I’ve been checking out the forecasts by FiveThirtyEight today, and their latest nowcast now shows Donald Trump winning the election. Here’s the data (there’s 6 images in the album, all data was print screened from the site): Click here.
Normally the predictions have been leaning towards Clinton until today (at least for the nowcasts, other polls have been tight), and I for one am getting interested in the recent developments. Any opinions on the data/situation?
NM
A post convention bump is expected, I don’t know how much that plays into that prediction. Hillary should get a similar bump this week returning the polls back to where they were a week ago. It’s been clear that a Trump win is possible because it is a tight race just like almost all presidential elections with undecided voters able to swing it one way or the other. But undecideds never all go for one candidate, it gets down to the nitty-gritty details of precincts and districts and state electoral totals.
It isn’t even August yet, there’s no way, especially this year, to make a solid prediction. This is not like any election I know of with two candidates with very high negatives and division within both parties. But I still expect it to look like most elections as November approaches, a tight race, with key swing states leaning towards the winner.
I’m a little surprised - I didn’t think Trump would make it this close a race. I think I underestimated how much damage the Republican lie machine has done to Hillary over the years - Trevor Noah was right to observe that they are each running against the only candidate they could possibly defeat.
I certainly didn’t expect any bounce out of that debacle of a convention.
There’s always a bounce. It may not be tied to anything at the convention itself, it’s just the time when the news focuses on the candidates and the election and people may just start responding to the polls. I don’t have the figures to look at, but I get the impression there’s no consistent pattern around the bounces and the results in the election.
I have my doubts. Some people are miffed that Bernie Sanders isn’t the nominee, and some are more miffed that he appeared to have been cheated out of the race. Now I’m pretty sure that the DNC convention will have some viewers, but I have my doubts it would provide enough viewership to create a reversal.
I wonder what would cause a shift in the poll?
Never underestimate Scott Baio fanbase.
What else are people going to pay attention to this time of the year? The Democratic Convention will get watched by all of the party’s base and all of the politically interested people just like the GOP version. Hillary will get a bounce too. I’d be surprised if the shift doesn’t disappear by next week and it goes back to where it was a week ago. A trend following the conventions can make a difference.
Both Democrats and Republicans fucked up monumentally. They each chose unelectable candidates. A Romney-type against Clinton would be a blow out win for Republicans. Biden would be a blow out win by Democrats. It really is an opportunity for a good third party candidate.
To do what? Throw the election to whichever major candidate they’re farther from? Because that’s what third parties do in the US.
Right, it’s an opportunity for a third party candidate to win a few electoral votes. I don’t really think people who say “it’s a good opportunity for a third party candidate” actually know how our state-by-state first past the post elections work.
As for convention bump, FiveThirtyEight explored that recently–their analysis is that convention bumps have been getting much smaller in the last few elections, largely because of the constant news cycle and internet news, the conventions aren’t really increasing exposure over the norm to the degree they once did.
It used to be that getting time on all the networks in prime time was an immense thing, it’s still an increase in publicity, but not to the degree it once was.
FiveThirtyEight also noted that Trump had started out in their first projection at 19% chance of win, and had been trending up more or less continuously since then, going into the convention. For that reason it’ll be difficult to assess whether he truly got a convention bump or not, since he was already trending upwards.
The Democrats can still change candidates. Clinton does not have a pledged delegate majority, does she? If not, then the superdelegates can vote against her and then they can select a new candidate.
It’s also possible that we’re still seeing the Comey effect rather than a convention bounce. As Silver pointed out, Trump started to gain on Clinton when Comey gave his press conference.
Meanwhile, over at the Kentucky Derby of the Dead, Secretariat is still ahead of Seabiscuit.
I’m not sure I’m understanding you correctly, but, yes, she has a majority of the pledged delegates, 2205 v 1846. But if you’re asking does she have enough to win just on pledged delegates alone, no. It’s 2382 to win with pledged delegates and superdelegates, so if most of them switched to Sanders, yeah, she could lose.
CNN is similarly reporting that “there hasn’t been a significant post-convention bounce in CNN’s polling since 2000.” Trump apparently got six points, which they’re saying is the largest since both Gore and Bush got 8-point bounces. Nothing seems predictable this year.
Plenty of Google-able news reports from 2008 that mention a post-convention bump of 5-10% for McCain and Obama.
Hell, just found one that had the two tied on September 17th, 2008, two days after Lehman Brothers.
To be clear, 538 has always said Trump could win the Presidency. There has never been any point since the nominees became apparent that it was not the case that 538 said either Clinton OR Trump could win. Neither has ever been an overwhelming favourite.
I am been pointing out for a long time that the election was close, but for reasons I don’t really understand, some people simply do not want to believe the plain facts; you will find folks, on this board and elsewhere, who will swear up and down that a Clinton victory is totally guaranteed, and you’ll find more still who swear a Trump victory is totally guaranteed. I am especially mystified by those on here who are certain Clinton will win, given that this board is both biased towards objective fact AND has a history of bringing up useful resources like 538 or the Princeton Election Consortium.
The history of people saying “science is wrong because of” and then insert your chosen excuse (GOTV, the polls don’t count people with cellphones, black people won’t show up to vote like they did last time, this and that and blah blah blah) is dreadful. The Wangs and Silvers and similar methodologies have been shockingly accurate, over and over.
I think it very important to remember that you should NEVER trust one poll. An aggregation of polls, done carefully, is a much better indicator of the truth. Sam Wang notes today that the total GOP convention bounce, if you look at all the evidence, is about two to three points, which is consistent with recent history.
Bounces or not, I don’t think it matters much. After every election someone will say the losing candidate would have won if the election were held on Aug. 20, or Sept. 7, or Oct. 12, or some date when they were ahead in the polls. Only the big poll on election day counts, post convention polling doesn’t.