Predictions for the 2020 US Presidential Election

Politics & Elections
101

Well.

Prediction: There will be a high profile death in Trump’s orbit because of the large number of people around him who have Coronavirus, which will throw water on his last ditch efforts.
Biden wins by 11 points (essentially a hardening of current polls)

Biden clearly wins Arizona, Florida, Minnesota, Michigan, Nevada, North Carolina, Pennsylvania and Wisconsin. He also picks up all of the votes in Maine and the Omaha area in Nebraska.

Ohio is tight but called for Biden after some time.

Iowa quickly goes for Trump.

Three states are left in a giant mess: Georgia, Texas, and surprisingly Alaska. But they don’t decide the election, or its outcome. Indeed, Trump’s arguments about election fairness bait him into an unrelated battle that he can’t let go of.

Alaska and Texas wind up being called for Trump, Georgia votes for Biden, all three of which are squeakers with thin majorities. Like everything else Trump does as President, these fights quickly spiral away from him and into the evil things he’s done. Trump gets his procedural fights, but there’s no fundamental dispute about who won the election, and the questions wind up leading to national reforms over ease of voting.

In late January, all television coverage is interrupted as the greatest perp walk in history begins across 1600 Pennsylvania Avenue, into the waiting hands of an FBI truck and a trip to the Southern District of New York. This trip, hours long, will be watched by tens of millions as a President of the United States is taken to court, denied bail, and, mercifully, removed from his means of tweeting.

102

“I have definitely seen that graph(from the first post in this thread?) and it looks very hopeful.”

Yes, the original was from Oct 08. Back then Biden was still in Hillary territory before her crash starting on Oct 14.

Now that we are at Oct 19, Biden seems to have escaped that downward trend and is in Obama 2008 territory.

Let’s see what the next two weeks hold.

(FYI, the plots for the different election years use the same settings, and this year’s prediction is remarkably stable compared to previous elections, likely due to people being much more polarized now)

103

This is cheesy as can be, but I am literally praying daily that Biden’s lead is stable, real, and it is an easy and unambiguous win for him Nov. 3.

I want Florida, Ohio, Michigan, Pennsylvania, and Wisconsin to go to him and pretty much end all hope for Trump.

104

Related Senate-contingent predictions:

If the GOP retains 51 or more seats, there will be a failed Biden presidency due to fiscal austerity, perhaps including government shutdowns, that helps turn the current depression into a great depression.

If the GOP retains 49 or fewer seats, most Republicans, including Mike Pence, will gracelessly acknowledge Biden as president-elect. Then, on January 20, the new administration, and a filibuster-free Senate, will bring Americans an improved version of FDR’s 100 Days, with massive spending for economic stimulus and battling COVID.

If the GOP retains exactly 50 seats, we are in trouble, due to Republican Senate control at the formal January 6 electoral vote count. Maybe I’m paranoid, but I think that Trump and Company will then tell Mitch McConnell to break with tradition by appointing partisan Republican tellers who would instruct Mike Pence to throw out swing state electoral votes by alleging election irregularities. What happens next would be one of the most dramatic episodes in U.S. political history.

105

Anybody think the U.S. might hit 70% voter turnout?

106

I reiterate the point: in a free and fair election, the Democrats would sweep the White House and both Houses of Congress. But this is not a free and fair election. Not even close.

107

No, but I’d be happily wrong. Do you mean overall or “of registered voters”?

108

Yeah, what’s the divisor? Is it registered voters or eligible voters? I’ve never been clear about how they measure “turnout.” I believe we’ll have 150 million votes cast this year, which is 70% of 214 million, but we’d need about 157 million to get 70% of 225 (and I don’t know if that’s the right baseline, either) and that might be a stretch in a pandemic.

109

Yes, 157 million. I’d like to see it.

110

Regrettably, I think a more likely scenario is: In mid-January, Trump sends a “fuck you; you can’t catch me” message from Russia, Saudi Arabia, or some other country with no extradition treaty

111

Most have done nothing to get more republicans to answer polls. The primary way they changed was simply to assign more weight to the responses of those (comparatively few) republicans who are willing to answer polls. Which sounds good, until you realize that if there is a social desirability bias – i.e. if Pubs are avoiding pollsters because they think saying “I’m for Trump” is socially disapproved – then your sample is going to wind up overrepresenting those pubs who aren’t for Trump. You haven’t fixed the problem, in fact, when you assign more weight to that skewed sample, you’ve compounded the problem.

And hey presto, if you look in the crosstabs of some of these polls, you see 10+% of Pubs voting for Biden. Not likely.

Trump wins 286 EVs.

112

I was musing during dog walking last night …
Everyone seems to be assuming / worried about “Shy Trump Voter” syndrome.
What if the reverse were true? There’s actually a shy Biden voter problem?
Kinda like the idea of a beard or the way that we keep seeing all the virulent anti-porn / anti gay / pedophilia fighters actually being closeted or over compensating?

113

My prediction is that it’s going to be a repeat of 2000 and the courts will be involved before the election is officially called in early December.

114

Nonsense.

8% of Republicans voted for Clinton in 2016, and Republicans hate Clinton.

115

I’ve got a question: I’ve seen a few comments here and there that there’s no real way to know whether the early voting turnout right now is significant, or if it’s just front-loading of “usual” voters. At what point can/will we know if there likely is/was a significant increase in actual turnout?

116

Just to continue the grump: as the recent claims about Iran and Russia demonstrate, the Republicans have long been setting up the plan that if the Democrats win, the election result will be painted as tainted. Court cases and street violence ensue.

117

It’s unlikely to be that low, given that Biden has considerably more bipartisan appeal than Clinton and given the number and prominence of Republican influencers (Lincoln Project, etc) supporting Biden.

Er, which is it? Is the “Republicans for Biden” vote significantly below 10% (in which case no plausible amount of alleged overrepresentation is going to noticeably affect the poll results), or not?

118

This logic seems a stretch.

What 20 points does Trump lose from 2016? Pennsylvania’s 20? That’s the one battleground state he’s gearing up to cheat the most in.

119

  1. There are not many places where supporting Biden incurs social disapproval. That’s the way the teachers/bosses/celebrities etc. want you to vote.

  2. Social disapproval bias is rooted in data, and is well-known to pollsters; you just can’t quantify it until after the fact. In 2016, for example, “Undecideds” broke for Trump something like 2:1. Several pollsters predicted that, precisely because of social disapproval. You can guess at it by looking at other indicators. Let’s say a respondent indicates that he bought his first gun this year, that he doesn’t like BLM, attends church, and says his most important issue is reopening the economy after coronavirus. Common sense tells you that guy is 90% a Trump voter … but obviously a pollster can’t just say that. So the pollster marks the guy as what he says and issues a warning.

120

If you wanna say 306, or even 316, I won’t argue. But given that the history of voter fraud in PA is mostly in Philly and Pittsburgh, I think the Dems hold PA.