Nate Silver

Nate Silver @NateSilver538

RT @laurabronner: The state of the race before the VP debate tomorrow is... almost exactly where we left off after last week's debate: -… — PolitiTweet.org

Nate Silver @NateSilver538

I'd also keep in mind that the while the initial diagnosis might have engendered some sympathy for Trump—along with some blame—subsequent developments (lack of transparency, the size of the WH cluster, mixed/poor/unpopular messaging) have been still less favorable for him. — PolitiTweet.org

Nate Silver @NateSilver538

This is another Biden +14, and the numbers are even worse for Trump in the part of the poll post his COVID diagnosis. I think we're starting to get more evidence now that COVID hurt him in addition to the debate. https://t.co/648Db5ziBC — PolitiTweet.org

Nate Silver @NateSilver538

RT @Nate_Cohn: In the Times/Siena national poll just a couple of weeks ago, voters overwhelmingly supported a new $2 trillion dollar stimul… — PolitiTweet.org

Nate Silver @NateSilver538

Wait, so Trump not only rejects stimulus funds that would probably have helped his re-election chances, but *also* does so in a way to make sure that he personally will take blame for it? — PolitiTweet.org

Nate Silver @NateSilver538

@barry I think he knows he's losing but he doesn't realize he's losing (at least in part) because of his strategic and policy choices and not in spite of them. — PolitiTweet.org

Nate Silver @NateSilver538

Also, the goal isn't solely to reduce COVID spread because physical isolation likely also has some very bad consequences. Any time we can reduce COVID spread without increasing isolation, or reduce isolation without increasing COVID spread, I would argue that is good. — PolitiTweet.org

Nate Silver @NateSilver538

A couple of implications: First, that's not a *false* sense of security. It's a *correct* sense of security and a response that indicates people are managing hard tradeoffs. I'd prefer to reserve the "false" label for "sanitization theater" interventions that have little effect. — PolitiTweet.org

Nate Silver @NateSilver538

Let's say masks reduce transmission by 60% other things held equal, but people respond by going out and being around people 30% more often. So instead of reducing COVID spread by 60%, you reduce it by about 45-50% instead. — PolitiTweet.org

Nate Silver @NateSilver538

Related to this, but I *hate* the "this will give people a false sense of security" argument, at least when applied to measures that *really are* effective (e.g. masking, testing) though not perfect. — PolitiTweet.org

Nate Silver @NateSilver538

There seem to be a few too many "the White House outbreak is proof that frequent testing alone can't stop COVID" takes given that it's neither clear how much testing the WH was actually doing nor how honest they've been about the results. — PolitiTweet.org

Nate Silver @NateSilver538

With Biden +5.8 in our Pennsylvania *forecast* (note: tighter than our polling average) but +6.1 in Wisconsin, we're fairly close to Wisconsin flipping back to being the tipping-point state. https://t.co/ajG88SznSA https://t.co/PSbIbWRwjB — PolitiTweet.org

Nate Silver @NateSilver538

Biden now up 6.6 in our Pennsylvania polling average. And that's on a fairly robust set of polling. Individual polls since the debate there have him up +9.5 (Monmouth, averaging their two models), +7 (YouGov), +7 (Times/Siena) and +5 (Ipsos). https://t.co/s32ZGD656G https://t.co/cu0jVZOK7R — PolitiTweet.org

Nate Silver @NateSilver538

The one saving grace for Trump is that Monmouth uses small-ish sample sizes (500 here, 400 in their previous polls) so they can bounce around a lot. They had Biden up 7-10 points in July but only 1-3 in August. Still, a really bad number for Trump here. — PolitiTweet.org

Nate Silver @NateSilver538

Monmouth has Biden with an 8 or an 11 point lead in Pennsylvania, depending on which version of their likely voter model you prefer. That's an awfully poor result for Trmup given that Monmouth's state polls had been a bright spot for him lately. https://t.co/s32ZGD656G — PolitiTweet.org

Nate Silver @NateSilver538

@jeffma @paulg @Betcris No, sophomoric is exactly the right term. It's people who think they're smart enough to beat 538/RCP etc but have not put in the many many hours that takes. I talk to these people, they are my friends, I like them, but they fall into that category. — PolitiTweet.org

Nate Silver @NateSilver538

@jeffma @paulg @Betcris Instead, you have people with some adjacent knowledge, e.g. in finance or sports betting or maybe people who understand politics but not polling/forecasting. While there are exceptions, those folks will tend to be pretty big whales betting in a domain where they're inexperienced. — PolitiTweet.org

Nate Silver @NateSilver538

@jeffma @paulg @Betcris There might be OK liquidity, but not fantastic, and another issue is that elections are rare enough that it would be hard to make a living betting on them (or to go broke by losing those bets). So you don't get the professionalization of the field. — PolitiTweet.org

Nate Silver @NateSilver538

@jeffma @paulg We have been on the right side of the markets in basically every election since FiveThirtyEight was founded with a huge ROI. Way different than sports where you have a much more sophisticated player base and more liquidity. — PolitiTweet.org

Nate Silver @NateSilver538

Trumps "it's just the flu!" messaging on COVID is likely to be quite unpopular, but it will be even more so if he's also still demonstrably suffering from personal health problems caused by COVID. — PolitiTweet.org

Nate Silver @NateSilver538

@paulg It's because political betting markets are populated by people with a sophomoric knowledge of politics. There are way too many excuses made for them when they have a poor track record. — PolitiTweet.org

Nate Silver @NateSilver538

Indeed, our national polling average is up to Biden +8.8 after adding the CNN poll and the USC tracker. Our popular vote *forecast*, which is mostly based on state polls and has some mean-reversion baked in, is at Biden +7.5 by contrast. https://t.co/cy51vc5isJ https://t.co/z62yVVXr9k — PolitiTweet.org

Nate Silver @NateSilver538

In fact, it's good! It's good that pollsters are publishing these results instead of sitting on them. And this is one of those times when a large movement in the race is *plausible*. But we have a lot of polling at it mostly suggests more modest gains for Biden. — PolitiTweet.org

Nate Silver @NateSilver538

Now, it's pretty clear that Biden has gained ground, from +7 before to +8 or probably more like +9 once we've thrown today's polls into our average, but with a +8 or a +9 you're going to get your occasional +3s (TIPP) or +16s (CNN) and that's OK. — PolitiTweet.org

Nate Silver @NateSilver538

I'm not sure people realize how much poll results will vary through chance alone. https://t.co/LnEazd9iNY https://t.co/HvBx2lRFEn — PolitiTweet.org

Nate Silver @NateSilver538

She's obviously got decent chances but appointed incumbents generally underperform and I think voters are *relatively* good at sniffing out a fake, and this is not among the more subtle ones. — PolitiTweet.org

Nate Silver @NateSilver538

Kelly Loeffler pivoting to #MAGA in an obviously-clumsy-and-consultant-driven way in the hopes of beating Collins when it's not likely to help her at all in the runoff has to be among the most... interesting... developments of the cycle. — PolitiTweet.org

Nate Silver @NateSilver538

RT @maggiekb1: It has been a hard year. Don't you think you need a T-shirt of America's cutest cartoon fox statistician? @FiveyFox https:… — PolitiTweet.org

Nate Silver @NateSilver538

For the Biden campaign to end its noncombative period on social media suggests it thinks these images are extremely damaging for Trump. https://t.co/QcjFnJqJLa — PolitiTweet.org

Nate Silver @NateSilver538

It probably doesn't end well, but it isn't clear yet in which way it ends poorly. — PolitiTweet.org