Nate Silver

Nate Silver @NateSilver538

Deaths are down again week-over-week, but I'm not sure if it's appropriate to compare a Monday coming out of a holiday weekend to a regular Monday. Possible we'll see a lot of deaths reported tomorrow and Wednesday. Plus, the high positive test rate today (9%) is not good news. — PolitiTweet.org

Nate Silver @NateSilver538

US numbers via @COVID19Tracking: ⚠️ CAUTION—HOLIDAY WEEKEND ⚠️ Newly reported deaths Today: 242 Yesterday: 209 One week ago (6/29): 330 Newly reported cases T: 47K Y: 43K 6/29: 36K Newly reported tests T: 518K Y: 656K 6/29: 569K Positive test rate T: 9.1% Y: 6.5% 6/29: 6.4% — PolitiTweet.org

Nate Silver @NateSilver538

@chrislhayes @cwarzel @TPCarney I think it is foolish to make too many short-term predictions, though. The best models project that deaths are more likely than not to begin rising modestly, but with a wide range of uncertainty. https://t.co/Qxu8rSPoV1 https://t.co/E1gcBFXGAV — PolitiTweet.org

Nate Silver @NateSilver538

@chrislhayes @cwarzel @TPCarney There *wasn't* actually a long lag in March/April because we were doing so little testing that the first time we knew when someone had COVID was sometimes when they showed up in the hospital (or dead). Presumably a longer lag *now* given there's more testing = earlier detection. — PolitiTweet.org

Nate Silver @NateSilver538

Please don't give me any "Trump can exploit 'cancel culture' going too far" takes. Maybe another Republican could. But Trump is not trusted *at all* on these issues. And he's not really even trying to be strategic about them; he operates from his impulses. https://t.co/GvTKt9jKqp https://t.co/rnBNvLh4eU — PolitiTweet.org

Nate Silver @NateSilver538

RT @ameliatd: The Supreme Court just handed down a *unanimous* decision on faithless electors. They rule that laws that take action against… — PolitiTweet.org

Nate Silver @NateSilver538

RT @JohnDelaney: Overdoses nationally jumped 18 percent in March, 29 percent in April and 42 percent in May - probably similar stats exist… — PolitiTweet.org

Nate Silver @NateSilver538

RT @nursekelsey: If you want to know how we get people to comply with wearing face masks, ask a sexologist. It’s not our first rodeo when i… — PolitiTweet.org

Nate Silver @NateSilver538

The positive test rate did come down. But again, the data is weird on holiday weekends and I think it would be more prudent to take a deep breath and wait 'til mid/late next week before really making very much of that. Hope everyone had a great 4th. — PolitiTweet.org

Nate Silver @NateSilver538

209 is the fewest deaths reported in a day since 3/23. But Sundays are slow, and a Sunday on a holiday weekend is perhaps likely to be especially slow. Some states didn't report data at all. So don't get too excited (or too freaked out by backlogs once the weekend is over). — PolitiTweet.org

Nate Silver @NateSilver538

US numbers via @COVID19Tracking ⚠️ Impacted by holiday weekend! ⚠️ Newly reported deaths Today: 209 Yesterday: 306 One week ago (6/28): 271 Newly reported cases T: 43K Y: 52K 6/28: 42K Newly reported tests T: 654K Y: 645K 6/28: 586K Positive test %: T: 6.5% Y: 8.1% 6/28: 7.2% — PolitiTweet.org

Nate Silver @NateSilver538

@mattyglesias Still not too late! Democratic delegates aren't technically legally bound and are free to vote their conscience. — PolitiTweet.org

Nate Silver @NateSilver538

The "Don't dismiss Kanye West because POLITICS IS CRAZY NOW!" people are pretty much the same people who were like "Don't dismiss Marianne Williamson because POLITICS IS CRAZY NOW!" — PolitiTweet.org

Nate Silver @NateSilver538

RT @ABCPolitics: Masks have proven to be a contentious issue among Americans, but are masks ultimately a smart economic and political polic… — PolitiTweet.org

Nate Silver @NateSilver538

RT @shearm: @realDonaldTrump's idea for a "statuary park" is not original. I wrote about one of the failed efforts to build one near Willia… — PolitiTweet.org

Nate Silver @NateSilver538

* Saturday reports mostly reflect data collection on Fridays. And this Friday was a holiday whereas last Friday was not. Death reporting is generally slower on weekends/holidays. Thus, the sharp decline in deaths vs. last Saturday is likely a reflection of that in large part. — PolitiTweet.org

Nate Silver @NateSilver538

US daily numbers via @COVID19Tracking BE CAREFUL WITH THIS ONE* Newly reported deaths Today: 306 Yesterday: 635 One week ago (6/27): 509 Newly reported cases T: 52K Y: 58K 6/27: 43K Newly reported tests T: 645K Y: 721K 6/27: 591K Positive test rate T: 8.1% Y: 8.0% 6/27: 7.4% — PolitiTweet.org

Nate Silver @NateSilver538

So, we're no longer in the period where we can regularly expect NY's numbers to show a decline, which itself is somewhat concerning. However, as to whether the numbers have begun to increase, I don't know about that. For now, a better case for increases upstate than downstate. — PolitiTweet.org

Nate Silver @NateSilver538

There's also hospitalization numbers, which have generally continued to decline, and NYC keeps its own data, which has generally looked pretty favorable. https://t.co/96duTCCKmV https://t.co/FzTtPAVIGR — PolitiTweet.org

Nate Silver @NateSilver538

NYC SUBURBS (Suffolk, Nassau, Westchester, Rockland, Putnam Cos.) Week ended 6/12: 156 cases/day, 1.1% + test rate 6/19: 128, 0.9% 6/26: 140, 1.0% 7/3: 135, 1.0% UPSTATE (everything except NYC metro) 6/12: 177, 0.8% 6/19: 168, 0.8% 6/26: 186, 1.0% 7/3: 205, 1.1% — PolitiTweet.org

Nate Silver @NateSilver538

NEW YORK CITY Week ended 6/12: 426 cases/day, 1.7% + test rate 6/19: 367, 1.2% 6/26: 339, 1.2% 7/3: 328, 1.2% — PolitiTweet.org

Nate Silver @NateSilver538

I've seen some speculation about whether cases in New York are rising, falling or steady. I don't think it's entirely clear, but FWIW here's a breakdown by region of the state. Specifically, the number of cases and the positive test rate over the past 4 weeks... — PolitiTweet.org

Nate Silver @NateSilver538

As a warning: with the holiday weekend, the data could be weird over the next several days, so be careful about jumping to hasty conclusions. — PolitiTweet.org

Nate Silver @NateSilver538

Deaths remain roughly at a plateau over the past several days. The NYT has a good article out on why the death count isn't increasing (though it may begin to rise soon) even as cases spike. https://t.co/L4Fi5HWQ0T — PolitiTweet.org

Nate Silver @NateSilver538

Cases are still rising, with yet another record set today. And although today was also a record testing day, the positive test rate (8%) is considerably higher than it was a few weeks ago. — PolitiTweet.org

Nate Silver @NateSilver538

US daily numbers via @COVID19Tracking: Newly reported deaths Today: 635 Yesterday: 670 One week ago (6/26): 636 Newly reported cases T: 58K* Y: 54K 6/26: 44K Newly reported tests T: 721K* Y: 665K 6/26: 603K Positive test rate T: 8.0% Y: 8.1% 6/26: 7.4% * new record — PolitiTweet.org

Nate Silver @NateSilver538

RT @NateDuncanNBA: Dunc'd On: How to Evaluate NBA “Sources” with 538’s @perrybaconjr; Schedule Analysis with @DannyLeroux Subscribe: http… — PolitiTweet.org

Nate Silver @NateSilver538

In 1880-1944, you have bigger and more varied economic shocks that are probably more comparable to what's happening today than anything (except perhaps 2007/08) in the postwar era. You also have higher levels of polarization vs. the mid/late 20c. So it's useful data for a model. — PolitiTweet.org

Nate Silver @NateSilver538

Overall, we find that since 1880, the economy explains about 30% of variation in the incumbent party's vote share if you adjust for polarization (26% if you don't). We believe this is a more realistic estimate than other models that are fit (often overfit) to more recent data. — PolitiTweet.org

Nate Silver @NateSilver538

One other dorky note. We've actually calculated an economic index for elections dating back to 1880. The figure expressed in the table below is basically a z-score combining several different variables. Estimates before 1948 are crude: https://t.co/LeIoYN5YjE — PolitiTweet.org