Register a SA Forums Account here!
JOINING THE SA FORUMS WILL REMOVE THIS BIG AD, THE ANNOYING UNDERLINED ADS, AND STUPID INTERSTITIAL ADS!!!

You can: log in, read the tech support FAQ, or request your lost password. This dumb message (and those ads) will appear on every screen until you register! Get rid of this crap by registering your own SA Forums Account and joining roughly 150,000 Goons, for the one-time price of $9.95! We charge money because it costs us money per month for bills, and since we don't believe in showing ads to our users, we try to make the money back through forum registrations.
Who will you vote for in 2020?
This poll is closed.
Biden 425 18.06%
Trump 105 4.46%
whoever the Green Party runs 307 13.05%
GOOGLE RON PAUL 151 6.42%
Bernie Sanders 346 14.70%
Stalin 246 10.45%
Satan 300 12.75%
Nobody 202 8.58%
Jess Scarane 110 4.67%
mystery man Brian Carroll of the American Solidarity Party 61 2.59%
Dick Nixon 100 4.25%
Total: 2089 votes
[Edit Poll (moderators only)]

 
  • Post
  • Reply
punishedkissinger
Sep 20, 2017

Jarmak posted:

That wasn't his stance, his "stance" was it was safe to vote, and IIRC his campaign never took an official position on the issue so we're picking at a couple of comments made in tv interviews at the start of the controversy that they never seriously promoted as a position.

It was a bad stance, dangerous even, but never was the stance of the Biden campaign "people should die to vote for me", that's you're opinion of the implications of his stance.

Moreover it was never reality that he actually did anything to make that happen. He had nothing to do with the decision, the opposite decision was made, and the GOP are the ones that made the vote happen.

When he lied and said it was safe to vote, he was explicitly supporting that position. It's kind of ridiculous to say it wasn't the campaign's position when he is the person in charge of the campaign.

Adbot
ADBOT LOVES YOU

VitalSigns
Sep 3, 2011

Biden simply tricked people into going out to die for his political ambitions, but he never not once went to anyone's house and forced them to the polls at gunpoint.

Well that's fine then!

Cpt_Obvious
Jun 18, 2007

Jarmak posted:

That wasn't his stance, his "stance" was it was safe to vote, and IIRC his campaign never took an official position on the issue so we're picking at a couple of comments made in tv interviews at the start of the controversy that they never seriously promoted as a position.

It was a bad stance, dangerous even, but never was the stance of the Biden campaign "people should die to vote for me", that's you're opinion of the implications of his stance.

Moreover it was never reality that he actually did anything to make that happen. He had nothing to do with the decision, the opposite decision was made, and the GOP are the ones that made the vote happen.

1. If Joseph Biden does not count as a representative of his own campaign, then the concept of representative politics itself is a fallacy.

2. If you tell people to fly a kite during a hurricane, you are partially responsible if they get hurt.

3. It is an incredibly bad faith argument to claim that public officials encouraging activity does not contribute to it. Democrats have spent years howling that Trump encouraging racists has contributed to hate crimes. That sword cuts both ways.

punishedkissinger
Sep 20, 2017

Sure, he took the same stance as the GOP on the issue, but he didn't run any campaign ads on it therefore

VitalSigns
Sep 3, 2011

Compare with: Trump just told people to take hydroxychloroquine and then his bad medical advice killed people, but he never held anyone down and forced pills into them, so no concerns about his judgment here.

Yeowch!!! My Balls!!!
May 31, 2006

Jarmak posted:

That wasn't his stance, his "stance" was it was safe to vote, and IIRC his campaign never took an official position on the issue so we're picking at a couple of comments made in tv interviews at the start of the controversy that they never seriously promoted as a position.

It was a bad stance, dangerous even, but never was the stance of the Biden campaign "people should die to vote for me", that's you're opinion of the implications of his stance.

Moreover it was never reality that he actually did anything to make that happen. He had nothing to do with the decision, the opposite decision was made, and the GOP are the ones that made the vote happen.

oh, his campaign was taking a very solid position internally. remote fundraisers, no public interviews. can't risk the septuagenarian catching the bug. this was a thing that was known, respected, and feared in the Joe Biden presidential campaign.

but ask Joe Biden what the... well, what would be a good word for them? they're like people, but, you know. less important. oh well something will come to me. ask Joe Biden what -they- should do, when the choice is between preserving their lives and preserving his power, and he doesn't loving blink.

all that poo poo being done to preserve HIS life is good, noble and proper.

Those fuckers out there? meh. what does the world lose if a couple thousand of them die

Cpt_Obvious
Jun 18, 2007

At this point, there is no moral defense of Democrats. Perhaps there is a tactical argument to be made, but at no point can any Democrat claim a moral justification for supporting Joe Biden.

Jarmak
Jan 24, 2005

punishedkissinger posted:

When he lied and said it was safe to vote, he was explicitly supporting that position. It's kind of ridiculous to say it wasn't the campaign's position when he is the person in charge of the campaign.

It's your opinion that he lied, and that's literally the opposite of what explicit means. That's not grammar pendantry, the literal point of the arguement I'm making is that it's dishonest to present implicit statements (which rely on analysis and opinion to interpret, and must be defended with evidence) as explicit (which is literally what happened and agreed upon facts).

It's also disingenuously reductive to act like there isn't a difference in something said off the cuff during an interview and the campaign explicitly pushing something as an explicit position. Particularly when talking about a politician especially famous for saying dumb poo poo because he says whatever he's thinking instead of stopping to think it through.

Cpt_Obvious
Jun 18, 2007

Jarmak posted:

It's your opinion that he lied, and that's literally the opposite of what explicit means. That's not grammar pendantry, the literal point of the arguement I'm making is that it's dishonest to present implicit statements (which rely on analysis and opinion to interpret, and must be defended with evidence) as explicit (which is literally what happened and agreed upon facts).

It's also disingenuously reductive to act like there isn't a difference in something said off the cuff during an interview and the campaign explicitly pushing something as an explicit position. Particularly when talking about a politician especially famous for saying dumb poo poo because he says whatever he's thinking instead of stopping to think it through.

Quick! I've been disproven! Call for help!

Up in the sky! It's a bird! It's a plane! No! It's a meaningless pedantic argument that addresses nothing!

Hooray! I'm saved!

Pick
Jul 19, 2009
Nap Ghost
Implicit vs. explicit positions are major distinctions in an environment as sensitive as a campaign.

Marxalot
Dec 24, 2008

Appropriator of
Dan Crenshaw's Eyepatch

Jarmak posted:

It's your opinion that he lied, and that's literally the opposite of what explicit means. That's not grammar pendantry, the literal point of the arguement I'm making is that it's dishonest to present implicit statements (which rely on analysis and opinion to interpret, and must be defended with evidence) as explicit (which is literally what happened and agreed upon facts).

It's also disingenuously reductive to act like there isn't a difference in something said off the cuff during an interview and the campaign explicitly pushing something as an explicit position. Particularly when talking about a politician especially famous for saying dumb poo poo because he says whatever he's thinking instead of stopping to think it through.

"Sure he may have said something off the cuff that may have led to a large amount of people getting seriously ill or dying, but he's also really stupid and people should know better than to listen to him or his campaign surrogates" isn't the defense of Biden that you think it is.

It's seriously loving damning.

e:
VVVVV It is technically not a literal crime within the United States of America to tell people that polling places are safe during a pandemic. You really got em!

Marxalot fucked around with this message at 17:11 on Jun 5, 2020

evilweasel
Aug 24, 2002

could someone fill me in on the full "biden is a criminal for getting people to vote in wisconsin" theory. what was the crime; what should have been done; etc.

there's too much referencing an assumed crime everyone knows about that's not actually said. assume i know nothing whatsoever about this argument outside the single post responding to this one, even if someone suggested part of it a few posts ago.

Mellow Seas
Oct 9, 2012
Probation
Can't post for 10 years!
BREAKING NEWS: BIDEN BAD

Oh, did I saw "breaking news"? I meant "tired argument".

Let's look at some numbers. Polling from Research Co.:

https://twitter.com/Politics_Polls/status/1268908918052438017

https://twitter.com/Politics_Polls/status/1268911914542907392

https://twitter.com/Politics_Polls/status/1268913465453920256

The success with which Republicans propagandized the nation to the point where the Republican candidate *always* leads on "the economy" and security is staggering. I guess it's at least good that Trump is well under 50% on those measures.

Polls also out today showing Trump +1 in IA and a tie in Texas. Generally speaking I'm still pretty drat sure Trump will win those states.

Cpt_Obvious
Jun 18, 2007

evilweasel posted:

could someone fill me in on the full "biden is a criminal for getting people to vote in wisconsin" theory. what was the crime; what should have been done; etc.

there's too much referencing an assumed crime everyone knows about that's not actually said. assume i know nothing whatsoever about this argument outside the single post responding to this one, even if someone suggested part of it a few posts ago.

There is nothing technically criminal with encouraging people to die. It's just loving horrendous, and indicative of how he would have handled the pandemic.

evilweasel
Aug 24, 2002

Cpt_Obvious posted:

There is nothing technically criminal with encouraging people to die. It's just loving horrendous, and indicative of how he would have handled the pandemic.

this doesn't give me any useful information. what exactly is it you think was wrong; what should have been done instead. i'm actually asking because i've just skimmed over all of this and feel like actually digging into whatever crazy rabbit hole this is.

Cpt_Obvious
Jun 18, 2007

Pick posted:

Implicit vs. explicit positions are major distinctions in an environment as sensitive as a campaign.

In no way does it address any of the concerns that the previous poster raised. It is a red herring meant to distract from a losing argumentative position.

Cpt_Obvious
Jun 18, 2007

evilweasel posted:

this doesn't give me any useful information. what exactly is it you think was wrong; what should have been done instead. i'm actually asking because i've just skimmed over all of this and feel like actually digging into whatever crazy rabbit hole this is.

Ok, long story short, there's a pandemic going on at the same time as a major presidential primary. People are told not to go out and gather in large groups. Biden tells people it is safe to go vote in person. This is a terrible thing to do because he is encouraging voters, who mostly includes the largest at risk group of people, to endanger their lives so that he can cement a victory he doesn't need.

He should have told people to stay home. He didn't need to win Wisconsin so badly that people should die for it.

Pick
Jul 19, 2009
Nap Ghost
Are you suggesting Sanders's blowout in Wisconsin was because he told his voters to stay home?

Cpt_Obvious
Jun 18, 2007

Pick posted:

Are you suggesting Sanders's blowout in Wisconsin was because he told his voters to stay home?

There it is. Another red herring combined with a straw man.

We are talking about the heinous nature of encouraging your voters to die. It is an understandably indefensible position. Maybe you should abandon it.

evilweasel
Aug 24, 2002

Cpt_Obvious posted:

Ok, long story short, there's a pandemic going on at the same time as a major presidential primary. People are told not to go out and gather in large groups. Biden tells people it is safe to go vote in person. This is a terrible thing to do because he is encouraging voters, who mostly includes the largest at risk group of people, to endanger their lives so that he can cement a victory he doesn't need.

He should have told people to stay home. He didn't need to win Wisconsin so badly that people should die for it.

is this it? because there's a major flaw in the logic. specifically, both the biden and bernie campaigns wanted bernie to stay in precisely because that election was very important: one of the wisconsin supreme court seats was up for election on the same ballot (the wisconsin supreme court had a 5-2 republican majority, and those republicans make the SCOTUS conservatives look tame) and both campaigns wanted to maximize democratic turnout in order to win that election.

which they did, and thanks to one of the remaining 4 conservatives having a tiny conscience on this specific voting rights issue it is possible that the GOP's attempt to throw hundreds of thousands of democrats off the voter rolls will be blocked (it depends on if the republicans can fast-track a decision before August 1). that offers at least the chance of helping to unfuck wisconsin in the next few years (though democrats already blew one of these seats post-trump's election).

that election was tremendously, tremendously important. it was why bernie stayed in until after wisconsin, but quit before the results were announced.

also, does this theory castigate bernie for continuing to campaign to get people to turn out and vote for him, knowing that he wasn't actually still competing?

Yeowch!!! My Balls!!!
May 31, 2006

evilweasel posted:

this doesn't give me any useful information. what exactly is it you think was wrong; what should have been done instead. i'm actually asking because i've just skimmed over all of this and feel like actually digging into whatever crazy rabbit hole this is.

you are asked "is it safe to hold a primary in the middle of a pandemic" on national television.

you, personally, are someone whose campaign has gone to considerable lengths to insulate you from a pandemic. your schedule has been completely reorganized to minimize personal interaction. however, your polling is pretty good at the moment, and for whatever reason there is considerable fear that a delay might hurt that position.

do you respond "the science says it's okay to go out there and vote"

if no, congratulations: you are less willing to kill people to preserve your own power than Joseph Robinette Biden.

Jarmak
Jan 24, 2005

evilweasel posted:

this doesn't give me any useful information. what exactly is it you think was wrong; what should have been done instead. i'm actually asking because i've just skimmed over all of this and feel like actually digging into whatever crazy rabbit hole this is.

This was the statement being represented as "Joe Biden said people should die to vote for him":

Joe Biden posted:

“A convention having tens of thousands of people in one arena is very different than having people walk into a polling booth with accurate spacing with 6 to 10 feet apart, one at a time going in, and having the machines scrubbed down,” Biden said. “I think you could hold the election as well dealing with mail-in ballots and same-day registration. I think it could be done… but that’s for them to decide.”

Cpt_Obvious
Jun 18, 2007

Jarmak posted:

This was the statement being represented as "Joe Biden said people should die to vote for him":

This is it. Thanks for digging up the lie for me.

It is a heinous disregard for human life and safety.

Pick
Jul 19, 2009
Nap Ghost
That's it?!?!?!?!

You've grossly misrepresented what was said.

Condiv
May 7, 2008

Sorry to undo the effort of paying a domestic abuser $10 to own this poster, but I am going to lose my dang mind if I keep seeing multiple posters who appear to be Baloogan.

With love,
a mod


Jarmak posted:

This was the statement being represented as "Joe Biden said people should die to vote for him":

this is him literally lying about the safety of polling places in the middle of an epidemic in order to keep turnout up

Yeowch!!! My Balls!!!
May 31, 2006

evilweasel posted:

that election was tremendously, tremendously important. it was why bernie stayed in until after wisconsin, but quit before the results were announced.

also, does this theory castigate bernie for continuing to campaign to get people to turn out and vote for him, knowing that he wasn't actually still competing?

HELL yeah. that's the liberal philosophy I know and love.

sure, we signed off on killing a bunch of people who were stupid enough to trust we wouldn't sacrifice -them-, but we killed them for a good cause. their reward will be great in Democrat Heaven, provided they pass the means test at the pearly gates.

incidentally, the bernie campaign was in fact calling for the primaries to be delayed on health grounds. one could, of course, note that this was just as much a strategic move to stall for time as anything else, and that its saving a couple thousand lives if implemented was just icing on the cake.

evilweasel
Aug 24, 2002

Mellow Seas posted:

BREAKING NEWS: BIDEN BAD

Oh, did I saw "breaking news"? I meant "tired argument".

Let's look at some numbers. Polling from Research Co.:

has fivethirtyeight or someone else started doing some analysis on how useful/predictive polls this far out generally are? i recall faintly there's some line 538 identified as when polls start being at least somewhat predictive and i thought it was basically "after the nominations are secured" but I can't find anything on the subject.

Mellow Seas
Oct 9, 2012
Probation
Can't post for 10 years!

evilweasel posted:

has fivethirtyeight or someone else started doing some analysis on how useful/predictive polls this far out generally are? i recall faintly there's some line 538 identified as when polls start being at least somewhat predictive and i thought it was basically "after the nominations are secured" but I can't find anything on the subject.

Great question!

I was able to find a 538 piece from summer '19 that just pretty much said that polls 18 months before an election aren't predictive, which is unsurprising. But it didn't talk about when they do become predictive.

I managed to find an Economist article that cites a paper by a couple of political scientists - the article and the paper are paywalled, so I can't really look at their methodology, but the intro of the Economist article does pinpoint "close of summer of the election year" as the point where the polls become very accurate.

Mean Paywalling Economist posted:

According to number-crunching from Christopher Wlezien and Robert Erikson, political scientists at the University of Texas and Columbia University, pre-election polls make for poor predictors until the close of summer in an election year, by which point both parties have held their nominating conventions. In their book “The 2012 Campaign and the Timeline of Presidential Elections”, they explain that candidates’ standing in the polls fails to account for even half of the variance in their eventual vote margins until the spring before the election.

So, according to this analysis... which we can't read... the polls are absolutely significant now, but not terribly predictive. I wonder about how "after the convention" compares to "after candidates are set" as a standard, but I don't have enough information to say one way or the other.

E: I also don't know if these numbers have shifted over time (which seems possible because of increased polarization). I figure they cover it in the paper, and The Economist's summary is speaking about current-day conditions.

Mellow Seas fucked around with this message at 17:59 on Jun 5, 2020

RBA Starblade
Apr 28, 2008

Going Home.

Games Idiot Court Jester

Jarmak posted:

This was the statement being represented as "Joe Biden said people should die to vote for him":

"Choking to death on their own blood" was the phrase from the last thread. I've actually never seen that whole quote before, I had at least believed people when they were saying he was telling his voters to go vote and ignore the risk of infection. That's an insane misreading of that.

Cpt_Obvious
Jun 18, 2007

evilweasel posted:

is this it? because there's a major flaw in the logic. specifically, both the biden and bernie campaigns wanted bernie to stay in precisely because that election was very important: one of the wisconsin supreme court seats was up for election on the same ballot (the wisconsin supreme court had a 5-2 republican majority, and those republicans make the SCOTUS conservatives look tame) and both campaigns wanted to maximize democratic turnout in order to win that election.

which they did, and thanks to one of the remaining 4 conservatives having a tiny conscience on this specific voting rights issue it is possible that the GOP's attempt to throw hundreds of thousands of democrats off the voter rolls will be blocked (it depends on if the republicans can fast-track a decision before August 1). that offers at least the chance of helping to unfuck wisconsin in the next few years (though democrats already blew one of these seats post-trump's election).

that election was tremendously, tremendously important. it was why bernie stayed in until after wisconsin, but quit before the results were announced.

also, does this theory castigate bernie for continuing to campaign to get people to turn out and vote for him, knowing that he wasn't actually still competing?

1. This is a shamefully Pyrrhic argument. If you truly believe that judge appointments are worth actual human lives, you need to seriously reevaluate your position.

2. This is disingenuous argument. You are casually ignoring the fact that Joe Biden is also running for president and asking people to vote for him because it is detrimental to your position, which is, frankly, absurd.

3. As mentioned above Bernie tried to delay the primary, because within his chest he has a soul unlike the rotting avatar of corruption that you are trying to defend.

evilweasel
Aug 24, 2002

Mellow Seas posted:

Great question!

I was able to find a 538 piece from summer '19 that just pretty much said that polls 18 months before an election aren't predictive, which is unsurprising. But it didn't talk about when they do become predictive.

I managed to find an Economist article that cites a paper by a couple of political scientists - the article and the paper are paywalled, so I can't really look at their methodology, but the intro of the Economist article does pinpoint "close of summer of the election year" as the point where the polls become very accurate.


So, according to this analysis... which we can't read... the polls are absolutely significant now, but not terribly predictive. I wonder about how "after the convention" compares to "after candidates are set" as a standard, but I don't have enough information to say one way or the other.

So if I understand that correctly, beginning in the spring of this year the polls start to "account for half the variance" in the eventual vote margins (of course, I have no idea what that means); but they are "poor" predictors until I guess end of August. So it's probably pre-March polls were utterly worthless (which makes sense; you would expect head to head polls to be not very predictive while the nominations are not set); polls start being somewhat predictive after nominations are set but before the conventions, and after the conventions people are "paying attention" and have most of their votes locked in.

So the polls probably mean something right now, much more than from a few months ago, but not "take it to vegas and put all your money on the leader" level. But once we get post-august polls we will start to have as high a degree of confidence as we're gonna get.

Do you have the actual economist link so I can see if I can get the article?

Mellow Seas
Oct 9, 2012
Probation
Can't post for 10 years!

evilweasel posted:

Do you have the actual economist link so I can see if I can get the article?

https://www.economist.com/democracy-in-america/2019/07/26/when-to-pay-attention-to-2020-forecasts

Yeowch!!! My Balls!!!
May 31, 2006

RBA Starblade posted:

"Choking to death on their own blood" was the phrase from the last thread. I've actually never seen that whole quote before, I had at least believed people when they were saying he was telling his voters to go vote and ignore the risk of infection. That's an insane misreading of that.

was it safe to go out and vote in the middle of a pandemic.

and what did Joe Biden tell you you should go do anyway.

once upon a time, it was possible to believe that the Democrats were just willing to kill other people to preserve their political power. Joe Biden has told you, to your face, that he's willing to kill you too. the white liberal who votes blue no matter who is just as pleasing a blood sacrifice to him as the Hispanic children hurled into dog cages for the crime of their blood-taint and the black men shot by police for using the same drugs as his son.

at least the evangelicals think they get something in return for their martyrdom.

How are u
May 19, 2005

by Azathoth

Mellow Seas posted:


The success with which Republicans propagandized the nation to the point where the Republican candidate *always* leads on "the economy" and security is staggering. I guess it's at least good that Trump is well under 50% on those measures.


It really is something, isn't it? I'm very confident that this is something that entirely exists within the minds of Boomers who wish they were still living in 1983 when Reagan told them it was cool and good to be bigots and their dicks still worked.

There's no way that millenials buy into this poo poo.

Majorian
Jul 1, 2009

How are u posted:

It really is something, isn't it? I'm very confident that this is something that entirely exists within the minds of Boomers who wish they were still living in 1983 when Reagan told them it was cool and good to be bigots and their dicks still worked.

There's no way that millenials buy into this poo poo.

They don't, but I think a lot of younger generations also don't buy into older conceptions of "handling the economy" either. A lot of millennials came of age at a time when unemployment numbers were low and GDP growth was high, and none of that meant squat to our job prospects and daily lives. Older millennials like me got out of college or grad school when the '07-'8 recession hit and have experienced first-hand how there really was no recovery from that for a lot of Americans. So when we see job numbers like the ones Trump is touting today, we don't even need to hear reports that the DOL hosed up the math to know that it's meaningless bullshit that doesn't actually represent facts on the ground.

evilweasel
Aug 24, 2002

Cpt_Obvious posted:

1. This is a shamefully Pyrrhic argument. If you truly believe that judge appointments are worth actual human lives, you need to seriously reevaluate your position.

2. This is disingenuous argument. You are casually ignoring the fact that Joe Biden is also running for president and asking people to vote for him because it is detrimental to your position, which is, frankly, absurd.

3. As mentioned above Bernie tried to delay the primary, because within his chest he has a soul unlike the rotting avatar of corruption that you are trying to defend.

in order:

1) that's not what "pyrrhic argument" means (if it means anything at all), and you yourself were the one who framed this argument about do people "need" the win. yeah, as a factual matter wisconsin needed that win, it was very important. did they need it enough to ask people to vote in a pandemic? well, that is a hard, complex question. when making the moral judgements, you need to recall democrats were dealing with an intentional republican strategy to try to depress turnout to win that election, so the moral judgements on the democratic response pale in comparison to what the republicans did. the problem is, that is a tough question: how much can and should you encourage people to risk their safety to exercise their right to vote? it doesn't just come up in pandemics, of course. but when you ask that tough question this ceases being the sort of gotcha, biden bad!!!!! argument you want it to be.

2) this doesn't make any sense at all. as best i can tell, you are trying to argue "well, what about joe biden wanting to win wisconsin???" he didn't really care, because the bernie campaign had already promised to drop out as soon as it was over. which they did.

3) that's dodging the question. and this question is important because it's forcing you to reason through your argument with the motivated reasoning changed. wisconsin democrats attempted to move the election, and it was blocked. knowing that, knowing that the election would not be moved, bernie sanders campaigned in wisconsin and encouraged people to vote for him, knowing their votes were pointless and that he had already agreed to drop out (and deliberately withheld that information from his voters). he did so for the same reason: driving up turnout in the wisconsin election, for the purpose of trying to win that judicial race.

what is your moral judgement on that decision, and why? it's a very similar decision, but with your motivated reasoning reversed such that you want to come out with the argument it's not bad instead of you want to come out with the argument it's bad. if you arrive at "both bernie and biden were bad" that's at least consistent; i still disagree with you but you bit the bullet and accepted the results you don't like about your logic and it clearly has some internal consistency, and we can discuss from there. if you arrive at "bernie good, biden bad" there might be a good reason - but i'd want to see it, because I don't see one, and i think your rationale for why bernie is not bad in this situation would be the very rebuttal to the "why biden bad" argument

evilweasel
Aug 24, 2002


looks like you can sign up with just an email account for five free articles. unfortunately, there is no useful information beyond what you quoted.

quote:

When to pay attention to 2020 forecasts
Polls tend not to predict with any accuracy until the summer of an election year

American politics
Democracy in America
Jul 26th 2019
BY G.E.M. | WASHINGTON, DC

PRESIDENT DONALD TRUMP is going to lose his re-election bid next year—or so recent surveys appear to suggest. A poll conducted between July 21st and 23rd on behalf of The Economist by YouGov, a pollster, found that Mr Trump would lose the national popular vote to a Democratic opponent by seven percentage points. A margin that large would be more than enough for a Democratic candidate such as Joe Biden, the former vice-president, or Elizabeth Warren, a senator from Massachusetts, to coast to a landslide victory in the Electoral College. However, it is currently more than 15 months before the election, and numbers released this early cannot accurately predict how candidates will perform.

According to number-crunching from Christopher Wlezien and Robert Erikson, political scientists at the University of Texas and Columbia University, pre-election polls make for poor predictors until the close of summer in an election year, by which point both parties have held their nominating conventions. In their book “The 2012 Campaign and the Timeline of Presidential Elections”, they explain that candidates’ standing in the polls fails to account for even half of the variance in their eventual vote margins until the spring before the election. At 330 days before the contest—roughly December of the year before the election—polls show virtually no correlation to final election outcomes; today, 467 days before the 2020 election, they are even worse. One could make better predictions by flipping a coin than by looking at the polls.

Pollsters can blame a few factors for the errors in their early estimates. First, at this point in the campaign, the eventual candidates have not even been decided. It could be that Joe Biden will win the Democrats’ current primary contest, but such a bet is no sure thing. How can voters be expected to know for whom they will cast a ballot? It is also too early to know the eventual economic and political conditions of the election. By November 2020, the country could enter a recession, crushing Mr Trump’s odds of re-election. Or America could enter a new boom. It is also too far in advance of the election for most voters to be paying much attention to national politics.

Forecasters seeking to predict elections with ample lead time may turn to information other than polling for more accuracy. Moody’s Analytics, a financial company, compiles economic data to use in predicting election outcomes, for example. The best projections also make use of economic indicators and the president’s approval ratings—which, while poll-based, are steadier and more frequently measured than the head-to-head match-ups. Nonetheless, these methods are far from perfect. Moody’s forecasts predicted with 95% confidence that Hillary Clinton would win the 2016 presidential election, and the best-known political-science model that relies on so-called “fundamentals” such as the economy and incumbency overshot Mr Trump’s popular vote margin by nearly three percentage points.

Other scholars are devising new approaches for election prediction, but they are untested. Rachel Bitecofer, who studies public opinion at Christopher Newport University, has developed a method that eschews poll-based models in favour of state-level demographic data, such as how many white voters live in a state, and how far left or right it typically leans in presidential contests. She predicts that Mr Trump will lose his re-election bid in 2020. Her similarly-devised method performed worse than poll-based methods in 2018. Ms Bitecofer’s model explained 70% of the variance in congressional races—simply using the results from the last election would have predicted 95%—and it is doubtful that the relationships between demographic factors such as age and education and vote choice will remain constant from the mid-terms to 2020, as her model assumes. Her work also does not actually predict the chance Mr Trump will win next year, only his chances in every state. When asked for such a probability, Ms Bitecofer said it “wouldn’t be mathematically based”. Still, she is confident that there is “little to no chance” that Americans re-elect the president.

The 2016 election instilled in voters a deep scepticism of polling, but is also served as a lesson to forecasters to be careful about how confidently they convey their predictions. There is danger in certainty. Two weeks before Mr Trump’s election, James Comey, then director of the FBI, made public an investigation into Hillary Clinton. He later said he did so because he “was making decisions in an environment where [Ms] Clinton was sure” to become the next president, a sentiment he attributed to pre-election polling and forecasting.

Given the uncertainty of predictions made far in advance, why are they often taken seriously? The work of Nobel Prize-winning psychologist Daniel Kahneman and Amos Tversky (who would have also won the prize had he been alive when Mr Kahneman received it) provides some insight. In 1993, in a book chapter titled “Probabilistic Reasoning”, the two authors stipulate that improperly-calibrated forecasts are often accepted because they play on peoples’ biases. When predictions favour one side of the political aisle, as Ms Bitecofer’s and Moody’s currently happen to do, people who support that side are more likely to believe them.

Messrs Kahneman and Tversky found that people often give too little weight to the underlying chance—the “prior probability”—of an outcome, favoring instead the most recent snippets of hard data that they see. In this case, the prior probability probably favours Mr Trump more than many forecasters believe. Punters at PredictIt, a website where users can wager on political events, give Mr Trump a roughly 50-50 shot of victory next year. With an average job approval of 43%, according to numbers crunched by the data-journalism website FiveThirtyEight, he is only just short of turning out the coalition that brought him to victory in 2016 with 46% of the vote. But many Democratic partisans will succumb to the temptation to believe soothing forecasts rather than confront such uncomfortable risks.

The very opacity of complex statistical models can also create an illusion of validity, particularly when they come from a respected expert. But as George Box, a famous statistician, once said, “all models are wrong, but some are useful”. Forecasters and observers alike would do well to remember that.

edit: that said, the underlying book/paper they're talking about is available for free on amazon, which i am now reading

evilweasel fucked around with this message at 18:28 on Jun 5, 2020

evilweasel
Aug 24, 2002

here is the underlying data about how accurate polls are prior to an election. sadly, as i last took statistics well over a decade ago, your guess is as good as mine about what an adjusted r-squared of about .5 means for the value of current polling about 150 days out from the election



(image is taken directly from the book, the red line i added, "The 2012 Campaign and the Timeline of Presidential Elections (Chicago Shorts)" location 207 if you get it on the kindle store)

evilweasel fucked around with this message at 18:35 on Jun 5, 2020

Mellow Seas
Oct 9, 2012
Probation
Can't post for 10 years!

evilweasel posted:

here is the underlying data about how accurate polls are prior to an election. sadly, as i last took statistics well over a decade ago, your guess is as good as mine about what an adjusted r-squared of about .5 means for the value of current polling about 150 days out from the election



(image is taken directly from the book, the red line i added, "The 2012 Campaign and the Timeline of Presidential Elections (Chicago Shorts)" location 207 if you get it on the kindle store)

Well that's kind of fun. Thanks for looking it up.

I really enjoy the drop in accuracy ~100 days before the election, which I presume is because of convention bounces.

Adbot
ADBOT LOVES YOU

Jarmak
Jan 24, 2005

evilweasel posted:

here is the underlying data about how accurate polls are prior to an election. sadly, as i last took statistics well over a decade ago, your guess is as good as mine about what an adjusted r-squared of about .5 means for the value of current polling about 150 days out from the election



(image is taken directly from the book, the red line i added, "The 2012 Campaign and the Timeline of Presidential Elections (Chicago Shorts)" location 207 if you get it on the kindle store)

It means half the movement in the dependent variable can be ascribed to movement in the independent variable.

Which in context I think means we can start to feel comfortable the that vote will fall within the middle two quartiles of the polls.

edit: (alternatively it might mean you can double to margin of error for the polls to get the margin of error between the poll and the election.)

Jarmak fucked around with this message at 18:41 on Jun 5, 2020

  • 1
  • 2
  • 3
  • 4
  • 5
  • Post
  • Reply