fairly typical report for recent days. Death tallies are still quite high, though we aren't seeing many 2K+ days lately so they have come down a bit. It's now common to see days with >300,000 tests and positive test rates <10%. — PolitiTweet.org

Nate Silver @NateSilver538

US daily numbers via @COVID19Tracking: Newly reported deaths Today: 1,741 Yesterday: 1,693 One week ago (5/7): 2,746 Newly reported cases T: 25K Y: 20K 5/7 28K Newly reported tests T: 367K Y: 324K 5/7: 304K Positive test rate T: 7% Y: 6% 5/7: 9% — PolitiTweet.org

Nate Silver @NateSilver538

Clinton advisers warn that Biden spending too much time in Wisconsin. (In all seriousness though I think the organizing part of Bernie's campaign is/was impressive. Strategy more lacking.) — PolitiTweet.org

Nate Silver @NateSilver538

@foxjust This model estimates ~1.1% currently infected in Sweden and ~6.3% ever infected. So that lines up. But that's countrywide & rates in Stockholm are much higher. If Stockholm is 2.5-3x higher than the country, you could get in the 15-20% ever infected range. https://t.co/q3BwQFENme — PolitiTweet.org

Nate Silver @NateSilver538

RT @kylegriffin1: Breaking via LA Times: Federal agents seized a cellphone belonging to Republican Sen. Richard Burr as part of the Justice… — PolitiTweet.org

Nate Silver @NateSilver538

@mattyglesias @datastatsnerd 1. I don't know what to expect in a few weeks and there's definitely the chance that things get a lot worse. 2. I also think people's priors have not been super robustly-considered or evidence-driven. — PolitiTweet.org

Nate Silver @NateSilver538

@Neoavatara @thehowie Also, to what extent is people going out to officially-sanctioned businesses substituting for hanging out in private? e.g. not sure that meeting a friend at a restaurant (following stringent protocols) is riskier than going over to their place. It might be safer, in some ways. — PolitiTweet.org

Nate Silver @NateSilver538

@Neoavatara @thehowie I think that's partly because the re-openings are pretty incremental and people are being cautious. So you're not going from 0 to 60. You're going from 20 to 25 maybe. (And, also in the states that are still "closed", you're going from 20 to 23 because people are getting antsy.) — PolitiTweet.org

Nate Silver @NateSilver538

@chrislhayes @MattZeitlin Also, Europe doesn't seem to have had big increases since its partial, careful re-openings. I wish the conversation in the US were less about spikes and more along the lines of "need to get the total # of cases down first, because we may hit a *plateau* upon starting to reopen". — PolitiTweet.org

Nate Silver @NateSilver538

@Neoavatara But those are truly large-scale events. I think people overestimate how easily you're going to see the signature of, say, 100 people not wearing masks at a state-capital protest in the data. Or a crowded day at the beach where people aren't being quite cautious enough. — PolitiTweet.org

Nate Silver @NateSilver538

@Neoavatara The evidence on Mardi Gras being a superspreading event seems pretty credible from what I understand. Maybe some European football matches where a lot of people traveled. — PolitiTweet.org

Nate Silver @NateSilver538

Not sure about this one. MD/DC/VA is one of the areas where it's really not clear whether cases are going down even if you *do* account for testing. Maybe the part of the country there's most reason to worry about right now. https://t.co/tfVflXE6Gw — PolitiTweet.org

Nate Silver @NateSilver538

We've added @youyanggu's model, which has performed quite well so far, to our interactive showing different COVID-19 projections. https://t.co/ksZhgvgzso — PolitiTweet.org

Nate Silver @NateSilver538

States and counties need to be able to increase testing without risking terrible headlines about a surge in cases (that only happened because they increased testing, but the article/headline doesn't mention it). — PolitiTweet.org

Nate Silver @NateSilver538

This is the Orange County testing data. A huge surge in tests; they're now testing 2-4 times as many people per week as they were just a few weeks ago. https://t.co/RtOCKvaG3Z https://t.co/XUnuU506n4 — PolitiTweet.org

Nate Silver @NateSilver538

You know why there's maybe been a surge of cases in Orange County, California, as this story focuses on? Because they're doing a ton more testing than before. https://t.co/EheL1pKykS — PolitiTweet.org

Nate Silver @NateSilver538

@conorsen @COVID19Tracking Plausible. I wish we had better data on spread in concentrated, vulnerable communities vs. generic "community spread". — PolitiTweet.org

Nate Silver @NateSilver538

@conorsen @COVID19Tracking At the same time... mask wearing + weather + a bit more testing/tracing could plausibly outweigh losses from mildly reduced distancing. Also questions about heterogeneities in transmission rates. — PolitiTweet.org

Nate Silver @NateSilver538

@conorsen @COVID19Tracking We will see. One thing that's plausible (since there seems to be around a 2-week lag between what the data is actually showing and the media narrative) is that it begins to sink in that the numbers are getting better at the same time they start getting worse again. — PolitiTweet.org

Nate Silver @NateSilver538

@conorsen @COVID19Tracking I think that was a largely correct interpretation of the data 2-3 weeks ago. I think it's probably not correct now, though of course things could take a turn for the worse again. — PolitiTweet.org

Nate Silver @NateSilver538

Excluding New York state, cases are down 11% week over week and the positive test rate is down 29%. — PolitiTweet.org

Nate Silver @NateSilver538

The progress in the case numbers is a bit clearer. The raw number of cases is down 14% in the previous 7 days as compared to the 7 days before that. And that's despite more testing; the positive test rate is down 31% week over week. — PolitiTweet.org

Nate Silver @NateSilver538

Wednesdays, like Tuesdays, often involve high death totals based on states' reporting patterns. This is another bad one (1,726), but better than recent Wednesdays, which have averaged around 2,300 reported deaths each. — PolitiTweet.org