Nate Silver

Nate Silver @NateSilver538

I don't think it's the best use of my time to bat down every dumb theory about why Trump can win ... in part because Trump can win (!!) even though it probably won't be for the particular reasons the theory espouses. But it's kind of silly season out there right now. — PolitiTweet.org

Nate Silver @NateSilver538

The race is not tightening if you use any sort of robust methodology or look at the better polling. The tightening is almost entirely confined to a group of 3 or 4 pollsters that release a lot of polls and so can dominate polling averages that don't adjust for house effects, etc. — PolitiTweet.org

Nate Silver @NateSilver538

RT @kkondik: this is the right interpretation. That Iowa and Ohio still seem to be really close tell us that Biden's leads in MN-MI-PA-WI a… — PolitiTweet.org

Nate Silver @NateSilver538

@IChotiner That was a vital piece of journalism but alas, no, it doesn't affect our model; shy Trump voters are accounted for with the Yard Signs Coefficient. — PolitiTweet.org

Nate Silver @NateSilver538

It really doesn't feel like we're gonna get much change in the polling averages between now and election day. The one potential (and important) exception is Pennsylvania, which it seems like a lot of major pollsters are holding out as the final state they'll poll. — PolitiTweet.org

Nate Silver @NateSilver538

But there is a professional class of tens of thousands of people who price equities for a living, and can therefore take advantage of naive traders, whereas there are very few professional election forecasters. — PolitiTweet.org

Nate Silver @NateSilver538

Since almost no people have the relevant expertise to build political forecasting models (it takes tons of work and even then is easy to get wrong), political betting markets are basically just a competition over what types of people suffer more from the Dunning–Kruger effect. — PolitiTweet.org

Nate Silver @NateSilver538

@dandrezner @DanRosenheck @JamesSurowiecki Hard to think of a better contrarian indicator than what people overseas think is going to happen in a US election — PolitiTweet.org

Nate Silver @NateSilver538

@JamesSurowiecki So you have some pro-Trump traders thinking it's all coming down to Hunter Biden and shy Trump voters and they can't believe they can get 2:1 on him. The pro-Biden traders are meeker, meanwhile, and maybe more aware they don't have the experience to properly price the odds. — PolitiTweet.org

Nate Silver @NateSilver538

@JamesSurowiecki 1) It's not that much money 2) Almost no people have the relevant expertise and have invested the relevant time into election forecasting 3) Makeup of the trading pool may matter some, I'd guess traders are also not highly discerning about which news sources are reliable or not — PolitiTweet.org

Nate Silver @NateSilver538

Only 5% still undecided, or really more like 3-4% once you account for the third-party vote. https://t.co/cy51vc5isJ https://t.co/loKJfWQD5D — PolitiTweet.org

Nate Silver @NateSilver538

@JamesSurowiecki If they were accounting for court shenanigans they'd have moved a lot more with the various SCOTUS rulings and they haven't done that. — PolitiTweet.org

Nate Silver @NateSilver538

@JamesSurowiecki No, they're just dumb. — PolitiTweet.org

Nate Silver @NateSilver538

What to make of that? Do these (OH excluded) count as good polls for Trump because they're less bad than the usual from Quinnipiac? Maybe, although it can also be the case sometimes that polls that had been outliers before tend to fall more in line with their final release. — PolitiTweet.org

Nate Silver @NateSilver538

Quinnipiac has had some very good/borderline outlier numbers for Biden this year, but these are much closer to the averages in each state, except in Ohio. — PolitiTweet.org

Nate Silver @NateSilver538

Quinnipiac: Biden +3 FL Trump +1 IA Biden +7 PA Biden +5 OH https://t.co/j0mhcreiJC — PolitiTweet.org

Nate Silver @NateSilver538

Highly similar to the Marist poll this AM. Florida's been gyrating between "toss-up" and "lean Biden" and this last round of polling is putting it more in the latter category. — PolitiTweet.org

Nate Silver @NateSilver538

I go to a lot of Knicks/Rangers/etc. and I wonder what kind of odds I'd have given in February to "the next time you'll set foot in Madison Square Garden will be in October, to vote." No line right now BTW. — PolitiTweet.org

Nate Silver @NateSilver538

I don't want to be too snarky but Trump needs to close a big polling deficit and this is the message on a day with 80k new COVID cases and a 940 point decline in the Dow. — PolitiTweet.org

Nate Silver @NateSilver538

On average, Biden has lost 0.5 points in national polls since the debate, but gained 0.7 points in state polls, closing a bit of a weird state vs national poll gap we'd seen pre-debate. No signs of a tightening race overall, and Biden is probably gaining ground in the Midwest. — PolitiTweet.org

Nate Silver @NateSilver538

RT @FiveThirtyEight: LOTS OF POLLS. *Little sign of the presidential race tightening. https://t.co/eG0FjQuRH9 — PolitiTweet.org

Nate Silver @NateSilver538

I decided to write about... polls! — PolitiTweet.org

Nate Silver @NateSilver538

The Supreme Court rulings seem consistent with "SCOTUS will break a tie for Trump so Dems had better hope it's out of the recount margin" and not "SCOTUS will go to outrageous lengths to hand Trump an election that he's clearly lost". — PolitiTweet.org

Nate Silver @NateSilver538

A 538 slack chat snippet from today, pertinent to the Georgia special election. https://t.co/SKeuCU3hnJ — PolitiTweet.org

Nate Silver @NateSilver538

There's so many polls today that it's hard to string together a tidy narrative, and you can cherry pick your way to a few better results for Trump/Republicans here and there. But overall a lot of data and most of it quite good for Biden and Democratic congressional candidates. — PolitiTweet.org

Nate Silver @NateSilver538

Dems' generic ballot lead now up to +8.1 following some 👀 numbers from CNN and YouGov. Makes sense that these numbers are converging with the presidental race. https://t.co/phHcIJyqv5 — PolitiTweet.org

Nate Silver @NateSilver538

@conorsen The model's prior is less bullish on Warnock in the runoff than the polls are but the polls still get a fair amount of weight. Also an outside chance he could hit 50 on Election Day. — PolitiTweet.org

Nate Silver @NateSilver538

@conorsen That's correct, he's almost certain to reach the runoff and has a fairly big lead in head to head runoff polls. — PolitiTweet.org

Nate Silver @NateSilver538

@TheStalwart Right, yeah. — PolitiTweet.org

Nate Silver @NateSilver538

3. I'm glad we got a couple of high-quality national polls in this not-quite-final period because the final final national polls often do a lot of herding and therefore aren't always that informative IMO. — PolitiTweet.org