Sign in with Twitter

Username:

Editor-in-Chief, @FiveThirtyEight. Author, The Signal and the Noise (https://t.co/EYTxvN6BLY). Sports/politics/food geek.

1,341 Following   3,465,616 Followers   32,765 Tweets

Joined Twitter 8/27/08


We have AOC forecasted to win by 57 points. https://t.co/HRnLqxx3Aq https://t.co/ALPXZMN5kr @NateSilver538 After a season of @FiveThirtyEight podcast listening I feel like that question was bait just for you.
Retweeted by Nate Silver
10/25
2020
@injuryexpert I don't know if they'll be biased but there are a lot of ways they could just be wrong.I'd be so careful with the early exits this year in fact that if I somehow had the option to guarantee that I would… https://t.co/jLpMA5kz8y(Also, the early exit polls are often decently predictive, FWIW, though I'd be very very very very careful with the… https://t.co/OYavJJV5OqThe early vote. Because those hot takes last for a few weeks and not just a few hours. https://t.co/iV3LHFCNfmGeorgia's moved up to a ~3% chance of being the tipping-point state. That isn't high but it's the same as say NV. A… https://t.co/TVrnI7CW2UWhat we can say from the early voting data is that Democratic turnout is likely to be very high. Will Republican tu… https://t.co/hSha4CtaEF @davidshor @Nate_Cohn I do think it's important to look at one's edge cases! But the Economist guys tend to bring u… https://t.co/ziDcbp2qE5 @davidshor @Nate_Cohn If, as a result of that, there can be a negative correlation in certain edge cases (e.g. MS a… https://t.co/jUENbFryqg @davidshor @Nate_Cohn Wasn't criticizing you, to be clear! It's a hard problem and our model leans heavily into ass… https://t.co/8HxG9uX8JXAs I wrote earlier, our model will also use the debate as an excuse to lop off some older polling data. That could… https://t.co/iEkno4bJPZThat Gravis result is a good one for Biden, as IVR polls haven't tended to be friendly to him, although Gravis hasn… https://t.co/NvEnNyQqYlIn terms of post-debate polling: We've got one day's worth of interviews from the 3 daily tracking polls. Biden mo… https://t.co/KTZ0tn6LLB @davidshor @Nate_Cohn Our correlations actually are based on microdata. The Economist guys continually make weird a… https://t.co/hm7hMIdWnZI think we need to bail out indoor businesses like bars & restaurants to give them the option to stay closed, but t… https://t.co/VScB3z8Xio @NickRiccardi Yeah.... I just think the turnout patterns look like they're going to be different this year. Biden v… https://t.co/PjuO2XusNOOr maybe some of them won't vote at all, in which case polls might be underestimating the D turnout advantage and R… https://t.co/zjhRX7uuNYMail is the only way to vote in Colorado and every registered voter gets a ballot, so it's not like these Republica… https://t.co/fGrnOuhJ30Colorado's another example of why there won't necessarily be a "blue shift" in late-counted ballots (and could even… https://t.co/y9YYIIbzKKI'd also note that this didn't have much effect on our *forecast*, which is mostly based on state polls and uses di… https://t.co/1x6EEG8rLdIt's not perfect, and we'll probably tinker with these mechanics post-2020, but keep in mind that our averages are… https://t.co/206ngdcbqHFollowing major events like debates, our average shifts forward the window of time that it considers, and more rece… https://t.co/yznX1dxS6XHey folks, our national polling average is now Biden +9.2, as compared with Biden +9.7 yesterday. What caused the c… https://t.co/3VOCsmTQMF @stevesingiser Yeah, sometimes there are non-intuitive changes based on polls dropping out of the window the averag… https://t.co/GfEt2wl8ro @NateSilver538 Rising cases is one thing, and certainly a reason to be worried. But when combined with the 35% inc… https://t.co/xsTpgj8ZQK
Retweeted by Nate SilverI'm not some sort of COVID alarmist... I've tried to stay pretty even-keeled and have been burned by being too opti… https://t.co/4gEb5q1Dfv @conorsen I'm trying to figure out how much Biden would have to win by for the conventional wisdom to become that p… https://t.co/bqosNNSfjgWithout adjusting for partisanship (probably more D internals than R in here I would guess), recent congressional d… https://t.co/xIjDiGacdBI just walked through Greenwich Village for an hour. It is PACKED. I’ve never seen it busier. NYC IS NOT A 👻👻👻 TO… https://t.co/No7UKIFffE
Retweeted by Nate Silver
10/24
2020
@Crimealytics The nowcast is dead, Jeff.One way to think about Trump's chances: if the polls move toward Trump in the closing days rather than Biden (50/50… https://t.co/olfpa2z3sx @chrislhayes Probably true in most places, but in NY and some other states with good testing programs, maybe less so.This site, which adjusts for testing volume, estimates that all but 4 states and DC have increasing caseloads right… https://t.co/Gs32OtgBFRThe good news is that it's on a record number of tests, 1.28 million. However, positive test rates have increased,… https://t.co/D06qnc4wN7.@COVID19Tracking, my go-to source, is reporting 83,010 newly-diagnosed COVID cases today, a record high for the US… https://t.co/xkMlPsZG6R @SeanTrende @tucker_doherty Yeah. Like some of those Eastern European countries that did amazingly well in the spri… https://t.co/rXGVnvjJvO @SeanTrende @NateSilver538 The way I think about this is that it looks like rurality (in the U.S. context at least)… https://t.co/iaeJMyGPLO
Retweeted by Nate Silver @NickRiccardi @SeanTrende They're pretty similar density-wise but much redder.... https://t.co/46aA1poj7GAlso: small average household size, not very religious (to the extent that church gatherings can serve a vector for… https://t.co/5kOErfP2bAOne starting point might be that they are among the only states in the country that are rural (indeed, *quite* rura… https://t.co/JtjLfv1GLXWhat are Vermont, New Hampshire and Maine doing right on COVID? With almost every other state having been hit hard… https://t.co/Q21AKQ3jVLWhat are you reading this weekend? As we get closer to #ElectionDay, check out @NateSilver538's The Signal and the… https://t.co/2iuYD10KRz
Retweeted by Nate SilverAs someone who was very sharply critical of how the press covered 2016, I think it has by and large done a good job… https://t.co/1eAvXNaZ7BStates will count those as mail ballots, so if some of those folks say they're voting early in person (and I imagin… https://t.co/QKHBJkw39VI wonder if people who plan to *drop off absentee ballots at a box or other in-person location* classify themselves… https://t.co/uaB85Ky7iJIt seems like North Carolina could be a real thorn in the side of Trump's re-election prospects. The polling's been… https://t.co/Mn4SDdiisQSo maybe the mediocre polling shows a tie in certain states. But because it's mediocre polling, you need to hedge a… https://t.co/K6aU6sGlboThe other thing is that polling in smaller states that aren't presidential swing states tends not to be great. You'… https://t.co/ZlokOrKpapAnd while it's tempting to think that all of these races will break in the same direction (so e.g. either Democrats… https://t.co/ifHRGhGB1ZIn general, our forecasts have been very steady in these states and the Upshot/Siena polls showing narrow Republica… https://t.co/YgpSlrYP6XThis is true but I think a lot of the problem is that expectations ran a little ahead of themselves in states where… https://t.co/MgmhVgVgllOur post-debate poll with @Ipsos shows that voters liked Biden's performance better, but there was basically no cha… https://t.co/Knrh3T9ZUsThere really is kind of a parallel universe of Trump-friendly polls that are on a completely different planet from… https://t.co/UZZRcQWxcB @BenjySarlin Happy Comey Day to you, sir.Hello! It's me. Professor Polls. Welcome to POLLING 101. Grab a seat, and let's talk about how polling works...… https://t.co/MBGixDHqvO
Retweeted by Nate Silver @BenjySarlin On top of that, the WSJ fiasco with the Hunter Biden story was probably the last chance for that story… https://t.co/IEqpYzryEC @BenjySarlin Yeah, I think the takes were right directionally, but this is a weird case where a non-event (the deba… https://t.co/BeQVVySCTtTrump can win. It's mathematically possible that polls could be waayyy off or something could come from totally out… https://t.co/ppunXlyQxUI know this point was made repeatedly last night including by me, but after sleeping on it, I wonder if people aren… https://t.co/Ox89kOZRE9It seems increasingly like the choice to keep elementary schools closed in places where the virus is contained and… https://t.co/yTmAd2ACV6
Retweeted by Nate SilverI realize this is beating a dead horse, but...SSRS, CNN's pollster, also conducts some of their polling online (the… https://t.co/oiGsR3m0Cu
Retweeted by Nate SilverThe margin here is close enough to the overall margin in the horse race than I wouldn't expect too much change in t… https://t.co/wqYz5wK9kcThis is about what I would have expected. Trump lost all 5 of his general election debates (2 against Biden and 3 a… https://t.co/2Q1rAYO77c @marcambinder Wanna guess what the post-debate polls will say? I think reporters pay *more* attention to style than… https://t.co/SOE0K3PeEQA lot of very Halperin-esque takes on here that elevate style above substance re: Trump's performance tonight.Hey folks: This is probably going to be the LAST PRESIDENTIAL DEBATE EVER FEATURING DONALD TRUMP. We're liveblogg… https://t.co/7ZSCiELGZgAnd you can order Fivey merch right here! https://t.co/LzBRZG6VTeYeah. Even if you take the account at face value (which, maybe you shouldn't given that the WSJ newsroom seemingly… https://t.co/aQWblMGxWVAlthough the Midwest (win back PA/WI/MI) path is Biden's path of least resistance, FL and NC are *close enough* to… https://t.co/OvPxuCULlpA few key takeaways: * The state polls haven't moved as much as national polls have. * Biden's gains in PA/WI/MI i… https://t.co/JrIvM6aPLSQuick snapshot of how the polls have changed since the first debate. https://t.co/AlOBLTrTin
10/23
2020
So the big WSJ story the Trump folks were buzzing about is going to be in the ... opinion section? https://t.co/wd9aCveKysTest positivity rates are slightly lower than the summer peak and considerably lower than the spring peak, but the… https://t.co/usM4KUTnLdIt would seem pretty likely that we're going to have some days with record-setting COVID case counts before the ele… https://t.co/0j1QgZQN3JNormie Democrats and the media are *paying a lot more attention* to voting rights and voter suppression now (long o… https://t.co/aNGeCxxWTsSo…I do think it's worth mentioning that the pendulum has swung somewhat from contracting voting rights to expandin… https://t.co/F2vdhPK0kNI think I'm making a more subtle point here, which is that a 15% vs. a 35% probability isn't highly actionable in t… https://t.co/soj0mnfNOh @jbarro Yeah, especially since it's those folks who are more likely to not ultimately have their vote counted. @jbarro I'm not sure I buy that conditional on not having voted yet, a Democrat who says they're going to vote on E… https://t.co/TE8g2GBytj @jbarro I mean I think the way to do it is just to use weighted turnout probabilities where "already voted" gets a… https://t.co/wKxSRQQvTbOTOH, some people who are deemed to be likely voters *don't* vote. So the upside case for Dems is if 100% (or more… https://t.co/swHeuaJRDFA bit less contrived: some "unlikely voters" actually do turn out to vote. So if you screen out a bunch of Republic… https://t.co/SNTGV7P9XRFor instance, if you had a fixed estimate of turnout (as X% of registered voters) and you put people who had alread… https://t.co/6Bs42ib2b6I think there's *probably* more upside risk than downside risk for Democrats here (i.e. that turnout will be bluer… https://t.co/NqClJhbmxzIt's definitely a source of uncertainty and different ways that pollsters are handling early and mail voting vis-a-… https://t.co/u5DAw96dsi @poniewozik Probably closer to 60-65% I'd guess. @darth It's hard to know. A fair number of the changes to state laws are supposed to be temporary. If the election… https://t.co/LBqS0HbrTi @darth My personal confidence level is no higher and no lower than the forecast. I think we mostly make fairly cons… https://t.co/uosINUsy9NOn the other hand, a 10 or 15% chance is *really* different than 0%, both mathematically (1 in 8 is a lot different… https://t.co/6agLu5pXzzAre people overcompensating in the other direction? Probably a bit. But here's the thing. Unless you're interested… https://t.co/1MLLlJydy3If there hadn't been a (not actually all that large!) polling error in 2016, you can imagine the conventional wisdo… https://t.co/SpJK8UbM7uClippy is Fivey Fox's father. https://t.co/1QNf8ibreG @joe_sheehan LOL. We're expecting turnout of around 65% of the voting- eligible population, which would be the highest since 1908.FYI, if you scroll down to the bottom of our forecast, you can find a link to download TONS of additional data that… https://t.co/JkwefSI7mkI'd note to pollsters that our pollster rating algorithm detects herding and it will result in a lower rating when… https://t.co/bXq56yhHePPlausible! Also, especially after the debate, we're going to start to get a lot of "final" polls, which is when the… https://t.co/bnk1hzeacbMaybe it helps rev up Trump's base a bit and gives him a higher floor, but Trump can't win if the Dem base is also… https://t.co/fNtt79DPp4Sorry, forgot the link. https://t.co/Ovh3Bmhpb6No indication at all of a decline in Joe Biden's favorability ratings, which is what Trump presumably would be hopi… https://t.co/Y5gfOwVxIa
10/22
2020

0