Nate Silver

Nate Silver @NateSilver538

@Crimealytics Where is that data from? — PolitiTweet.org

Nate Silver @NateSilver538

@SeanTrende But also: 4) People who are bought into a narrative of American institutions failing. The "tell" that someone is in category 4 is that that they tend to emphasize how the US is doing uniquely poorly compared to, say, Europe, when that just objectively isn't true. — PolitiTweet.org

Nate Silver @NateSilver538

@SeanTrende I think there are kind of 4 camps: 1) People with poor news consumption habits, e.g. too much cable, which tends to emphasize bad news and isn't very numerate. 2) People who don't want to get others' hopes up. 3) People for whom pessimism is a coping mechanism. — PolitiTweet.org

Nate Silver @NateSilver538

Choose your quarantine house. https://t.co/e6bD8hgJPx — PolitiTweet.org

Nate Silver @NateSilver538

@SethS_D I'm not necessarily sure that's true. Some states and countries (e.g. France) are going back to try to count deaths in nursing homes or otherwise outside of hospitals that they might have missed. https://t.co/gIJwtszOYk — PolitiTweet.org

Nate Silver @NateSilver538

Newly-reported deaths, on the other hand, probably lag infection by more like 4 weeks, on average. Go back 4 weeks, and there are few, if any, social distancing measures in place. The NBA is still playing. Restaurants are busy. People are mostly going into work. — PolitiTweet.org

Nate Silver @NateSilver538

All of this is in line with what you might expect. Newly-reported cases lag infection by something like 2-3 weeks. Go back 2-3 weeks, and that's when a lot of stay-at-home orders went into effect. You are really starting to see their impact. The new-cases curve is now quite flat. — PolitiTweet.org

Nate Silver @NateSilver538

The number of new cases looks pretty decent, though: 31,263. That's higher than Monday (29,023) or Sunday (26,553) but lower than Friday (33,767) and Thursday (32,889). 137K new tests reported, which is about average for recent days. — PolitiTweet.org

Nate Silver @NateSilver538

After surprisingly low death tolls the past two days—which I was worried were partly caused by day-of-week effects (reporting can be slower coming out of weekends)— today was a bad one: 1,941 new deaths reported in the US, the highest yet by some margin, surpassing 1,352 on Sat. — PolitiTweet.org

Nate Silver @NateSilver538

In the halfway-to-normal stage, these proposals are potentially more viable and prudent. That's not to say you can't find fault with them. But a lot of people, institutions, governments and businesses probably ought to be thinking about what the halfway stage looks like. — PolitiTweet.org

Nate Silver @NateSilver538

But we're not going to be facing the current situation forever. Eventually, these restrictions are going to be lifted. When? I dunno. But almost certainly it will come in stages. So there will be an interim period, perhaps lasting months, where things are halfway back to normal. — PolitiTweet.org

Nate Silver @NateSilver538

One thing re: sports getting re-started. People seem to be interpreting proposals by the various leagues in light of the current situation, i.e. where there are rather strict lockdowns almost everywhere. In the current situation, these proposals are dumb and/or far-fetched IMO. — PolitiTweet.org

Nate Silver @NateSilver538

RT @maggiekb1: "...the thing I’m most concerned about is the emergence of a new virus that is highly transmissible from person to person, a… — PolitiTweet.org

Nate Silver @NateSilver538

Also, things are still getting worse in the UK, while improving a bit in Italy and Spain. The IHME model projects the UK to eventually have the most deaths in Europe (66K) by some margin, as compared to 20K in the next-most-affected country, Italy. https://t.co/I7c7v7mvps — PolitiTweet.org

Nate Silver @NateSilver538

@CarlBialik Yeah, it's deaths attributed to COVID-19, so I can imagine a lot of circumstances where the attribution takes time even after you know that someone has died. — PolitiTweet.org

Nate Silver @NateSilver538

Highest case fatality rates in the world for countries with 1+ M people. Note: this generally reflects places where the situation is/was quite bad *and* which are doing a poor job of testing. Italy: 12.6% UK: 11.1% Netherlands: 10.7% Spain: 9.9% France: 9.5% Belgium: 9.2% — PolitiTweet.org

Nate Silver @NateSilver538

RT @FiveThirtyEight: Trump has touted hydroxychloroquine as a way to fight COVID-19. But scientists don’t know the risks of giving this d… — PolitiTweet.org

Nate Silver @NateSilver538

Yeah I don't think people realize how stochastic some of this is. Like, literally 1 or 2 super-spreaders arriving in the wrong place at the wrong time (say mid Feb.) could double the number of cases that a state/metro area ultimately gets. — PolitiTweet.org

Nate Silver @NateSilver538

"Pandemic risk was underpriced in Europe and N. America (though not in Asia) and will now be approximately properly priced" is a reasonable hypothesis, but the societal changes that flow from that are probably an order of magnitude less sweeping than these polemics would assume. — PolitiTweet.org

Nate Silver @NateSilver538

I'm especially perplexed by the "NEW PARADIGM!!" stuff when it comes from people who think markets (in a broad sense) are reasonably efficient because it's not at all obvious why the equilibrium 12-18+ months (post-vaccine) from today should greatly differ from the one today. — PolitiTweet.org

Nate Silver @NateSilver538

@mattyglesias In between Q and S. — PolitiTweet.org

Nate Silver @NateSilver538

In general, people badly overestimate trends and underestimate mean-reversion. So you should expect that people *really* over-extrapolate from "things that have changed for a few weeks because of mass social distancing" to "OMG A BRAND NEW PARADIGM!!". — PolitiTweet.org

Nate Silver @NateSilver538

That's on 34K newly-reported tests which is a pretty high number for Italy. Less than 10% of newly-reported tests came up positive in other words. — PolitiTweet.org

Nate Silver @NateSilver538

No Tuesday hangover in Italy either. 3,039 newly-reported cases, which looks like their lowest number since March 13. https://t.co/WqQjcpwvnB — PolitiTweet.org

Nate Silver @NateSilver538

Maybe that anecdotal evidence doesn't mean much. But it did affect my priors a bit since the data on this is also a bit dubious, often failing to account for confounders like population density and sometimes overemphasizing relatively small differences from place to place. — PolitiTweet.org

Nate Silver @NateSilver538

The context: my partner and I needed to go to Kansas City to see family ~4 weeks ago. As the COVID-19 situation was getting worse, we wound up staying there for a bit. Drove back to NY this weekend. There was lots of social distancing in KC + the red states we drove thru. — PolitiTweet.org

Nate Silver @NateSilver538

Shared the context for this on the podcast yesterday. But for what it's worth, here's a photo of a gas station near Gassaway, West Virginia (pop. 908) this weekend, where they were quite conscientious about social distancing. https://t.co/padjqKvfOk https://t.co/m5voYtg2k4 — PolitiTweet.org

Nate Silver @NateSilver538

The bad news, as Cuomo said, is that NY recorded its highly daily death toll, 731, vs. an average of ~600 for the previous four days. As he also said, deaths are likely to lag behind other indicators that show the slope bending. https://t.co/T7WuZp1xGr — PolitiTweet.org

Nate Silver @NateSilver538

In many places, Tuesdays are bad since you're catching up from slow reporting on the weekend. So that's an encouraging number. Meanwhile, net hospitalizations are up (+656) from the weekend, but much slower than last week (~1,200 per day) and lowest ICU number (89) in ~2 weeks. — PolitiTweet.org

Nate Silver @NateSilver538

In many states and countries, Tuesdays are often pretty bad days since you're catching up from slow reporting on the weekend. So that's an encouraging number. Meanwhile, net hospitalizations are up, but only modestly, and daily ICU admissions are down. https://t.co/S61a5eFwN2 — PolitiTweet.org