Author Topic: US Elections 2020  (Read 259115 times)

merithyn

  • Octogon Champion
  • *************
  • Posts: 13240
  • Unabashed Anglophile
Re: US Elections 2020
« Reply #7095 on: September 20, 2020, 01:00:22 pm »
I didn't dig too deeply into this, and we all know how photos can be manipulated, so I've no idea how many people were here, what they were doing, etc. Nonetheless, it appears to go against the nature of a free and fair election.

Trump Supporters Disrupt Early Voting in Virginia

Quote
A group of Trump supporters waving campaign flags disrupted the second day of early voting in Fairfax, Va., on Saturday, chanting “four more years” as voters entered a polling location and, at one point, forming a line that voters had to walk around outside the site.

County election officials eventually were forced to open up a larger portion of the Fairfax County Government Center to allow voters to wait inside away from the Trump enthusiasts.

Election officials said that the group stayed about 100 feet from the entrance to the building and, contrary to posts on social media, were not directly blocking access to the building. But they acknowledged that some voters and polling staff members felt intimidated by what some saw as protesters.

“Citizens coming into and leaving the building did have to go by them,” Gary Scott, the general registrar of Fairfax County, said in a statement. “Those voters who were in line outside of the building were moved inside and we continued operations. Some voters, and elections staff, did feel intimidated by the crowd and we did provide escorts past the group. One of the escorts was the county executive.”

In an unnerved electorate, where concerns about voting rights and safely voting amid the coronavirus pandemic are at a fever pitch, the demonstration outside of a polling place served as preview of a likely contentious election season, and how groups may be utilizing tactics that rattle or even deter voters over the next six weeks.

The disruption came as President Trump has repeatedly sought to undermine confidence in the upcoming election, spreading falsehoods about voting by mail and declaring the election “rigged” before any votes have even been cast.

The demonstration originated from a “Trump Train” parade that began in nearby Prince William County and featured Tommy Hicks Jr., the current Republican National Committee co-chairman. The event was set to end in the parking lot of the government center, which was also serving as the polling location on Saturday. Some of the people who attended the parade walked over to vote. Others gathered outside and began chanting, “four more years, four more years!”

Sean Rastatter, a vice chair at the Fairfax County Republican Committee who was at the polling location, said that he did not think any actions came close to voter intimidation, and that many of the discussions from members of the group were with journalists.

“I don’t think there was any way to need or feel intimidated in any form,” Mr. Rastatter said. He added that county officials asked the group on several instances to back away from the curb, and that the group complied.

Bryan Graham, the chairman of the Fairfax County Democrats who was also at the polling center, saw it differently, writing on Twitter that “the Republicans are straight-up attempting to intimidate voters at the government center.”

In an interview, Mr. Graham said he’d “never seen or heard of anything like this happening before.”

“I was there when the county executive was there and I saw him walk multiple people through the crowd because they didn’t feel safe,” Mr. Graham said. “I don’t think it’s appropriate. We shouldn’t be doing things to make people feel unsafe.”

Steve Descano, the county attorney, did not directly address the demonstration in a statement posted to Facebook on Saturday, but said that he was “instructing my office to pursue cases of voter intimidation that may occur.”

Virginia election law states that it is illegal to “hinder or delay a qualified voter in entering or leaving a polling place,” and that it is also prohibited to perform any kind of political advocacy within 40 feet of any entrance to a polling place.

Videos of the demonstrators quickly went viral on social media, and show them gathering outside the entrance, holding up Trump campaign signs and chanting as voters walked past them.

Some election rights groups said that the Trump group might have still crossed a legal line.

“In Virginia, the safe zone around the polling location is only 40 feet, but that safe zone is for campaigning and trying to change a person’s vote,” said Sylvia Albert, director of voting and elections at Common Cause, a voting rights group. “Outside of that, in general, there are laws against intimidation. So I would say that even if they have a right to campaign, which they absolutely do, they do not have a right to interfere with someone’s right to vote or to intimidate them.”

“To me, this went beyond campaigning and they should have been removed.”

The first two days of early voting in Virginia were marked by high turnout and long lines, with voters beginning to line up early in the morning, hours before polls opened. The state offered expanded early voting in an attempt to alleviate crowding at polling centers during the pandemic.

Ms. Albert added that the demonstration underscored the need for counties and municipalities to develop contingency voting plans for all situations.

“Contingency plans aren’t only if there is a hurricane or a pandemic,” she said. “It’s also if something is making my polling location inaccessible to voters. That could be power going down, that could be people protesting and blocking the door.”
« Last Edit: September 20, 2020, 01:01:55 pm by merithyn »
Yesterday, upon the stair,
I met a man who wasn’t there
He wasn’t there again today
I wish, I wish he’d go away...

Tyr

  • Blessed by Valmy
  • **************
  • Posts: 20930
  • In the end we will win
Re: US Elections 2020
« Reply #7096 on: September 20, 2020, 02:47:37 pm »
On all the models predicting a certain Clinton dictionary - wasn't it well observed at the time that the polling methods were very wrong and they've spent the past few years trying to fix the?
██████
██████
██████

DGuller

  • Blessed by Valmy
  • **************
  • Posts: 22904
Re: US Elections 2020
« Reply #7097 on: September 20, 2020, 03:48:09 pm »
On all the models predicting a certain Clinton dictionary - wasn't it well observed at the time that the polling methods were very wrong and they've spent the past few years trying to fix the?
I don't think it was the issue with the polling methods, but rather with the interpretation of the polls.  There is a big level of complexity between the poll results and the prediction of electoral college votes, and that's the part that 538 got right by using sophisticated statistical methods, and most others didn't even realize they had to think about.

Sheilbh

  • Blessed by Valmy
  • **************
  • Posts: 21043
Re: US Elections 2020
« Reply #7098 on: September 20, 2020, 04:13:40 pm »
On all the models predicting a certain Clinton dictionary - wasn't it well observed at the time that the polling methods were very wrong and they've spent the past few years trying to fix the?
I don't think it was the issue with the polling methods, but rather with the interpretation of the polls.  There is a big level of complexity between the poll results and the prediction of electoral college votes, and that's the part that 538 got right by using sophisticated statistical methods, and most others didn't even realize they had to think about.
I thought that the polls themselves got things wrong in 2016 (in much the same way as they did in the UK in, I think, 2015), I think they made mistakes in the adjustments they were making or something along those lines, no?

But all the polling companies have been working to adjust and update their methodology since then. In the UK I think the pollstsers all collectively commissioned an independent academic review to identify the flaws and produce recommendations. I've not read of anything similar by the US pollsters but I imagine they will all have tried to fix their mistakes.
Let's bomb Russia!

Valmy

  • Neil
  • ***************
  • Posts: 49676
  • Champions du Monde
Re: US Elections 2020
« Reply #7099 on: September 20, 2020, 10:51:28 pm »
Again did I live through a different fucking 2016 than the rest of you did? Clinton had a tiny lead going into the election, like 3 points or something. It was hardly predicted to be a landslide and I was nervous as fuck. And then we had the FBI re-opening the Email investigation right before the election at that. Feel free to go back and look at the posts right before the election if you are misremembering this. Now the polls did flub a few states, like Wisconsin, but the problems with those polls were correctly pointed out by 538 before the election.

I just feel like the whole 'Polls were all TOTALLY WRONG' is just a piece of the Trump legend and Trump propaganda everybody has accepted over the actual events of 2016. I mean I was there following this shit pretty closely and at no point in the last few weeks leading up to the election did I see any evidence that made me think Clinton victory was certain.

It just infuriates me to no end seeing this kind of thing repeated when I know it was not true.
If we can hit that bull's-eye, the rest of the dominoes will fall like a house of cards. Checkmate!

Valmy is practically french. :frog:

Honorary gay award from Martinus

FunkMonk

  • Can I Be: Ottoman Empire
  • ***********
  • Posts: 6717
  • Donald the Carpathian
Re: US Elections 2020
« Reply #7100 on: September 20, 2020, 10:53:36 pm »
Many US pollsters now weight for education for the many state and national polls we're seeing now, which was a big miss back in 2016. Non-college voters heavily tilt toward Republicans and college-educated voters tend to tilt to Democrats now.

The other thing was that several of the Midwest states that Trump won were barely polled back in 2016 (looking at Wisconsin). The assumption was those states were safe Clinton. Few data points to draw from. Whereas now all those states have been polled and polled and polled into the ground.

Pollsters largely got things right for the 2018 midterms, iirc.
Person. Woman. Man. Camera. TV.

DGuller

  • Blessed by Valmy
  • **************
  • Posts: 22904
Re: US Elections 2020
« Reply #7101 on: September 20, 2020, 11:42:10 pm »
Again did I live through a different fucking 2016 than the rest of you did? Clinton had a tiny lead going into the election, like 3 points or something. It was hardly predicted to be a landslide and I was nervous as fuck. And then we had the FBI re-opening the Email investigation right before the election at that. Feel free to go back and look at the posts right before the election if you are misremembering this. Now the polls did flub a few states, like Wisconsin, but the problems with those polls were correctly pointed out by 538 before the election.

I just feel like the whole 'Polls were all TOTALLY WRONG' is just a piece of the Trump legend and Trump propaganda everybody has accepted over the actual events of 2016. I mean I was there following this shit pretty closely and at no point in the last few weeks leading up to the election did I see any evidence that made me think Clinton victory was certain.

It just infuriates me to no end seeing this kind of thing repeated when I know it was not true.
I was there with you in 2016, if that helps things.  The polls were not wrong, Clinton did win by 2 points, which is more or less where the polls were.  The interpretation of the polls was fucked up by everyone except 538, because they did not account for state-by-state errors being correlated.

Sheilbh

  • Blessed by Valmy
  • **************
  • Posts: 21043
Re: US Elections 2020
« Reply #7102 on: September 21, 2020, 01:07:25 am »
Many US pollsters now weight for education for the many state and national polls we're seeing now, which was a big miss back in 2016. Non-college voters heavily tilt toward Republicans and college-educated voters tend to tilt to Democrats now.
Exactly - I thought it was something like that. And I think it was a very similar issue in the UK 2015 election.

Quote
I was there with you in 2016, if that helps things.  The polls were not wrong, Clinton did win by 2 points, which is more or less where the polls were.  The interpretation of the polls was fucked up by everyone except 538, because they did not account for state-by-state errors being correlated.
Including by the pollsters.

I've just looked it up and it was like the UK, three of the big pollsters jointly commissioned an independent review of what went wrong by the American Association of Public Opinion Research. I don't think the pollsters thought they weren't wrong. The big issues were state level polling failures and, which is very similar to the UK polling issues in 2015, a failure to deal with voters deciding late, a failure to weight college v non-college educated voters and Trump voters not revealing their preferences.

As I say I think the polling industry have done a lot of work to fix/improve their polls since 2016 - no doubt this'll happen again, but probably not in 2020.
Let's bomb Russia!

The Minsky Moment

  • Octogon Champion
  • *************
  • Posts: 13213
  • Requiem for a Scheme
Re: US Elections 2020
« Reply #7103 on: September 21, 2020, 07:36:14 am »
Sheilbh you and Guller are talking about two different things.  You are talking about errors within state level polls and Guller is talking about using the polls to create a predictive model of who will win.  The polling companies try to improve their product (at some some do) and each major election brings new data points and information they can use to do that.  That's part of the process but to some degree you are fighting the last war because relevance and salience partisan characteristics can change over time. In any event, the 2016 results within each state were mostly within the sizable margin of error for individual state polls.  The election modelers who forecasted high 90s probabilities for HRC did not do so because of errors in the individual polls, but for the reason DGuller gave - they assumed that those errors were uncorrelated when in fact they were correlated.
The purpose of studying economics is not to acquire a set of ready-made answers to economic questions, but to learn how to avoid being deceived by economists.
--Joan Robinson

DGuller

  • Blessed by Valmy
  • **************
  • Posts: 22904
Re: US Elections 2020
« Reply #7104 on: September 21, 2020, 07:49:31 am »
Let's say you own a portfolio of 10 mortgages.  Each mortgage has a 20% probability of default.  What's the probability that all 10 mortgages default?

If you answered 0.00001024%, welcome to 2005.  With the information you have, the answer is somewhere between 0% and 20%.

celedhring

  • Octogon Champion
  • *************
  • Posts: 11978
  • On lockdown
Re: US Elections 2020
« Reply #7105 on: September 21, 2020, 09:24:30 am »
Yeah, the idea is that if the PA polls were wrong, because - say - they underrepresented certain white constituencies that would flip from Dem to GOP, this same error was probably replicated in MI and WI given some shared demographics. Most predictive models, besides 538's, didn't account for that.

The Brain

  • Neil
  • ***************
  • Posts: 30236
  • All man
Re: US Elections 2020
« Reply #7106 on: September 21, 2020, 09:40:54 am »
Is that how crappy their models were back then?
Women want me. Men want to be with me.

celedhring

  • Octogon Champion
  • *************
  • Posts: 11978
  • On lockdown
Re: US Elections 2020
« Reply #7107 on: September 21, 2020, 09:42:32 am »
Is that how crappy their models were back then?

As DGuller pointed out, that's the same logic that led to the subprime crisis, although back then there was a perverse incentive for adopting faulty logic - it was good for business.

The Minsky Moment

  • Octogon Champion
  • *************
  • Posts: 13213
  • Requiem for a Scheme
Re: US Elections 2020
« Reply #7108 on: September 21, 2020, 10:57:24 am »
The subprime error can be explained through the maxim that it is difficult to get a man to understand something when his salary depends upon his not understanding it.  Clearly there were people in the industry whp understood the problem with the assumption, it's no accident that a shop like Goldman Sachs was quick to trade out of positions when things started to go south.

The election error in 2016 is harder to explain as 538 which was a notably outlier that was quite explicit about their modeling and the reasons for it.  There are lots of reasons to think that state poll errors would be correlated - not just because of similarities in sample composition across state lines but also for the simple reason that even recent polls are retrospective and don't capture last minute swings.
The purpose of studying economics is not to acquire a set of ready-made answers to economic questions, but to learn how to avoid being deceived by economists.
--Joan Robinson

FunkMonk

  • Can I Be: Ottoman Empire
  • ***********
  • Posts: 6717
  • Donald the Carpathian
Re: US Elections 2020
« Reply #7109 on: September 21, 2020, 11:05:20 am »
Yeah late undecided voters I think broke for Trump by a good margin and that had a lot to do with his upset victory.

There are a lot fewer undecided voters this time around.
Person. Woman. Man. Camera. TV.