Is this complete bull **** ? (Page 1/3)
82-T/A [At Work] MAY 20, 06:56 AM
I literally do not know anyone who says that Biden is doing a good job... and most of my friends are Democrats.

Biden is getting polling results that consistently show things like... Hispanic approval dropping to something insane like 20%, and black approval rating dropping to 28%, etc... etc... etc... and yet, 538 continues to show Biden above 41% approval rating. Like this one... approval among young people... 18 to 34, dropped to 27%: https://www.newsweek.com/jo...y-falls-poll-1708434
That's an absolutely HORRIBLE approval rating.

Is 538 just completely full of **** ? I don't understand their "weights..." maybe I'm just jaded, but they seem to be heavily relying on the polling that shows Biden over 43%... which is completely absurd.
olejoedad MAY 20, 07:32 AM
538 being a biased website?

Say it ain't so!

Google results from various political bias evaluating websites rate it at 'center left', that it reports factual information but uses 'loaded' words in an attempt to influence the reader.

So, if those websites rate it 'center left', my view would be that it is much farther left than that, due to the pervasive left bias of most internet 'rating' websites.

From my own experience, watching Senate committee hearings and then checking the various 'news' analysis of said hearing, I would rate them LEft wing, vs Mother Jones as LEFT wing.

YMMV
theBDub MAY 20, 11:24 AM
I think it's high. I think I see how it happened though.

Looking through the recent polls, it appears they adjust the true poll up or down depending on the target group polled:
https://projects.fivethirty...-rating/?cid=rrpromo

Now, I'd assume Likely Voters for the midterm are going to skew Republican. It seems 538 assumes the same thing, so if you look at LV, they appear to adjust the weight up 2% for the 538 number. I think this is flawed because people will say they're likely to vote, but won't actually show up on election day. Meaning, the lower number is a more accurate number than 538 is giving it. But fine, they have a methodology. Where I take bigger issue is they also seem to adjust Registered Voters up 1%. I don't think there is any logical reason to do that. They also adjust All Adults down 1%, but I don't think that really makes up for the other adjustment.

So I would think the real number is around 2% lower than Nate's number.
82-T/A [At Work] MAY 20, 12:09 PM

quote
Originally posted by theBDub:

I think it's high. I think I see how it happened though.

Looking through the recent polls, it appears they adjust the true poll up or down depending on the target group polled:
https://projects.fivethirty...-rating/?cid=rrpromo

Now, I'd assume Likely Voters for the midterm are going to skew Republican. It seems 538 assumes the same thing, so if you look at LV, they appear to adjust the weight up 2% for the 538 number. I think this is flawed because people will say they're likely to vote, but won't actually show up on election day. Meaning, the lower number is a more accurate number than 538 is giving it. But fine, they have a methodology. Where I take bigger issue is they also seem to adjust Registered Voters up 1%. I don't think there is any logical reason to do that. They also adjust All Adults down 1%, but I don't think that really makes up for the other adjustment.

So I would think the real number is around 2% lower than Nate's number.




It seems as though they are constantly adjusting the weights to keep it just above Trump's ratings... which seems to be a really important factor for a lot of people. Biden is below Trump's approval ratings right now, but Trump was under investigation by the FBI for committing treason, which we know now was a completely made-up story. Not sure how history will look at this... but it's frustrating that people are pulling out all the stops for Biden, but were totally against Trump. I know this isn't just "my perspective." Having worked in Government, I know that the left generally rules most agencies and departments, and slow-rolled everything Trump tried to push, while they expedited like flies on **** anything that Biden pushed out.

I know also from working in the market research industry (which dabbled in polling), that polls could be easily manipulated to get a desired results based on wording. I can't imagine that with everything going on... that Biden's polling is this high. Literally, we had none of these problems under Trump, but he was still viewed as an ******* , and his polling was exceptionally low. We have Biden, who can't even complete most sentences and reads the cut lines on the teleprompter, who's failed at literally everything his administration has attempted to fix... most of which was as a result of previous failures he committed... and his polling seems to match Trump's.
Hudini MAY 20, 09:20 PM
Remember the Hillary vs Trump polls? They all had her leading by 12% or so. But then if you looked at the details of who was polled it stated they polled 12% more Democrat voters. They told you the poll was skewed but nobody looked at the details. What you saw on the news was "Hillary leads by 12%". After the election they were all surprised how the polls had failed when the answer was there all along.
randye MAY 20, 10:07 PM

quote
Originally posted by 82-T/A [At Work]:

I literally do not know anyone who says that Biden is doing a good job...


Is 538 just completely full of **** ?.





Three answers:

1.) Yes. 538 is indeed full of **** .

2.) You can almost always find a poll somewhere to comport with whatever you want to believe.

3.) If you can't find a poll you want you can always get someone to conduct a poll to provide you with whatever narrative you want to push,
("push poll" pun intended).


Ok, what the hell. I'll toss in a fourth answer closely related to #1 just for good measure:

4.) Nate Silver is an idiot who should have stuck to doing baseball statistics.

Silver's claims to practice "analytical rigor" but, at a very basic level, he doesn't really know what a poll is. In his conception and demonstrated practice, polls of opinion and sentiment aren't a snapshot of the present as informed by the past, but rather a prescient view into the future. Not a measurement, but a prediction.
This is why political polls now routinely include such ridiculous qualifiers as "likely voters" and "registered voters"

[This message has been edited by randye (edited 05-21-2022).]

rinselberg MAY 21, 10:28 AM
Don't forget the "Hidden Trump Voter" phenomenon.




quote
Election outcomes can be difficult to predict. A recent example is the 2016 US presidential election, in which Hillary Clinton lost five states that had been predicted to go for her, and with them the White House. Most election polls ask people about their own voting intentions: whether they will vote and, if so, for which candidate. We show that, compared with own-intention questions, social-circle questions that ask participants about the voting intentions of their social contacts improved predictions of voting in the 2016 US and 2017 French presidential elections. Responses to social-circle questions predicted election outcomes on national, state and individual levels, helped to explain last-minute changes in people’s voting intentions and provided information about the dynamics of echo chambers among supporters of different candidates.


"Asking about social circles improves election predictions"
Galesic, et al; nature human behavior; February 26, 2018.
https://www.nature.com/articles/s41562-018-0302-y

[This message has been edited by rinselberg (edited 05-21-2022).]

82-T/A [At Work] MAY 21, 11:26 AM

quote
Originally posted by randye:

Silver's claims to practice "analytical rigor" but, at a very basic level, he doesn't really know what a poll is. In his conception and demonstrated practice, polls of opinion and sentiment aren't a snapshot of the present as informed by the past, but rather a prescient view into the future. Not a measurement, but a prediction.
This is why political polls now routinely include such ridiculous qualifiers as "likely voters" and "registered voters"





I want to give him the benefit of the doubt, but I realize that people sometimes are unable to control their ideology and emotions. I just got a grade back from one of my classes. I don't mean to sound like a piece of crap, but I ended up with an A- in one of my classes. I've always gotten a 4.0 in all of my degree seeking programs except my earliest college classes I took when I was 18 and had dropped out. This is for a law degree that I'm pursuing. The teacher took a ton of points off my final paper because I discussed Rowe v. Wade. I had written and submitted my paper at least 2 weeks before this whole leak happened. It was totally unintentional. I wasn't even being political about it... the paper was about privacy law. I only mentioned Rowe v. Wade because that court decision was the catalyst for what would become the Privacy Act of 1974... since the decision was based on the right of privacy. It's the only reason why I mentioned it. Teacher really took this badly... between the time I submitted my paper and she graded it, the whole leaked court decision happened. Ugh...



quote
Originally posted by rinselberg:

Don't forget the "Hidden Trump Voter" phenomenon.

"Asking about social circles improves election predictions"
Galesic, et al; nature human behavior; February 26, 2018.





Well, if I'm being really introspective... most of my friends are Democrats. Of my Democrat friends, none of them like Biden, but they hate Trump more. I don't know where these rabid / fervent supporters on Twitter come from, because I've never met a single one. I have friends that are borderline communist who even tell me every chance they get how Government is good, and the U.S. is responsible for the failure of Communist countries because of our embargoes, etc. They don't like Biden either. So I wouldn't be shocked if Biden gets primaried... I've never seen anything like that in my lifetime, but I guess we'll see.

None of my friends, except maybe one, would vote for Trump as their "first choice." They all prefer Ron DeSantis (which includes all the ones who aren't from Florida either). Me personally, I am even more committed to Trump. I know that may warrant an eye-roll... but Trump likely has everything already planned out. He remembers everything (like an elephant) that wronged him... every agency that stone-walled him, every department that tried to thwart his actions... and he's already got a plan to go in and totally reshape everything. I really like Ron DeSantis... but he would be coming in new, and likely would want to separate himself somewhat from Trump... which means he's going to have first term blues all over again.

Trump on the other hand, will have had 4 years to contemplate his "revenge," if you will. Not trying to be a jerk, but with the abject failure we've seen over the past 4 years, I don't think most Democrats are going to immediately jump on the "Trump is the end of the world" bandwagon again when literally things were quite good under his administration. You compare the last 1-1/2 years today, to Trump's... it's night and day difference ... success and failure.

So maybe you're right... but I really have no idea what's going on right now. The world is crazy.
randye MAY 21, 08:24 PM

quote
Originally posted by 82-T/A [At Work]:


I want to give him the benefit of the doubt, but I realize that people sometimes are unable to control their ideology and emotions.






I don't know where Silvers' personal politics are but, just like the two forum Leftists that have posted in this thread, Silver does clearly hold to the dumb idea that there is "predictive science" to opinion polls.

Leftists share that belief because like Herbert Marcuse, a Leftist philosopher of the Frankfurt School, who posited that if you understand economics, you can actually write down what will happen in the future with as much confidence as you write down the history of the past, they desperately cling to fortune telling pseudoscience as a replacement for their innate inability to objectively evaluate data simply for what it is and not for what they wish for it to be.

Silver's, Leftist's and the "media chatterati's", faith in polls as "predictive science", ( “you can actually write down what will happen in the future, with as much confidence as you write down the history of the past. Because it’s science!”), has been thoroughly debunked time and time again, (i.e. Silver's complete screw up of predicting the 2010 elections in the U.K.; a somewhat accurate prediction of the 2012 results in the U.S.; and then a wildly wrong prediction of the 2016 elections that he bizarrely continues to attempt to explain away by claiming that he was "slightly less overconfident about a Donald Trump loss than everybody else".

[This message has been edited by randye (edited 05-22-2022).]

theBDub MAY 23, 01:58 PM

quote
Originally posted by randye:
I don't know where Silvers' personal politics are but, just like the two forum Leftists that have posted in this thread, Silver does clearly hold to the dumb idea that there is "predictive science" to opinion polls.

Leftists share that belief because like Herbert Marcuse, a Leftist philosopher of the Frankfurt School, who posited that if you understand economics, you can actually write down what will happen in the future with as much confidence as you write down the history of the past, they desperately cling to fortune telling pseudoscience as a replacement for their innate inability to objectively evaluate data simply for what it is and not for what they wish for it to be.

Silver's, Leftist's and the "media chatterati's", faith in polls as "predictive science", ( “you can actually write down what will happen in the future, with as much confidence as you write down the history of the past. Because it’s science!”), has been thoroughly debunked time and time again, (i.e. Silver's complete screw up of predicting the 2010 elections in the U.K.; a somewhat accurate prediction of the 2012 results in the U.S.; and then a wildly wrong prediction of the 2016 elections that he bizarrely continues to attempt to explain away by claiming that he was "slightly less overconfident about a Donald Trump loss than everybody else".




It’s patently ridiculous to conflate the idea that opinion polls may give an indication of what to expect in an election with the idea that someone can predict the future based on the past. It’s a fun science-fiction concept (and was Asimov’s basis for the wonderful Foundation series), but it’s nothing more than that. I’ve never spoken with any person ever who has believed this.

You’ve done very well in taking down your own straw man, though. Good job!