[Discussion] Hope to Remember The Trump Administration Thread as being 'transparent and honest'

Let's follow and discuss what our newest presidential administration gets up to, the good, the bad, the lawsuits, and the many many indictments.

Chairman_Mao wrote:
Robear wrote:

Interesting article on polling.

Wapo wrote:

How can pollsters estimate how the country feels after talking to only 1,000 people?

The more people you ask, the more accurate your results. To a point.

It seems counterintuitive that you could talk to 600 or even 1,000 people and get a good sense for what a country of 320 million people are thinking. Until you look at it another way: When you go to the doctor, does she have to remove all of your blood to conduct tests, or does she just take a sample?

The answer, assuming you are alive right now, is the latter. She takes a representative sample of your blood and runs the tests. That’s what pollsters do with polls.

Comparing to taking a blood sample is an interesting analogy.

Especially seeing as your blood is a much more homogeneous pool to sample from. I think it would be much more analogous if the doctor took a random 3g of substance from you (and I bet the test results in that case would be much less likely to be an accurate representation of your current state).

Clumber wrote:

One path to victory in Michigan I see is Barr writes up arrest indictment for Joe Biden and a willing sheriff perp walks him in handcuffs out of a Michigan rally/speech. What the polls simply can't capture is how not normal this election has the potential to be.

As of March, Biden has Secret Service protection. I don't see a situation where they let a county sheriff take their charge away in cuffs. Part of their effectiveness is based on belief. That would damage it immensely.

Polling is not homeopathy.

Remember, the polling discussion is directly related to the earlier stuff about whether or not Trump is in a good place in the race.

Yeah. That blood analogy is terrible.

I don't have a problem with the math behind sampling. It's Hawthorne Effect, Goodhart's Law, and corollaries that bother me.

For instance: If polling is very accurate and timely, and available to democratically elected leaders who want to hold onto their power, are they going to shift their "views" to match the followers? That's not government, that's a chatbot heuristic partially running on wetware.

I think the same principle applies to any algorithm that aggregates data to infer trends to populate recommendation lists and news feeds. In both cases, recommendations and political polls, I also do not trust a well-sampled poll to not be manipulated in ways beyond sampling. What mechanisms decide which polls I see? Who authors the questions? Do they take into account regional differences in connotations? How many people involved started in advertising? If I feel I can trust a particular polling entity, what's the average shelf-life of that trust? Ten years? Two?

Give me a magic crystal that can answer all statistical questions about a populace instantly, with no side-effects. Should any of the answers effect how I vote? I feel like the answer is no. Perhaps an argument can be made that I should use the statistical results to better pick the right bloc.

Give me one Robear's words for ten years, and it isn't hard to find value. Questionnaire 10,000 Robears in a day, I don't think I can find the same value. Thirty Helens agree.

Danjo Olivaw wrote:

For instance: If polling is very accurate and timely, and available to democratically elected leaders who want to hold onto their power, are they going to shift their "views" to match the followers? That's not government, that's a chatbot heuristic partially running on wetware.

There was a Neal Stephenson book with basically that premise, called Interface, where the candidate has a chip in his brain that pipes in polling results.

I remember reading it and thinking "isn't this a perfection of democracy? Don't we want our representatives to know what we want them to do?"

Stele wrote:

Maybe because this isn't the election thread? We got that one shut down remember? This is just about Trump as president and all the awful things he continues to do.

Yeah, that's fair. We could use a general election thread separate from this one. The Democratic primary thread was shutdown, but the primary is essentially over at this point anyway.

Djinn wrote:

I'm certainly not saying that Biden's victory is guaranteed. Nothing is guaranteed in life. What I'm saying is that the numbers are extremely good for Biden and we should all be optimistic about November. Everyone is so gloomy in this thread. It's like someone being worried on their honeymoon that they might die in a car crash. Sure, it could happen, but try to enjoy the moment.

I just wanted to chime in and point out the reason so many people seem gloomy despite the polling numbers is that it's been proven time and time again, the rules don't apply to Trump. For what reason, I don't know. But when Dukakis posed for a picture in a tank 1988, it tanked his campaign. When Howard Dean got a little too shouty during a speech in 2004 it halted his campaign. Trump made numerous missteps throughout his campaign that were far worse, but it didn't effect him. Then he won when no one thought he would. Then in office he did multiple things that all could be impeachable. The democrats picked the one thing they thought they'd have the best chance at and impeached him, but it made no difference. He's still there and the constant news stories about how he's terrible at calls with foreign leaders, or is trying to get foreign governments to help his election, or whatever, aren't telling me anything new that I didn't already know 3 years ago- and it didn't make difference then so why now.

So until Trump loses the electoral college, concedes the election AND moves out of the White House in January, I (and I imagine many others) will be sleeping uneasy.

NYT: Data on Financial Transfers Bolstered Suspicions That Russia Offered Bounties
Analysts have used other evidence to conclude that the transfers were likely part of an effort to offer payments to Taliban-linked militants to kill American and coalition troops in Afghanistan.

American officials intercepted electronic data showing large financial transfers from a bank account controlled by Russia’s military intelligence agency to a Taliban-linked account, which was among the evidence that supported their conclusion that Russia covertly offered bounties for killing U.S. and coalition troops in Afghanistan, according to three officials familiar with the intelligence.
Tscott wrote:
Djinn wrote:

I'm certainly not saying that Biden's victory is guaranteed. Nothing is guaranteed in life. What I'm saying is that the numbers are extremely good for Biden and we should all be optimistic about November. Everyone is so gloomy in this thread. It's like someone being worried on their honeymoon that they might die in a car crash. Sure, it could happen, but try to enjoy the moment.

I just wanted to chime in and point out the reason so many people seem gloomy despite the polling numbers is that it's been proven time and time again, the rules don't apply to Trump. For what reason, I don't know. But when Dukakis posed for a picture in a tank 1988, it tanked his campaign. When Howard Dean got a little too shouty during a speech in 2004 it halted his campaign. Trump made numerous missteps throughout his campaign that were far worse, but it didn't effect him. Then he won when no one thought he would. Then in office he did multiple things that all could be impeachable. The democrats picked the one thing they thought they'd have the best chance at and impeached him, but it made no difference. He's still there and the constant news stories about how he's terrible at calls with foreign leaders, or is trying to get foreign governments to help his election, or whatever, aren't telling me anything new that I didn't already know 3 years ago- and it didn't make difference then so why now.

So until Trump loses the electoral college, concedes the election AND moves out of the White House in January, I (and I imagine many others) will be sleeping uneasy.

This

What percentage of our armed forces and law enforcement would join Trump and welcome a Russian invasion if it was billed to rid the US of the deep state and anyone they define as Antifa?

Jayhawker wrote:

What percentage of our armed forces and law enforcement would join Trump and welcome a Russian invasion if it was billed to rid the US of the deep state and anyone they define as Antifa?

Probably a much smaller percentage than that of ICE, CBP and police forces. But still much higher than 0%.

Tscott wrote:

I just wanted to chime in and point out the reason so many people seem gloomy despite the polling numbers is that it's been proven time and time again, the rules don't apply to Trump.

(edit) I really should have rolled this into new rules vs. old rules, but here's one case where Trump suffered consequences and the polls were right: the 2018 midterms, and it's worth keeping the reference because it's an important one.

They were a tidal wave where the polling numbers turned out to be right. As I remember they tightened a little towards the end, then widened again, and you had the predicted wave. Bitecofer's model (not a poll) got it right down to almost the individual congressional seat.

For what reason, I don't know.

I think I do--Negative Partisanship (again, Bitecofer's models, and the work of a lot of political researchers documenting this shift in the American electorate). There are fewer and fewer swing voters, so the rules have changed. All the things you listed (tank photo/Dean scream/impeachable offenses/etc.) are the kind of things that would cause voters to swing. They don't anymore. It's all about turning out your base.

Also, again--he didn't win when no one thought he would. It may be the fault of a bad job by the media understanding the polls and informing their audience, but the polls always gave him a chance to win. People mistook the chance that he would get blown out to mean there was no chance he could win.

There's a difference between the outcome being in doubt, and there being a bigger range of possible outcomes. In 2016 Trump had a chance to win, but also a chance to get blown out. In 2012, Romney had little chance to win, but also a smaller chance of a blowout. Some of the people reporting on the polls were trying to get that message across, but for whatever reason they didn't.

So until Trump loses the electoral college, concedes the election AND moves out of the White House in January, I (and I imagine many others) will be sleeping uneasy.

Whatever else is the basis of that (and I'd agree there are all kinds of dirty tricks he might get up to) the basis of that unease can't be "the rules don't apply to Trump and we don't know why" or "the polls were wrong." The polls never said he couldn't lose, and we know why the rules don't apply to Trump: because they were the old rules, but the new rules say he will lose this time, and for the same reason he won last time--Negative Partisanship.

Danjo Olivaw wrote:

I don't have a problem with the math behind sampling. It's Hawthorne Effect, Goodhart's Law, and corollaries that bother me.

For instance: If polling is very accurate and timely, and available to democratically elected leaders who want to hold onto their power, are they going to shift their "views" to match the followers? That's not government, that's a chatbot heuristic partially running on wetware.

Chatbot Heuristic Partially Running on Wetware 2020!

538 in their discussion had some very interesting breakdown commentary on why we should believe the polls. Specifically, it is how they are showing the erosion of Trump support among critical demographics. The two that were particularly hard hit were older white voters who appear to be alienated by all the "sacrifice grandma" talk and educated white women who have had three and a half years of Trump being Trump. Both of these demographics had issues with Clinton, but appear at least for now to find Biden a lot more palatable.

Another question that came up was whether or not Biden could court the Sanders voters and Nate made the observation that there really aren't very many of them this time around and that they are far more polarized than they were four years ago.

Paleocon wrote:

538 in their discussion had some very interesting breakdown commentary on why we should believe the polls.

Yeah but their business is polling. Bankers think banking is great too!

I feel like it’s counterproductive to assume Trump’s defeat because if that becomes the default narrative of this election amidst everything else going on it will give Dem voters and the Dem establishment a reason to let the campaign coast until November and the outcome of that happening is far less predictable.

ruhk wrote:

I feel like it’s counterproductive to assume Trump’s defeat because if that becomes the default narrative of this election amidst everything else going on it will give Dem voters and the Dem establishment a reason to let the campaign coast until November and the outcome of that happening is far less predictable.

Maybe, but it also gives motivation to push on stretch races.
Totally agree not to be complacent.

cheeze_pavilion wrote:

Trump is behind by a lot: "And as you can see in the chart below, both Biden’s average support and margin over Trump are historically large — the largest of any contender since Bill Clinton in 1996", including the states that matter. 9.6 in Michigan, 7.6 in Wisconsin, 5.6 in Pennsylvania. With 7.4 in Florida.

It's good not to be complacent, but the numbers are the numbers. There's a difference between complacency, and well-founded optimism.

If Trump wins Florida and Pennsylvania he becomes president. Sure, 7.4% is not nothing. But polls being 3% off would be considered reasonably accurate. So we are talking a shift of 4% in 4 months for Trump to be within polling error distance. That is well within possible movements.

cheeze_pavilion wrote:

Everything indicates that the Democratic party has been adding voters since 2016.

The there is the news stories about new voter registrations grinding to a halt under covid-19, with whatever consequences that might have.
https://www.npr.org/2020/05/26/86045...

cheeze_pavilion wrote:

I'd say "people's mistaken ideas about 2016" like "all the polls said she was going to win" when they didn't. People *reporting* on the polls did a bad job so I can see where they got that idea, but as I remember it, there was a lot of rancor directed at 538 for giving Trump such a high chance of winning. Criticism that ignored 538's explanations.

In any case, people now have a better idea of what actually happened in 2016 and how it differs from what happened in 2018 and why 2020 looks more like the latter than the former, so that should change some minds, at least a little?

Agreed. The polls did really well in 2016, especially at 538. It was mostly the news reporting that was bad.
The polls also moved quite a bit in the months leading up to the election. If we were a month or weeks from the election with the current polls, it would be more assuring. But 4 months is a long time, especially in 2020.

Shadout wrote:
cheeze_pavilion wrote:

Trump is behind by a lot: "And as you can see in the chart below, both Biden’s average support and margin over Trump are historically large — the largest of any contender since Bill Clinton in 1996", including the states that matter. 9.6 in Michigan, 7.6 in Wisconsin, 5.6 in Pennsylvania. With 7.4 in Florida.

It's good not to be complacent, but the numbers are the numbers. There's a difference between complacency, and well-founded optimism.

If Trump wins Florida and Pennsylvania he becomes president. Sure, 7.4% is not nothing. But polls being 3% off would be considered reasonably accurate. So we are talking a shift of 4% in 4 months for Trump to be within polling error distance. That is well within possible movements.

Granting that for the sake of argument, it still means he's behind by a lot. Margin of error *completely* in his direction (remember, he could actually be *down* by an additional three points!), AND a four point swing. That's a lot.

edit: oh, and also in your scenario I think Biden can't win either North Carolina, or Arizona and Maine's District 2, after quickly playing around with 270toWin.

cheeze_pavilion wrote:

Everything indicates that the Democratic party has been adding voters since 2016.

The there is the news stories about new voter registrations grinding to a halt under covid-19, with whatever consequences that might have.
https://www.npr.org/2020/05/26/86045...

I'm aware, but remember that we've already had an election already in 2018 that the Democrats surged in. Democrats already banked a whole lot of registrations before this hit.

cheeze_pavilion wrote:

I'd say "people's mistaken ideas about 2016" like "all the polls said she was going to win" when they didn't. People *reporting* on the polls did a bad job so I can see where they got that idea, but as I remember it, there was a lot of rancor directed at 538 for giving Trump such a high chance of winning. Criticism that ignored 538's explanations.

In any case, people now have a better idea of what actually happened in 2016 and how it differs from what happened in 2018 and why 2020 looks more like the latter than the former, so that should change some minds, at least a little?

Agreed. The polls did really well in 2016, especially at 538. It was mostly the news reporting that was bad.
The polls also moved quite a bit in the months leading up to the election. If we were a month or weeks from the election with the current polls, it would be more assuring. But 4 months is a long time, especially in 2020.

I'm not sure it is when you combine the polls with Bitecofer's model. Voting now isn't about some large pool of independents in the middle, it's about turning out your base. Four months isn't long enough to turn Democrat partisans into true independents.

There's a reason the GOP is against mail-in ballots. Voter suppression is the only way Trump wins again.

What Stele said ^

Jonman wrote:
Danjo Olivaw wrote:

For instance: If polling is very accurate and timely, and available to democratically elected leaders who want to hold onto their power, are they going to shift their "views" to match the followers? That's not government, that's a chatbot heuristic partially running on wetware.

There was a Neal Stephenson book with basically that premise, called Interface, where the candidate has a chip in his brain that pipes in polling results.

I remember reading it and thinking "isn't this a perfection of democracy? Don't we want our representatives to know what we want them to do?"

I think what you're asking for there is government by id. (as in, no ego or superego.)

Looking forward to the first strong AI so we can just elect it president.

ruhk wrote:

Looking forward to the first strong AI so we can just elect it president.

"Skynet 2020: Because what have you got left to lose?"

Prime Intellect for President: “For I desire nothing.”

ruhk wrote:

Looking forward to the first strong AI so we can just elect it president.

The one almost exclusively written by white guys with six-figure incomes?

You are assuming that China doesn't leapfrog us in technological advances. No, this is not a "ZOMG China!" post.
I just recall reading that they are severely outpacing and out spending us in technology fields like AI. Where I saw it is eluding me at the moment...

Mixolyde wrote:
ruhk wrote:

Looking forward to the first strong AI so we can just elect it president.

The one almost exclusively written by white guys with six-figure incomes?

Precisely.