2024 Pre-Election Political Polls | POLL - Trump would have had 7 point lead over Biden

  • Thread starter Thread starter nycfan
  • Start date Start date
  • Replies: 6K
  • Views: 144K
  • Politics 
Political forecasting isn't necessarily about who wins. A forecast that gets 50 states right but misses the % in every one by 5 points isn't as good as one that nearly nails all the %s but gets two tossup states wrong.
That sounds like a quote straight from Nate Silver. But I agree.

Speaking of Mr. Silver, he came up in John Oliver's show last night. There is no bootleg video available yet, but here is the quote:

Clearly, we don’t know what’s ahead on Election Night, though I can confidently predict a few things. First: a little khaki nerd armed with a touchscreen will inspire some of the most depraved thirst tweets the internet’s ever seen. Second, regardless of the results, Nate Silver will tweet, “Actually, this is exactly what I said could happen,” in a heated argument with a 14-year-old with a Spongebob avatar who only tweets, “Nate Silver pees upside down.” And finally: if Trump loses, he will not concede gracefully. He’s already said he’ll only accept the results “if it’s a fair and legal and good election.” And he’s been laying the groundwork to cry foul if he loses.
 
I don’t disagree with that. But as I read the poll, Trump isn’t running far enough ahead of the senate candidates to win the election. It will be close, but he’s behind.
One way we fail ourselves by obsessing about polls is that we convince ourselves that there are a whole bunch of ways to look at an election, when there is really only one. All the rest is repackaging.

You see it with Latino support. It's bad news for Dems that they are only getting 58% of Latino vote now instead of 66%. Dems can't win with 58%, they say. Why not? Look at the polls. They've got Harris leading with Latinos at 58%, because Harris is doing better with white college grads than ever before. The cross-section might be useful for understanding why something is happening, but not the what.

It's bad news for the Dems that Trump is leading on handling the economy and the economy is the most important issue. OK, so look at the polls. They've got Harris leading even with voters saying the economy is the most important issue. It just means that actually, people care about a mix of issues and not just the most important one.

It's bad news for Trump that the Senate candidates are trailing. But look at the polls. Despite these Senators trailing badly, Trump is only 2 points down. That's all baked in. It just means there are people telling pollsters that they will split ticket. Maybe people don't do that as much now . . . unless they start doing it, which is what these polls are telling us. There's no reason to think it can't happen.

All of these cross-sections and different angles are just the same thing sliced differently. The polls might not be accurate (probably aren't completely accurate) but why privilege the Senate polls? The proposition that "Trump isn't doing as well as his polls suggest, because the Senate candidates' polling is so bad" doesn't seem to be superior, in any way, as the proposition that "Trump is running unusually far ahead of the Senate candidates." Either way, we're seeing something different than in the recent past. We don't know if it means that the Senate candidates are being underestimated, Trump overestimated, or more people splitting.
 
That sounds like a quote straight from Nate Silver. But I agree.

Speaking of Mr. Silver, he came up in John Oliver's show last night. There is no bootleg video available yet, but here is the quote:

Clearly, we don’t know what’s ahead on Election Night, though I can confidently predict a few things. First: a little khaki nerd armed with a touchscreen will inspire some of the most depraved thirst tweets the internet’s ever seen. Second, regardless of the results, Nate Silver will tweet, “Actually, this is exactly what I said could happen,” in a heated argument with a 14-year-old with a Spongebob avatar who only tweets, “Nate Silver pees upside down.” And finally: if Trump loses, he will not concede gracefully. He’s already said he’ll only accept the results “if it’s a fair and legal and good election.” And he’s been laying the groundwork to cry foul if he loses.
I've been saying that since before I knew who Nate Silver was. That is pretty funny btw from John Oliver.
 
I've been saying that since before I knew who Nate Silver was. That is pretty funny btw from John Oliver.
I wasn't suggesting a copyright violation -- more a great minds think alike kinda point.

But if 538 says there will be 227 Republicans and Bouzy says there will be 213 Republicans and there ends up being 219 Republicans, I'd say 538 was more right than Bouzy. When the deltas are that close, the prognosticator that picks the actual winner is entitled to a significant credibility bump.
 
I wasn't suggesting a copyright violation -- more a great minds think alike kinda point.

But if 538 says there will be 227 Republicans and Bouzy says there will be 213 Republicans and there ends up being 219 Republicans, I'd say 538 was more right than Bouzy. When the deltas are that close, the prognosticator that picks the actual winner is entitled to a significant credibility bump.
Fine, but it's not the horrible fail you made it out to be originally.

Unlike what will happen if Bouzy actually goes to the mat on Florida going blue.
 
Fine, but it's not the horrible fail you made it out to be originally.

Unlike what will happen if Bouzy actually goes to the mat on Florida going blue.
I've never made it out to be a "horrible" fail, just refuting it as evidence of Bouzy's supposed savant status per @theel4life (and to a lesser degree @aGDevil2k).

Bouzy is not horrible because he missed on the 2022 election. He is horrible because he is a scammer, and he has glommed onto the "tell liberals that his secret sauce analysis predicts much rosier outcomes than the mainstream press is reporting" schtick in order to induce such panicky liberals to boost his algorithms, which he in turn uses to scam and grift.
 
I've never made it out to be a "horrible" fail, just refuting it as evidence of Bouzy's supposed savant status per @theel4life (and to a lesser degree @aGDevil2k).

Bouzy is not horrible because he missed on the 2022 election. He is horrible because he is a scammer, and he has glommed onto the "tell liberals that his secret sauce analysis predicts much rosier outcomes than the mainstream press is reporting" schtick in order to induce such panicky liberals to boost his algorithms, which he in turn uses to scam and grift.
"Not to mention he completely failed in 2022. "

I guess a complete failure isn't actually the same as a horrible failure, but I did conflate them in my mind. Anyway, I suspect we would agree about Bouzy, again if I bothered to look for what you've seen.
 
Can we really just not get into the Bouzy shit? It's been debated about 50 pages by now.

Meanwhile:


Interesting POV. I do imagine if she only loses by Ohio by 4, she's winning the race.
 
Last edited:
I think it is true, we don't really know and won't until election night or even a few days after as absentee and provisionals are counted in some places. I just really feel Trump's support is overstated in the polls. I think because it was understated in 2016 and 2020, I think they are trying to adjust methodology. I think Harris will win the national popular vote by 5% or better and win 300+ electoral votes. I really believe that.
 
I think it is true, we don't really know and won't until election night or even a few days after as absentee and provisionals are counted in some places. I just really feel Trump's support is overstated in the polls. I think because it was understated in 2016 and 2020, I think they are trying to adjust methodology. I think Harris will win the national popular vote by 5% or better and win 300+ electoral votes. I really believe that.
I could see her with a smaller popular vote win just from some softening in New York etc
 
Same here. I am thinking a popular vote win by six points or more and an electoral college victory of 319-219.
My "gut instinct" for months has been a feeling that Harris is going to win, and win by a larger margin than Biden did in 2020. I just get the sense that Trump fatigue is real and that his support is somewhat softer this year than in 2020 or 2016. OTOH, I also keep having flashbacks to 2016 and just can't convince myself at all that she's actually going to win. And I won't feel confident until (or even if) she's actually declared the winner after the election. Usually I trust my instincts, but not this time, probably because of everything that is at stake if he wins.
 
My "gut instinct" for months has been a feeling that Harris is going to win, and win by a larger margin than Biden did in 2020. I just get the sense that Trump fatigue is real and that his support is somewhat softer this year than in 2020 or 2016. OTOH, I also keep having flashbacks to 2016 and just can't convince myself at all that she's actually going to win. And I won't feel confident until (or even if) she's actually declared the winner after the election. Usually I trust my instincts, but not this time, probably because of everything that is at stake if he wins.
I feel identical to this.
 
I think it is true, we don't really know and won't until election night or even a few days after as absentee and provisionals are counted in some places. I just really feel Trump's support is overstated in the polls. I think because it was understated in 2016 and 2020, I think they are trying to adjust methodology. I think Harris will win the national popular vote by 5% or better and win 300+ electoral votes. I really believe that.

I've seen this stated multiple places (that pollsters are over adjusting their methodology) but I've never ready anything from a pollster stating they are adjusting their methodology to account for Trump over performing earlier polls. Can anyone point to an actual pollster stating that?
 
I've seen this stated multiple places (that pollsters are over adjusting their methodology) but I've never ready anything from a pollster stating they are adjusting their methodology to account for Trump over performing earlier polls. Can anyone point to an actual pollster stating that?
They always talk about model adjustments. They never give specifics
 
In re polls: I have no idea how polling works anymore. I don't answer phone calls from unknown numbers. I no longer have a land line. I don't respond to unknown emails. I don't respond to unknown texts. I throw junk mail away unopened. I think about 90% of the population acts like I do. How can someone conduct a poll when the only people they can contact are the insane 10% of the population who are so desperate for any sort of human contact that they will talk to a pollster?
 
Last edited:
Can we really just not get into the Bouzy shit? It's been debated about 50 pages by now.

Meanwhile:


Interesting POV. I do imagine if she only loses by Ohio by 4, she's winning the race.

As someone from Ohio I agree
 
Back
Top