US Elections 2020 - very self such moderated

[Cnut237]

the alphabetical order conspiracy might be onto something

I can 100% guarantee you that position in a list is an important factor in which item people pick from that list.

There is certainly evidence to support this, even in US presidential elections. Quotes below from this article.

"There is a human tendency to lean towards the first name listed on the ballot," says Krosnick, a politics professor at Stanford University. "And that has caused increases on average of about three percentage points for candidates, across lots of races and states and years."

Whilst most people are not affected, it does impact the decisions of people who are unsure who to pick, but also want to (or feel obligated to) vote for someone. This does make sense. Universal suffrage is a hard-won right, and I can imagine there are people who will make absolutely certain they take advantage of that right, even if they have no real preference of candidate.

In 2016,

"In the states where Trump won very narrowly, his name was also listed first on the ballot in most of those states,"

Also:

In 1996, Bill Clinton received 4% more votes in the regions of California that listed him first in the ballot papers than in those where he featured lower down the list.
Research by Robert Darcy of Oklahoma State University shows that, given the choice, most election officials tend to list their own party's candidates first.
In one famous example of this, Florida's rules meant that Republican governor Jeb Bush's brother George W Bush was placed at the top of the list of candidates in his state, in the 2000 presidential election.
Bush went on to win Florida - which turned out to be a decisive state - by a very narrow margin.

"Because of the fact that different states in the US order candidate names differently and idiosyncratically, and almost none of the states do what Ohio and California do which is to rotate candidate name order across ballots to be fair, we have unfortunately had at least two recent election outcomes that are the result of bias in the name ordering," says Krosnick.
"If all of those states had rotated name order fairly, most likely George W Bush would not have been elected president in 2000, nor would Donald Trump have been elected president in 2016."

Apparently (as of 2017) California and Ohio are the exception:

Some always list parties in the same order. Some allow the state's officials to make a new choice each time. Some put the party that lost in the last election at the top of the ballot. Some list alphabetically.

The "state's officials" bit could be important there - particularly if this is happening in swing states.


Bit of an aside, but presumably (and in fact from personal experience, although not in the US) the effect is more pronounced in committee elections - I have received ballot papers where you can vote for 3 candidates, but there are 10 standing, 5 from each major party. If you just want 3 from your choice of party, but aren't bothered which ones, there's an obvious benefit to being top of the list.

[1714181827]

1714181827

[1714181827]

1714181827

[1714181827]

1714181827

[DaveF]

Bloomberg spent / spending up to $100 million against trump in Florida.

https://www.politico.com/news/2020/10/20/bloomberg-trump-florida-spending-430236

As everyone here in NY knows Trump / Bloomberg really don't like each other.
But it is interesting to see how much he is willing to spend just to make Trumps life more difficult.

Also, kind of shows the point that Trump really does not have any supporters with deep pockets and ready cash. If he did countering that kind of cash dump since mid September would be simple.

-Dave

[nutildah]

This is my new projection of how things are going this year, with the election (unsurprisingly) hinging on Florida:



If the polls are even slightly more accurate this time around then theoretically Florida should go to Biden, if only by a few chin hairs. Here is a link to the unadulterated electoral map as things currently stand:

https://www.realclearpolitics.com/epolls/2020/president/2020_elections_electoral_college_map.html

You can also create your own map scenarios here to fantasize how you think its going to play out:

https://www.realclearpolitics.com/epolls/2020/president/create_your_own_president_map.html

Right now I think its a genuine toss-up. The statistical probability that the election will be decided by as many or fewer votes as it was in 2000 is almost nil, but I think it will be the closest election we've had since then.

[suchmoon]

~

I think WI and AZ are blue-er than Florida... maybe PA too. Even Rasmussen puts Biden ahead in those states, just barely.

[eddie13]

I hear/read that the early voting results so far in FL are looking very promising for a Trump win there, and Biden just greatly hurt his chances in PA and OH with his comments about "phasing out"/shutting down the oil industry...

Starting to get more confident in a Trump win again..

[Gyfts]

https://twitter.com/NateSilver538/status/1320127823512821765

I'm not too optimistic.

Democratic voter turnout has to be historic considering there was a large chunk of voters that stayed home in 2016 and regretted it. Republicans in Florida has had surprisingly decent numbers based on early voting, but Michigan, Wisconsin, and Pennsylvania I'm afraid won't even be close. I REALLY hope I'm wrong but I doubt dem's are going to stay home.

A part from people staying home last cycle, maybe voters in Michigan, Wisconsin, and Pennsylvania  resonated with Trump on manufacturing jobs. Combine this with his energy independent strategy, could he possible snag one of the three? Who knows. PA and WI might be possible, MI isn't close according to the RCP. iirc MI wasn't even within the margin of error for some of the polls.

Silver's model gave Trump a 12 in 100 chance compared to a roughly 1/4 chance last year. His chances were roughly 1/4 one month ago but of course his campaign took a nose dive with the first debate and COVID diagnosis.

Trump's been on the campaign trail hosting multiple rallies a day so Biden is sitting comfortably right now while Trump is on the trail. This complacency is what caused Clinton to lose.

[TwitchySeal]

I hear/read that the early voting results so far in FL are looking very promising for a Trump win there, and Biden just greatly hurt his chances in PA and OH with his comments about "phasing out"/shutting down the oil industry...

Starting to get more confident in a Trump win again..

Looks like for in person early voting they're at  ~ 640k for republicans, 500k for dems.

Mail in is 1.64m to 1.04m in favor of democrats.

So democrats actually have a significant lead now, but that's expected to shrink pretty consistently up till election day since a higher % of Republicans have yet to vote (they're more likely to vote in person).

https://www.tampabay.com/news/florida-politics/elections/2020/10/23/florida-republicans-surge-ahead-with-in-person-early-voting/

[KonstantinosM]

I look at all the material that has Biden winning but then I look outside at all the Trump people. I haven't seen people gather for Biden anywhere here. I believe Florida at least is going to Trump.

There are some houses with Biden/Harris stuff on them though. It's not as one sided as 2016.

[TwitchySeal]

I look at all the material that has Biden winning but then I look outside at all the Trump people. I haven't seen people gather for Biden anywhere here. I believe Florida at least is going to Trump.

There are some houses with Biden/Harris stuff on them though. It's not as one sided as 2016.

The real question is who do you think is more qualified to figure out the opinion of ~150 million people?  Polling companies with a proven track record over thousands of elections  - some spanning more than 5 decades, or you when you look outside?

[suchmoon]

The real question is who do you think is more qualified to figure out the opinion of 50-100 million people?  Polling companies with a proven track record over thousands of elections  - some spanning more than 5 decades, or you when you look outside?

Speaking of things people see, I do hope that the lines at precincts will impart the same herd mentality (not to be confused with herd immunity) and sense of urgency like lines at Apple stores or TP shortage in March did and will encourage more people to vote. At the end of the day, at the end of November 3 in particular, that's really the only thing that matters regardless of crowds, yard signs, polls, and forum debates. As little as a 10% change in turnout of a specific demographic can in theory flip the whole election.

[theymos]

The whole Republican strategy seems broken to me. They're talking about all the wrong things. At this point, I'm thinking that Biden is going to win by a landslide. He's currently ahead in Florida and Arizona, which Trump needs to win. The polls may still tighten, but I suspect that pollsters are more likely to have overcorrected than undercorrected after 2016; in other words, I expect polling error to go in the opposite direction to 2016.

On election night, we should watch Florida and Arizona, since they're apparently going to have some solid results early. If Biden wins either, then it's over. If Trump wins both, then it'll go into a big mess of delays and litigation in the other swing states. If Trump wins FL&AZ, then it also implies a fairly tight election, which gives Trump the opportunity to try for various tricks such as trying to have mail-in ballots thrown out in the remaining states. But I think Biden wins Florida and Arizona, which will be a bit of an anticlimactic end to the whole thing.

Ticket-splitting has become rare, so Trump's disaster of a campaign is likely to lead to the Democrats taking the Senate as well. It's always bad when either party has full control, since then the government can actually do stuff, and everything the government does is harmful. Disappointing.

Biden represents a return to the neoliberal status quo, which I would not have expected. The neoliberal ideology and post-WWII world order has been falling apart worldwide, and we seemed (still seem?) headed inexorably toward something else. The protracted and disastrous wars in the middle east were major blows to the political strength of the US and neoliberalism's strength in domestic politics, as was the 2008 financial crisis. Trump's open and loud rejection of neoliberalism seemed like the final nail in the coffin. But now Biden will attempt to "go backward". People voting for Biden want and expect a return to something "ordinary", but I think it'll be moreso Biden fighting to recreate something that can no longer exist in this world.

[eddie13]

I'm thinking that Biden is going to win by a landslide.

Wanna bet me another 0.01? Or?

[Spendulus]

The whole Republican strategy seems broken to me. They're talking about all the wrong things. At this point, I'm thinking that Biden is going to win by a landslide. He's currently ahead in Florida and Arizona, which Trump needs to win. The polls may still tighten, but I suspect that pollsters are more likely to have overcorrected than undercorrected after 2016; in other words, I expect polling error to go in the opposite direction to 2016.
....

That's rational assuming the slant isn't purposeful and an attempt to create an impression Biden's ahead.

[PrimeNumber7]

Polling companies

polls

It has become difficult to conduct polling with the advent of the cell phone, and the outrageous number of spam calls in recent years have made it even more difficult.

I don't think pollsters have been able to fix their problems from 2016. In 2018, there was one poll that had Andrew Gilliam up by 7 points right before election day (FL governor), and DeSantis ended up winning. The other polls right before election day had DeSantis down by 4-5 points.

Part of the problem is the strength of the Democrat propaganda machine and its strategy of labeling Trump supporters and Republicans as racist. This makes Republicans more likely to hide their political views, especially from strangers (pollsters). I would trust voting data regarding party affiliation more than polls, but this is infallible, for example, approximately 30% of people at Trump rallies are Democrats, and may vote for Trump.

[nutildah]

~

I think WI and AZ are blue-er than Florida... maybe PA too. Even Rasmussen puts Biden ahead in those states, just barely.

After what happened last time, I don't have faith in the pollsters at all. To come up with my results what I did was look at an average of polls from 2016 at this time of year and find the difference between what they projected vs. what actually happened (what I referred to as "M.O.E." in the table below). It's basically working on an assumption that they will be equally as wrong this time around. If they are even just slightly more correct, Biden has a good shot of winning Florida; WI and AZ not so much.

State	M.O.E.	Biden has	Biden needs
Pennsylvania	5.9	5.1	0.8
Florida	2.0	1.5	0.5
Wisconsin	7.2	4.6	2.6
Michigan	6.5	7.8	-1.3
North Carolina	6.2	1.5	4.7
Arizona	5.0	2.4	2.6

[TwitchySeal]

Polling companies

polls

It has become difficult to conduct polling with the advent of the cell phone, and the outrageous number of spam calls in recent years have made it even more difficult.

I don't think pollsters have been able to fix their problems from 2016. In 2018, there was one poll that had Andrew Gilliam up by 7 points right before election day (FL governor), and DeSantis ended up winning. The other polls right before election day had DeSantis down by 4-5 points.

Part of the problem is the strength of the Democrat propaganda machine and its strategy of labeling Trump supporters and Republicans as racist. This makes Republicans more likely to hide their political views, especially from strangers (pollsters). I would trust voting data regarding party affiliation more than polls, but this is infallible, for example, approximately 30% of people at Trump rallies are Democrats, and may vote for Trump.

I know this is all stemming from Trumps 'fake polls' rhetoric, and I'm never going to convince you to not be a poll-denier, but...

The polls for the 2018 House, Senate and Gubernatorial races were over remarkably accurate.  As in they got lucky.  The average poll was within 2.8% with a slight bias for Democrats of 0.4%.   And that included state and district polling which is far less accurate historically than the major polls you see in the days before a presidential election.

Looking at just the polls that were off the most and then using those handful of poll to calculate the reliability of the thousands of other polls doesn't really make sense.  You have to look at all the results.

[suchmoon]

Let's try to get back on track. Democrats said this, Republicans said that - doesn't matter much anymore with about a week to go.

I find it quite amazing that 40% of 2016 turnout have already voted... 55 million people by some estimates. Democrats up by about 8-10% (by party registration numbers where available) but as TwitchySeal noted about Florida, Republicans will likely keep catching up as the election day draws closer. Makes me wonder what the total turnout is going to be.

After what happened last time, I don't have faith in the pollsters at all. To come up with my results what I did was look at an average of polls from 2016 at this time of year and find the difference between what they projected vs. what actually happened (what I referred to as "M.O.E." in the table below). It's basically working on an assumption that they will be equally as wrong this time around. If they are even just slightly more correct, Biden has a good shot of winning Florida; WI and AZ not so much.

I don't believe the margin of error is going to be exactly the same... if for no other reason than demographic and other changes in the 4 years. Some polling issues specific to Trump (underweighing non-college-educated voters for example) have probably been addressed so I would expect smaller and less-one-sided errors this time around.

[DaveF]

Quoting myself from another Trump thread:

I think a lot comes back to the saying "it's the economy stupid" I'll add it's also about the money. There were another 900k +/- people filing for unemployment this week. A lot of people are seeing their jobs leaving and not coming back.

The people who were doing well, had $$$ in the bank and a secure job and skills to get another one are fine. But, there are a lot of people who are seeing their unemployment run out, having to worry about how they are going to pay for food and other things and accurately or not blaming the current administration.

I also said it a while back, with the changes to the SALT deduction there were a lot of people, outside of what people considered the "blue cities" with high taxes that took a big hit.

I have a good friend who has been a 100% Republican line voter over since 1988 voted 100% democratic in 2018 and is going to do it again in 2020. The entire community where he lives in a solid red state is now blue and that says something. Why? because it's all large private land owners. Even at dirt cheap property tax rates, due to the large plots of land they all own it was still a big number they they could write off. Now they can't and it hurt. $5 an acre for unimproved land in taxes is nothing. Until every plot is miles and miles on each side. The surrounding area also used to be red was purple in 2018, going to be interesting to see how 2020 turns out there.

Just my view.

-Dave

With a week and a couple of days to go I still see this as a major issue. And I really think a lot of the republicans' are going to get blindsided by it.
Beyond Covid there are a lot of people who have been hit hard by the current administrations economic policies and they are pissed.

theymos said it well too.

The whole Republican strategy seems broken to me. They're talking about all the wrong things.......

But I will add to it. They are talking about the wrong things in the wrong areas.
If you are a Republican in PA you care about JOBS, coal / steel / manufacturing, whatever yeah you might have opinions / views even strong ones about immigration but for most of them it's probably not their main concern. Reversing it in AZ, more probably care about immigration then manufacturing and farming jobs. But you are not seeing the targeting of ads where they should be in my opinion.

Putting the 2 together:

Having ads that say Biden bad, Trump good. Biden raise taxes, Trump lower taxes works in some areas. But not the ones where people have been looking at their tax bill that went up big time for the last 2 years.

Same with many small businesses that sprung up along the US / Canada border. In the grand scheme of things NAFTA might have been good / or it might have been bad.
For the people who setup businesses that relied on it and certain functionality of it, well they are not happy. Saw a bunch on TV the other day their congressman was talking about how much better it was now. Yeah, closed border due to covid and 50% loss of cross border traffic before that. It's sooooo much better. Long term it might be better. But for now they seem to feel that long term is so far down the road it's not coming.

-Dave

[squatz1]

Would probably make more sense to put the two top parties on top though, right?

What's a top party though? Technically you can't know until the votes are counted and ballot design shouldn't rely on polls.

Then again your decision on who to vote for shouldn't change based on where they are on the ballot.

Precisely.

Oh 100%

I'm pretty sure that like, in regards to how stuff gets listed on a ballot, everyone is randomized in terms of order based on the county that you're in. I know that's been a common thing to complain about like:

"WHY IS TRUMP ON THE TOP OF THIS BALLOT" (You can insert Biden for that too)

"WHY IS JO JORGENSEN ON THIS ONE, IT CONFUSES PEOPLE TO VOTE FOR THEM INSTEAD OF JOE BIDEN" - wtf?

People are amazing.

[theymos]

It has become difficult to conduct polling with the advent of the cell phone, and the outrageous number of spam calls in recent years have made it even more difficult.

If Trump wins, it will indicate that polling has utterly failed, which would be interesting. However, the 2016 polls weren't actually that far off, with most being within their margins of error; it was mainly an issue of people ignoring the margins of error and "just knowing" that Trump couldn't possibly win. Now the situation is more-or-less reversed: a lot of people "just know" that the polls are wrong and Trump will somehow pull it off, so they're baking in a several-point swing from current polls. However, if the election were held today, polls would have to be much more wrong than they were in 2016 for Trump to win, and if pollsters overcorrected as I suspect, then the real result may swing against Trump compared to polls, not even for him.

While there are several very bad things about Trump, I think that a Trump victory would be far better than a Biden victory. Trump is more anti-war and anti-regulation, and he's much less effective at actually getting the government to do things. It'd also lead to increased distrust of government and perhaps some serious anti-federalist and secessionist movements among blue states, which is excellent from a libertarian perspective. And on a personal level I'd also find it hilarious to see Democrats' reaction if Trump pulls off a surprise victory again. But I just don't think that it's going to happen. The best we can probably hope for is that the Republicans retain the Senate and we get at least 2 years of gridlock, though even this is in serious doubt.