Nate Silver

Nate Silver @NateSilver538

@stevesingiser @SeanTrende I think any epidemiologist would say there are for sure 100 of thousands of unreported COVID infections in the US. In fact there could be millions. That's the good news. The bad news is that deaths are a lagging indicator, and deaths can also be underreported. — PolitiTweet.org

Nate Silver @NateSilver538

@PollsterPatrick @ASDem Even for a snapshot, forcing people to make a choice between 2 candidates often results in their being honest about how they feel, especially if the 2 candidates are both well-known (as they are in this case). — PolitiTweet.org

Nate Silver @NateSilver538

@ASDem @PollsterPatrick That's become a fairly consistent pattern. Opposite of 2016 when most of the (many) voters who disliked both Clinton and Trump went for Clinton. — PolitiTweet.org

Nate Silver @NateSilver538

@ASDem @PollsterPatrick I think it may be people who are acting graciously toward Trump in a time of national crisis, but have no intention of voting for him. A lot of Trump's approval bounce has come among D's and D-leaning indies since he was already maxed out among his base. — PolitiTweet.org

Nate Silver @NateSilver538

@ASDem @PollsterPatrick Topline numbers are underrated and underlying numbers are overrated. — PolitiTweet.org

Nate Silver @NateSilver538

But you know what? It doesn't matter. Sanders needs *landslide wins everywhere* to win the delegate race, and Wisconsin should be one of his best states. A big landslide loss vs. a small landslide loss vs. a narrow loss vs. even a narrow win doesn't change things much. — PolitiTweet.org

Nate Silver @NateSilver538

This is a very good pollster but we should wait for more confirmation that Biden's lead is really that large. Our polling average (mostly based on older data) has it at 15 points instead. https://t.co/9n7ri67e8Q — PolitiTweet.org

Nate Silver @NateSilver538

ThE DeMoCrAtiC PriMaRy Is OvEr — PolitiTweet.org

Nate Silver @NateSilver538

To some extent there's a fine line between underreporting problems that all countries are having, deliberate underreporting (if you stop testing people, the numbers will fall!) and outright fabrication/manipulation of data. — PolitiTweet.org

Nate Silver @NateSilver538

I have no doubt that China's coronavirus numbers are fishy, but I find myself a little bit perturbed so far in the lack of specificity in the reporting over what they did exactly; hopefully that will come later. — PolitiTweet.org

Nate Silver @NateSilver538

The best part is NY seems to have done a really admirable job of ramping up its hospital capacity; though as Cuomo says, the limiting factor may now be staff (which is harder to scale up) rather than beds. You also wonder if other cities will be able to scale up as quickly. — PolitiTweet.org

Nate Silver @NateSilver538

The worst part of this is that if NYC had acted even 2-4 days sooner, it would have made a *lot* of difference. — PolitiTweet.org

Nate Silver @NateSilver538

My impression, and you can sort of hear this in Cuomo's tone, is the New York numbers have really been in the middle between the more pessimistic scenarios and the relatively more optimistic (under the circumstances) scenarios. — PolitiTweet.org

Nate Silver @NateSilver538

For context, those death tolls are about **2-3x worse on a per capita basis** than the number reported in the United States yesterday, which was quite bad. — PolitiTweet.org

Nate Silver @NateSilver538

Three European countries that took a laissez-faire, "herd immunity' approach to coronavirus early on, the UK, the Netherlands and Sweden, are paying the price for it. They reported, respectively, 563, 134 and 59 deaths today. https://t.co/9fFIJJjqhz — PolitiTweet.org

Nate Silver @NateSilver538

@bruce_arthur I have talked to people who work on these kinds of projections in other cities. I also know how complicated this sort of modeling exercise is. Those projections come with huge amounts of uncertainty and the more policymakers understand that, the better job they're doing. — PolitiTweet.org

Nate Silver @NateSilver538

@bruce_arthur No that's the correct response. There are so many unknowns that it's a bad idea to lay out overly specific time frames beyond "weeks to months" and likely with some measures being relaxed sooner than others. — PolitiTweet.org

Nate Silver @NateSilver538

@skepticalsports @karlbykarlsmith Comparing the US to the EU seems reasonably appropriate and ... well, we'll have to see in a few months who did worse. One thing the US does seem to be doing right is that we're doing more testing in the worst-affected areas, whereas sort of the opposite is true for the EU. — PolitiTweet.org

Nate Silver @NateSilver538

@skepticalsports OK, but if we have 8x more metros of a certain population size relative to some midsize European country, that's also 8x more opportunities to have the some metro that has the worst outbreak. — PolitiTweet.org

Nate Silver @NateSilver538

@skepticalsports Wouldn't it say that the US is doing well handling the outbreak well in some areas and poorly in others? So, averaged across the entire US, it's doing mediocrely? — PolitiTweet.org

Nate Silver @NateSilver538

This is sort of like saying California is a redder state than Wyoming because it has 25x as many Trump voters. (It's true: CA had 4.5m Trump voters in 2016 vs. 175K for Wyoming.) — PolitiTweet.org

Nate Silver @NateSilver538

I don't have much patience for claims about which countries are doing better/worse on coronavirus that don't even bother to adjust for population size. — PolitiTweet.org

Nate Silver @NateSilver538

@skepticalsports Yep — PolitiTweet.org

Nate Silver @NateSilver538

As I mentioned yesterday, there do seem to be some day-of-week effects, where reporting on many metrics is a little slower on weekends and coming out of weekends (i.e. on Monday) and then you make up ground with worse numbers Tuesday-Thursday. — PolitiTweet.org

Nate Silver @NateSilver538

Not a very encouraging data on the data front. * Number of tests in the US has stagnated. * 807 deaths recorded today vs. 511 yesterday. * The rate of increase in detected new cases has slowed (partly because the number of tests has flattened), but they're still increasing. — PolitiTweet.org

Nate Silver @NateSilver538

FWIW the idea is to be evidence-driven, and the uncertainty in the outlook is greater as you go out further in time, so I don't know that I love cities cancelling things this far in advance. — PolitiTweet.org

Nate Silver @NateSilver538

Biased since it's from 538, but this is the best explainer you'll find on why building COVID-19 models is challenging. — PolitiTweet.org

Nate Silver @NateSilver538

RT @maggiekb1: I couldn't figure out why the CDC said to wash your hands for 20 seconds and the WHO said 45 seconds. So I called handwashin… — PolitiTweet.org

Nate Silver @NateSilver538

Please answer only if you currently live in the United States. Do you or someone you know personally have coronavirus? — PolitiTweet.org

Nate Silver @NateSilver538

The UK has tested only 135,000 people *total*. That's quite poor. The US is likely testing almost that people many *per day* now. (We have more people, but surpassed them on per-capita testing a few days ago.) — PolitiTweet.org