I suppose they could just listen to Trump speak. Although my MAGA mother confided in me that she doesn’t actually listen to Trump speak she reads transcripts. I should have asked where she gets those transcripts. Every one of us can be fooled. I actually believed George Bush, Jr. when he told me Saddam had weapons of mass destruction and I continued to believe him even when experts told us this was a false narrative. Part of the reason I believed Bush was because I simply couldn’t fathom the idea that he’d lie about something that big when it would obviously be uncovered. Pretty naive I guess. But eventually I had to face the facts and see that I had been tricked. There were no WMDs.
While Fox news, Trump himself, and other outlets bear the brunt of the responsibility for their lies, at some point the people being lied to need to share some of that responsibility as well. It’s not easy to admit you’ve been fooled but it sure beats letting someone continue to make a fool of you.
Ah, but “the line” of just voting for a Black Man is only one of many “lines.” And for many people at the end of the day really not that huge. Everyone has different “lines”, some that are existential, irreversible; some that are situational, contingent.
In 2008/12 people were seeing someone exciting and charismatic promising hope & change vs Yet Another Standard Corporate White Guy. (And remember a lot of Obama’s support was individual to him. His coat-tails were notoriously short especially when HE was not in the ballot… ) Meanwhile, though, the Tea Party were flexing and building up at the state and district level 5 years before Trump showed up. The “transformational” thing stopped way short of what some hoped for.
It would not surprise me if some of those who went Obama —> Trump are yet another electorate segment I did not mention earlier: people whose “HOPE” is just for someone who will disrupt the system. And everyone has ample proof Team Donald really wants to disrupt the f… out of the system.
Yeah, I find myself unwilling to agree that nearly half the voters are evil fascists. But whatever their motivation, it is exceedingly hard for me to come up with a positive descriptor.
I’m very happy to write off the 30% MAGAs. 70% sane folk should easily outweigh them. But I find myself distressed and confused about the remaining 15-20% who will vote for Chump.
I’m glad that the models exist as it’s good to have smart people looking at these things in depth, but they are so full of moving parts that in turn are based on judgments of varying levels of arbitrariness that I can’t quite take them seriously.
An example: as someone said above, Silver is currently suppressing Harris’s convention bounce, which puts Trump on top. Does a surprise candidate like Harris even get a convention bounce, or is any bounce just part of the quick arc up? Silver has no particular insight into that question, since it’s an unprecedented situation. Also, Trump got no bounce at all–is that owing to his status as a repeat candidate, perhaps? Or maybe convention bounces aren’t a thing any more.
Multiply the above kind of issue by a big number, and the problem with the models becomes apparent. I don’t think they offer much value-add beyond a grokking of where the swing state polling is and a simple count of likely electoral votes.
And, unfortunately, the fucking part means they are reproducing…
More seriously, I think there are some who might simply prefer him for some single issue like abortion, or wealthy folk believing they will profit financially. While it is not praiseworthy that they would support a dangerous fascist to achieve those goals, I think it not entirely accurate to call THEM fascists. Might be a subtle distinction, but one I feel worth making.
There’s one thing that makes me very skeptical of the polls, which is that the general mood of the country, to the extent that one can suss it out, seems to be strongly pro-Harris.
I’ve read a LOT of posts on the SDMB and reddit and so forth saying things like “I live in a neighborhood that had tons of Trump signs in 2016 and 2020, and there are hardly any now” and “young people are registering to vote in overwhelming numbers”.
Now, could that all be anecdotal and wishful thinking? Possibly. But the “buzz”, for want of a better word, seems to be vastly better for Harris than it was for Clinton in 2016, or even for Biden in 2020… and certainly better than it was for Biden in 2024 before he dropped out. But the polls seem to hardly reflect that at all.
It’s certainly possible that the entire concept of large-scale opinion polling is now propped up by so many layers of correction/drift/assumption/whatnot that it’s basically ceasing to be useful at all. At least, I hope that’s the case, because basically everything I observe leads me to believe that Harris doing great… except for those stubborn polls.
The old math lesson about operations and reporting significant digits comes to mind: as a general rule the accuracy of a calculated result is limited by the least accurate measurement involved in the calculation.
We have assumptions made in these models and data going in that we know are +/- huge fractions. The typical systemic polling error alone is three to four percent, if not more. Reporting these model results as precise odds, let alone taking changes of 10% or less with any seriousness, is probably being a bit innumerate. @Aeschines you are right I think: to some degree the models are punditry with a mathematical facade. To any degree of significance all we can say is that the data is not good enough at this point to make much a prediction at all other than some extreme results being very unlikely.
I remain cautiously optimistic only because I see Team Harris executing well, better than Teams Clinton or Biden had (even if the polling doesn’t reflect that), and Team Trump both unable to find any solid footing so far and looking unlikely to suddenly find it. But I recognize the risk that I may be seeing what I want to see …
Mind you I will still obsess over ever poll and deep dive cross tabs! But I know it isn’t very rational to take those dives too seriously.
Here’s a little more detail about Silver’s accounting for post-convention bounces, complete with data from 60 years of past general elections (non-paywalled, below). Something that Silver elects not to spotlight is that in the 21st century, post-convention bounces have dropped WAY down – like within-margin-of-error low (with 2008 a bit of an outlier).
What you’re getting at is an idea into which I put considerable stock: Polls cannot capture results from the type of people who would never respond to polls. In this day and age, that largely means people who opt out of capital-P Polling by (a) rarely answering unsolicited cold calls/texts on their smart phones and (b) not explicitly signing up to regularly supply poll responses by online polling outfits.
Flip the coin around: Most polling response are from those truly motivated for one reason or another to get their opinions into polls. That’s a powerful form of self-selection.
Furthermore, this is an effect that, to me, may have easily increased between 2016 and 2024 – at least increased enough to matter. How many landlines have gone out of service since then? How many “when the phone rings, you answer it!” elderly folks have passed away in eight years?
Other posters have mentioned other types of polling, such as of specific demographics, and Harris seems to be killing it in those but then not in the national or state polling (even though she is doing at least pretty well in these–but I agree she should be slamming it considering whom she is up against). So it all doesn’t really seem to be adding up.
Absolutely. 538 gives its result without a decimal, which still might be pushing things, but it’s simply ridiculous for Silver to express his result to one decimal place. And the man knows better than to do so, I’m certain.
Another issue is that it’s impossible to validate the models. The sample size of general elections is too small, and politics is constantly changing, so it’s a moving target anyway. Finally, they can’t be falsified. I mean, Silver gave Trump a 25% chance in 2016, so his fanboys can say, “He wasn’t wrong!” OK, then, how can he be wrong? (People have said that he would be justified if, after a ton of elections, his percentage correct was such and such; but here again we run into the issue of sample size. And the models are only useful in the first place because the Electoral College makes things complicated; they are not useful in elections where straight up polling tells you what you need to know with maybe just a bit of aggregation.)
Again, I’m glad the models exist, but an Internet browser should issue a grain of salt every time someone navigates to one of them.
Yes, the enthusiasm for Harris-Walz is unambiguously big. Trump’s support seems solid but tepid, and he continues to go nuts in military cemeteries, etc., which would not seem to endear him to “independents.” Considering that Trump lost in 2020 and new registrations by young people, Black women, etc., are off the charts right now, the idea that Trump will just coast to victory strikes me as absurd.
I think it would make sense for convention bounces to be way down since the rise of the Internet. The conventions used to be a way for people to get to know the candidates much better and all at once with a very favorable manner of presentation. Now, however, the good and bad news about a candidate is already out there long before the convention.
Clinton received a significant poll bounce from the convention, due to both the perceived success of the convention, as well as Ross Perot announcing he was withdrawing from the campaign just as the convention was ending (Perot got back into the race in October).
The convention bounce gave the Clinton/Gore ticket a lead that only shrank insignificantly when Ross Perot re-entered the race. Clinton and Gore went on to defeat President Bush and Vice-President Quayle, as well as independent candidate Ross Perot and his running mate, James Stockdale, in the general election.
Well, before the UN inspection team went in- everyone thought Saddam had WMD- in fact he even bragged he did. But when the team found nothing- Bush invaded anyway. That’s the point- threatening SH to let the Inspection team in- GOOD. Invading anyway- BAD. Later they did find quite an amount of old stuff buried and rusting in the desert- no longer much of a danger as weapons, but only to the environment. SH had no WMD that were a danger to the free world.
