Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

One pollster’s explanation for why the polls got it wrong

vox.com

Post reply on HN
Print view View original

281–290 of 565 posts

Re: One pollster’s explanation for why the polls got it wrong

#281

Post by jungletime »

ZephyrBlu wrote: post #272
jungletime wrote: post #257

Dr.SHIVA Ayyadurai, MIT PhD. "Inventor of Email" Analyzed the results in Michigan and came to the conclusion that 69k votes were taken from Trump and given to Biden. His reasoning here: https://www.youtube.com/watch?v=Ztu5Y5obWPk&feature=youtu.be

Do you have a TL;DW? I would watch it all if it was a bit shorter or there were timestamps, but 1hr10min is a lot.

From my understanding:

He found that the more a republican a county is, the more republicans voted for Gop candidates, and against Trump. But only after the first 20% or so of the vote was in. In the first 20% of the vote Trump was doing better than the Gop candidates on the ticket.

Comes to the conclusion that an algorithm is syphoning votes from Trump to Biden. In a predictable linear fashion after the first 20% of the vote is counted. And flips more votes in more republican counties in an almost perfect linear way.

It doesn't make sense because Trump was polling better than the Gop candidates before the election, and was popular with republican voters.

The tallying software has a weighted vote feature. Where votes can be weighted for example 2:1 for one candidate over the other. Thinks that this was kicked in after 20% of the vote was in. Which would produce a linear drop that the graphs show.

Re: One pollster’s explanation for why the polls got it wrong

#282

Post by saeranv »

To me, this article highlights how much of a moving target weighting polls, and political prediction is. Shor here speaks about how Democrats get oversampled as both in 2016, and 2020 due to different levels of engagement and trust over the election:

>> It used to be that once you control for age and race and gender and education, that people who trusted their neighbors basically voted the same as people who didn’t trust their neighbors. But then, starting in 2016, suddenly that shifted. If you look at white people without college education, high-trust non-college whites tended toward [Democrats], and low-trust non-college whites heavily turned against us. In 2016, we were polling this high-trust electorate, so we overestimated Clinton. These low-trust people still vote, even if they’re not answering these phone surveys.

Is this shift partially because Trump and then QAnon have shifted Trump-voters trust in institutions - thus the voters that don't respond to pollsters are shifting as the Republican party becomes more radicalized?

Re: One pollster’s explanation for why the polls got it wrong

#283

Post by smnrchrds »

AnimalMuppet wrote: post #213
smnrchrds wrote: post #33

Earlier quoted context omitted.

> what does the general public get out of polling? The laypeople, not much. But for politically active people, not to mention political parties, polls are the best tool to determine where to do what. If you are way ahead in one race but neck and neck in another, it makes sense for parties and activists to focus their efforts and donations to the competitive race. There is never enough resources to focus on every race…

OK, but when some random pollster calls me, what do I get out of it if the parties can allocate their resources more effectively? That's not anywhere on my list of concerns. It lets you campaign more effectively? Great, go bother someone else.

If political parties see, through polls, that their platform is not resonating with the populous, they will change their platform and policies. Decisions about future direction of the parties are not made in a vacuum. Polls are one of the feedback methods determining them.

Re: One pollster’s explanation for why the polls got it wrong

#284

Post by Enginerrrd »

sgc wrote: post #60
losteric wrote: post #45

Earlier quoted context omitted.

> Basically, if I ever got a call from a number not in my contact list, I would never answer. I know I'm not the only one like this, but I don't have any numbers to back it up. huh... I have literally never met a single person that answers cold calls on personal lines, this was already SOP when I was a kid. Seriously, who's picking up those calls?

Anyone with a job that requires them to talk to people with unknown numbers calling. There is one bias right there. Probably people who don't get that many cold calls outside elections, which must fit some sort of demographic as well. But the better pollsters don't just call. They use various online methods as well. 538 lists the following methods, and rates a pollster higher if they use more of them: Live — Live tel…

>Anyone with a job that requires them to talk to people with unknown numbers calling.

Exactly this. I'm forced to answer a lot of calls even when I think it's 80% likely that it's a spam call. I own a business and need to keep my employees fed with work. My likelihood of landing a contract with a person drops by about half if I don't pick up the phone right away.

Re: One pollster’s explanation for why the polls got it wrong

#285

Post by darawk »

snowwrestler wrote: post #229
rayiner wrote: post #116

> I reached out to the smartest polling guy I know: David Shor, an independent data analyst who’s a veteran of the Obama presidential campaigns who formerly operated a massive web-based survey at Civis Analytics before leaving earlier this year Before being fired for pointing out that violent protests alienate people: https://nymag.com/intelligencer/2020/06/case-for-liberalism-... Schor’s firing is a tremendous examp…

Sadly typical of HN today that a topic that is interesting technically and politically neutral (the challenge of accurate polling) is derailed up top with an inaccurate culture war comment (there is no correlation between when a Dem condemned violent protest and whether they won their election). FWIW, Shor’s theory about poll respondents is shared by Robert Cahaly at Trafalgar, but Trafalgar polls missed badly too in…

> there is no correlation between when a Dem condemned violent protest and whether they won their election

You got a source on that? It's clear that democrats lost ground, despite the Republican party's abject failure to control the coronavirus. This should have been the easiest win in world history, and somehow, the Democrats barely escaped with the presidency. There is pretty clearly something to explain here.

Re: One pollster’s explanation for why the polls got it wrong

#286

Post by AnimalMuppet »

PeterisP wrote: post #262
free_rms wrote: post #243

Earlier quoted context omitted.

And when you consider that concessions to the lower class by both parties are entirely cultural, linguistic and non-material..

Yes, that's true, and because of that the lower class systematically vote for parties and policies that harm lower class to benefit the wealthy. Actually, this - class consciousness / false consciousness; the understanding (or lack of it) of the role and interests of the class you're in - was one of key concepts of Marx writings, too bad that this association currently makes many people automatically discard these id…

Marx's problem was that he thought of everything in terms of class. If I'm poor, and I know I'm poor, but I also think about something other than class when making political decisions (no matter what that "something" is), then I have "false consciousness". That is, if I don't think about this totally the way Marx says, I'm stupid. Nice.

If you think about your life that one-dimensionally, you're poor in a way that Marxism can't fix.

Re: One pollster’s explanation for why the polls got it wrong

#287

Post by euank »

DecoPerson wrote: post #261

“The kind of people who answer polls are really weird, and it’s ruining polling.” This is true everywhere. I’ve been involved polling in for a popular AAA game, a small retail business, school events, online forums, and more; they all fail to account for the “who chooses to answer” bias (is there a name for this?). I speculate that the “average 4.7 million viewers“ rating Firefly received when it aired was totally of…

> is there a name for this bias?

I think it's response bias. https://en.wikipedia.org/wiki/Response_bias

Re: One pollster’s explanation for why the polls got it wrong

#288

Post by darawk »

boomboomsubban wrote: post #220
rayiner wrote: post #116

> I reached out to the smartest polling guy I know: David Shor, an independent data analyst who’s a veteran of the Obama presidential campaigns who formerly operated a massive web-based survey at Civis Analytics before leaving earlier this year Before being fired for pointing out that violent protests alienate people: https://nymag.com/intelligencer/2020/06/case-for-liberalism-... Schor’s firing is a tremendous examp…

So you're defending the person who is trying to explain why polls got it wrong by using polling data to show that their argument belittling the BLM protests was correct. This is ridiculous on it's own, but also ignores things like how wrong it is to blame the protests for violence often instigated by the police, how dumb it is to trust polling data from the Nixon campaign when we know he was involved in all sorts of…

> So you're defending the person who is trying to explain why polls got it wrong by using polling data to show that their argument belittling the BLM protests was correct.

He's not belittling them. He's supporting them and their cause, and offering course corrective advice so that they are better able to achieve their stated goals.

> This is ridiculous on it's own, but also ignores things like how wrong it is to blame the protests for violence often instigated by the police

Violence was instigated on both sides, but regardless of what the underlying reality was, the perception was that leftists were responsible for much of the violence, and that is problematic electorally.

The Twitter thread you linked is exactly why the Democrats lost power this cycle. This thinking, right here:

> And for those pointing out that Dems may have lost some House seats in conservative districts, at least in part due to BLM: If you define your party based on what you think might work in one district in Oklahoma, you're doing it wrong.

You may not like the fact that you have to think about the median Oklahoman voter, but the reality is that that voter is much more important than the median say, NYC voter. Because you already have NYC in the bag. Oklahoma, on the other hand, is up for grabs. Whether or not this system is how things ought to be, this is how it actually is.

Re: One pollster’s explanation for why the polls got it wrong

#289

Post by nl »

karpierz wrote: post #263
nl wrote: post #252

Earlier quoted context omitted.

While it's true that the average salary of people who complete a college degree is higher than those who don't there are large numbers who don't complete the degree (which ends up with little salary benefit), or who do degrees that don't provide great salary outcomes[1] [1] https://www.bls.gov/careeroutlook/2020/data-on-display/educa... - note the associate degree and "some college" levels. > He also pledged to fund…

> The stockmarket is a comparative market where higher profitability compared to other companies is what affects stock price. Can you provide evidence for this? I was under the impression that price is dictated by long-term underlying company value. That is to say, time-discounted, risk-adjusted dividends, including buy-outs. Dividends are just paid out via stock-buybacks these days because the dividends are taxed mo…

I was under the impression that price is dictated by long-term underlying company value. That is to say, time-discounted, risk-adjusted dividends, including buy-outs.

It's true that price is determined by long-term value, but long-term value isn't sorely dictated by dividends. Expected price appreciation itself is a factor for growth stocks. Growth stocks include some very large companies, including Google and Facebook.

Earnings is seen as an indication of future value (and future dividends if it is an income stock rather than a growth stock)

https://finance.zacks.com/profits-reflect-stock-prices-8263....

Re: One pollster’s explanation for why the polls got it wrong

#290

Post by nl »

nl wrote: post #208
kjeetgill wrote: post #188

Earlier quoted context omitted.

Do you think the bulk of voters for Trump believe one of these things? He won a considerable number of votes.

I think a large proportion of Trump voters believe parts of QAnon-pushed conspiracy theories - especially that he is "taking on a Deep State conspiracy", and data indicates some support for that[1][2]. Surely they don't believe the "blood drinking paedophile" part though? It seems outrageous, and yet I know a bunch of people on HN believed PizzaGate back in 2016, so maybe I'm underestimating it. [1] https://www.usato…

The downvotes on this are very aggressive and I'd love to know why. Do people think my summary of the data is wrong or that I'm underestimating the depth of it?
Post reply on HN