Indecision 2024

[dyqik]

For some reason, a Cruz ad has just played here in Massachusetts. You can tell how Cruz is doing by the fact that he doesn't appear in his own ads.

[Woodchopper]

For those that don’t know the interesting thing about one Iowa survey. It’s conducted by Ann Selzer and she is a) an expert on polling in Iowa, b) is widely considered to be one of the best in the US, and c) has a long history of finding things that other pollsters in Iowa missed, and being right.

Four years ago, the race in Iowa was looking extremely tight. FiveThirtyEight’s polling average actually had Joe Biden practically tied with Trump in the Hawkeye State. At the time, this hardly seemed unusual; polls consistently said Biden was poised to make a big recovery with the Midwestern working class whites who had abandoned Hillary Clinton, and Iowa was the epicenter of that shift. The Selzer poll from September had found both candidates tied, 47–47, reinforcing this belief.

Then, Selzer dropped a bomb in the final week, with a survey that showed Trump ahead 48–41. This was a finding that suggested Trump was actually well-positioned to replicate his 2016 performance in Iowa, with big implications for the rest of the Midwest. It was Trump’s best poll in months, and arguably his best one of the entire year. And at the end of the day, Selzer turned out to be right, just as she was in the 2008 Democratic caucuses, in the 2014 Senate election, and the 2016 election — her outlier polls, once again, continued to consistently catch something that no other pollster was seeing.

This was, however, a huge deviation from basically every other poll of the cycle, and came with a number of caveats. Selzer is one of the most respected pollsters in the country, but no pollster is free from the laws of the universe that govern statistics. Sampling error is always a possibility, and her polls, even with their phenomenal track record, are not infrequently off by a point or two. And, of course, it was just one poll — the data in aggregate said something different.

All of these caveats apply to the astonishing poll that Selzer released this evening, where Kamala Harris leads Donald Trump, 47–44, in Iowa. It is inarguably one of the best results for Harris of the cycle. Although Iowa polling has been sparse (neither candidate is treating it as a swing state, given Trump’s 8-point margin there in 2020), there is scant evidence anywhere of the surge in support Harris would need to flip Iowa.

[…]

If this poll is anywhere near correct, the implications are vast. It is difficult to imagine a universe in which Kamala Harris achieves the level of white support necessary to be competitive in Iowa, but does not win Michigan, Wisconsin, Pennsylvania, and Nebraska’s second congressional district – and with it, the election.

While no other poll has shown quite this monumental of a shift, if you squint, there are perhaps hints of something similar happening in polls of similar states. Harris has polled exceptionally well in Nebraska’s second congressional district, and some polls of Nebraska statewide show a shift toward her as well. There was also a recent poll of Kansas that only had Trump up 48-43, a seeming outlier, but one perhaps worth taking a second look at in the wake of this poll.

Does this poll imply a Harris landslide? That’s one interpretation we’re skeptical of — even setting aside the outlier nature of this poll, it is worth noting that even a perfectly accurate Iowa poll cannot say much about states like Georgia or Arizona, where the whites vote differently from the Midwest. Barack Obama, for example, won Iowa easily en route to losing both of those states, and the idea of Harris possibly replicating something closer to the Obama coalition seems far-fetched. (But, then again, so does the idea of Harris being competitive in Iowa at all).

We have criticized pollster “herding,” a dynamic in which polls crowd around the expected result, instead of being willing to publish outliers or other surprising results. One thing is clear: Ann Selzer is not herding. It is possible that this becomes the first Selzer poll to seriously miss, through no fault of its own — again, sampling error can be unavoidable. On the other hand, perhaps Selzer is once again correct, and pollsters have been seriously underestimating Harris, an outcome that would remind people, at the very least, that polling error is unpredictable.

In three days, we will have our answer.

https://split-ticket.org/2024/11/02/the ... s-victory/

Anyway, this is just one poll in one state. We can come back on Wednesday and see whether Selzer was right again.

[Edited to fix quote]

[Woodchopper]

Iowa? Iowa!

It's not a poll about 6 EVs, it's a poll about women.

Yes, and especially older women and those who are self-described independents.

[headshot]

Kamala goes on SNL and nails it. (Albeit to people already voting for her…)

https://youtu.be/e6Funs6yyEw

[lpm]

The Iowa poll is unexciting when dug into.

Sample size only 800 but it pretends to give useful info on "white women over 65" and the like, then everyone is stretching to apply these sub-sub-categories to other mid west states.

[Woodchopper]

The Iowa poll is unexciting when dug into.

Sample size only 800 but it pretends to give useful info on "white women over 65" and the like, then everyone is stretching to apply these sub-sub-categories to other mid west states.

Nothing else to do while we wait.

[lpm]

I saw someone on twitter say they were nauseously optimistic.

[dyqik]

The Iowa poll is unexciting when dug into.

Sample size only 800 but it pretends to give useful info on "white women over 65" and the like, then everyone is stretching to apply these sub-sub-categories to other mid west states.

The story it tells lines up with recent polls/surveys showing Kansas at only Trump +5 and Ohio at Trump +3.

And it's not just white women over 65, but white men over 65 as well. There's a good few corn and soybean farmers etc. in the Midwest who don't like the idea of tariffs and trade wars, and don't like Trump's tone, or threats to medicare and social security. They probably do like Harris's proposal to fund elder care through Medicare.

Kansas has a twice voted for a Dem governor, and rejected an abortion ban by a 20% margin.

[Woodchopper]

Final results from NYT Siena College

NYT Siena.jpg (163.64 KiB) Viewed 1974 times

Both candidates have multiple pathways available to capture the 270 Electoral College votes required to claim victory, assuming that polls are not dramatically underestimating the support for one or the other. In such a close race, even a small systemic polling error could tip the contest decisively in either direction.

But there are signs that late deciders are breaking for Ms. Harris: Among the 8 percent of voters who said they had only recently decided on their vote, she wins the group by 55 percent to 44 percent. (With Election Day nearing, 11 percent of voters remained undecided or persuadable, down from 16 percent about a month ago.)

The polling comes as more than 70 million Americans have already voted, according to the University of Florida Election Lab.

Roughly 40 percent of those surveyed by the Times/Siena poll across the seven states said they had voted. Ms. Harris wins those voters by a margin of eight percentage points, the polls found. Mr. Trump has an edge among voters who say they are highly likely to vote but have not yet cast a ballot.

[...]

Rates of early voting are particularly high in North Carolina, where more than half the voters said they had already cast a ballot. Ms. Harris wins early voters in the state by 8 percentage points, perhaps contributing to her three-percentage-point edge in the survey of the state. Despite recent devastation there from Hurricane Helene, more than nine out of 10 North Carolina voters said that the storm and its aftermath have had no impact at all on their ability to vote.

[...]

The only state where the poll found Mr. Trump winning with people who said they had already voted was Arizona. Forty-six percent of voters there said they had already voted, and Mr. Trump wins that group, 50 percent to 46 percent.

The polls also reveal a shift in the issues being prioritized by voters in the final stretch of the race. The economy still remains their top concern, but in states like Wisconsin, where Ms. Harris has held a consistent edge, abortion now nearly matches the economy as voters’ most important issue. And in Arizona, where Mr. Trump leads, immigration also continues to rise as a crucial issue driving voters’ choices.

[...]

The survey shows that Mr. Trump has continued to hold on to the core of the coalition that supported him in his past two presidential bids — white voters who did not attend college, and men — while expanding his support among younger, nonwhite and newer voters. He is exceeding his 2020 vote share in Arizona and Michigan, both states he did not win four years ago.

Ms. Harris is underperforming relative to President Biden’s performance in 2020 with younger voters, Black voters, particularly Black women, and Latino voters. But she has improved on his numbers with these groups since he dropped out of the race in July.

The gender gap remains wide across all seven states, with Ms. Harris the favorite of women and Mr. Trump preferred by men. For women and younger voters, abortion now surpasses the economy as the most important issue driving votes.

https://www.nytimes.com/2024/11/03/us/p ... -poll.html

We'll get some good information from the exit polls about whether there has actually been a decline in support from young people and persons of colour. IMHO if more young people are voting for Trump that would be a similar phenomenon as found in Europe - angry young right wing men rather than people radicalized by Gaza. Most of the supporters of the Palestinians don't live in swing states, and among those that do, a Trump white house would probably be worse.

Worth noting that based upon the above there are still several plausible routes to the White House, and some don't involve PA. For example:

MI, WI and NV plus either GA or NC
WI and NV plus GA and NC
WI, MI, GA and NC

The picture should be clearer by Wednesday morning, but I expect that counting in some close states will go on for a while.

[bob sterman]

Final results from NYT Siena College

NYT Siena.jpg

Not that encouraging. PA and MI far too close for comfort.

[lpm]

Yes, but you'd rather have Harris's numbers. On the assumption polls are accurate she's slightly better than 50/50.

[lpm]

The picture should be clearer by Wednesday morning, but I expect that counting in some close states will go on for a while.

It will be unlikely to need this long.

2016 was very close, but we knew Trump had won by about 3 or 4 a.m. UK time.

2020 was even closer, but it did look OK by 3 or 4 a.m.

US projections come fast, thanks to the piecemeal reporting and the correlations between districts. Early Virginia results - though not a swing state - will show the direction North Carolina and Georgia are heading. Ohio, IA, PA, MI and WI will likely show similar swings to each other, though different to the VA, NC, GA zone. Then the AZ and NV zone comes surprisingly early, they must close the polling booths at 4 pm or something ridiculous. None of this voting till 10 pm for Americans.

[Woodchopper]

The picture should be clearer by Wednesday morning, but I expect that counting in some close states will go on for a while.

It will be unlikely to need this long.

2016 was very close, but we knew Trump had won by about 3 or 4 a.m. UK time.

2020 was even closer, but it did look OK by 3 or 4 a.m.

US projections come fast, thanks to the piecemeal reporting and the correlations between districts. Early Virginia results - though not a swing state - will show the direction North Carolina and Georgia are heading. Ohio, IA, PA, MI and WI will likely show similar swings to each other, though different to the VA, NC, GA zone. Then the AZ and NV zone comes surprisingly early, they must close the polling booths at 4 pm or something ridiculous. None of this voting till 10 pm for Americans.

Fair enough, I guess it'll depend upon whether a crucial state, eg NC, MI, PA or NV, is so close that there is a recount. Back in 2020 it took a couple of weeks before the GA result was certified. But Biden had won anyway.

If there is a decent margin across the swing states we'll know by Wednesday Morning.

[FlammableFlower]

I guess one issue is how much Trump is looking to gum things up in the courts and use that time to cause trouble crying foul.

[bjn]

Don't forget the threat of violence as well. He did it before, he'll do it again.

[El Pollo Diablo]

538's model then:

https://projects.fivethirtyeight.com/20 ... id=rrpromo

Modal result is approx. Trump 312-226 Harris. Median is around Trump 277-261 Harris. Mean is Trump 276-262 Harris.

But what do models know? Harris's ceiling is higher than Trump's, and that Iowa poll is a seductive one.

[lpm]

Don't forget the threat of violence as well. He did it before, he'll do it again.

Sure, but the violence and fraud and court cases won't prevent us knowing who won the election.

For example the networks called the 2012 election for Obama at 11:12 pm ET. Karl Rove got cross, in a forerunner of Trumpism, and demanded Fox withdrew their call. Romney didn't formally concede until 1 a.m.

But technically Obama hadn't won the Presidency until the electors voted (17 Dec this year), travelled by horse to Washington, submitted their tally to Congress, and then Congress counted - which we'll forever now know takes place on Jan 6th.

[lpm]

Last time at 00:46 ET Biden said he was on track to win. Things were looking OK by about 3 a.m. ET.

But bl..dy hell, the opinion polls massively underestimated Trump in 2020.

[Woodchopper]

538's model then:

https://projects.fivethirtyeight.com/20 ... id=rrpromo

Modal result is approx. Trump 312-226 Harris. Median is around Trump 277-261 Harris. Mean is Trump 276-262 Harris.

But what do models know? Harris's ceiling is higher than Trump's, and that Iowa poll is a seductive one.

Down at the state level he has PA, WI and MI as too close to call. The result comes down to those three and neither Nate Silver nor the rest of us have a good idea about which way they'll fall.

[El Pollo Diablo]

There's a good chart here which shows the average polling error in US elections past:



A trend going back to the 70s is that the Republicans have been underestimated twice in a row, and then are overestimated for one election. Obviously, past doesn't predict future, etc., but the last two elections underestimated the Republicans...

[lpm]

Here's when previous elections were known/called.

2020 following Saturday
2016 2:45 a.m. ET
2012 11:12 p.m.
2008 9:18 p.m. (Ohio called for Obama, formal network call at 11 p.m. on closing of California polls)
2004 following Weds 2 p.m. (Ohio again)
2000 never
1996 9 p.m. (WI and MI)
1992 9 p.m.
1988 9 p.m.

Clearly it could be another all nighter. But more normal to have an outcome by 11 p.m., with a concession speech before midnight.

[lpm]

It is unknowable.

But on one hand

- polls underestimated Trump in 2016 and 2020
- Americans are as mad as f.ck

On the other

- plausible dyqik arguments on polls overestimating Trump post Dobbs
- hilarious final week for Trump
- no GOP ground game because Trump nicked the money and Musk thought he could do it
- Trump is behaving like he thinks he's lost

So on the whole got to be optimistic.

Dyqik, you've got the most optimism, what odds do you reckon?

[Woodchopper]

Here's when previous elections were known/called.

2020 following Saturday
2016 2:45 a.m. ET
2012 11:12 p.m.
2008 9:18 p.m. (Ohio called for Obama, formal network call at 11 p.m. on closing of California polls)
2004 following Weds 2 p.m. (Ohio again)
2000 never
1996 9 p.m. (WI and MI)
1992 9 p.m.
1988 9 p.m.

Clearly it could be another all nighter. But more normal to have an outcome by 11 p.m., with a concession speech before midnight.

Good points. I guess this time it feels a lot closer. But maybe I’m wrong about that. Come Wednesday we can count how many states are within 1%.

[Chris Preston]

My thoughts for the virtually no value they are worth. Although, I am worried that I am allowing my biases to sway my thinking, but I feel looking at the data that there are a lot of potential negatives for Trump and a number of positives for Harris.

Perhaps the first one is that there seem to be a large number of polls with less than robust methodology that are favouring Trump and I think this influencing how the media is portraying the election. I strongly suspect there are groups deliberately publishing dodgy polls to muddy the waters. The recent Iowa poll showing Harris ahead was a surprise, but I suspect indicated the race in Iowa was much tighter than it seems. Even though this was a small poll, it came from a pollster with a good past record in Iowa. The next day there was a poll by Emerson College Polling/RealClearDefense poll giving Trump a 10% lead in Iowa. This poll is clearly nonsense and has likely been fudged.

What I take from looking at the polling over the last 6 weeks is that I don't feel support for Trump is being under-estimated this time around. There is about a 6-10 point difference in favourability between Harris and Trump that does not seem to be influencing voting intentions. I find this strange. (Actually when I look at the polls involved most polls come in the 8 to 15% unfavourable for Trump, with a few pollsters always in the 2 to 3% range. This is one of the data points influencing my thoughts about deliberately dodgy polls.)

Secondly, there seems to be a lack of enthusiasm for Trump in this election compared to 2016 and 2020. Trump had always prided himself on the size of his rallies. However, the rallies are having fewer people show up and the people that do show up are leaving early.

Virtually no ground game for the Republicans. Trump burgled the RNC coffers to pay his lawyers and there has been no organised ground game in most states. This has been exacerbated by the infighting in the Trump team between Corey Lewandowski and Chris LaCivita. Lewandowski, as a means of trying to oust LaCivita, complained to Trump that LaCivita was spending funds on the ground rather than on television advertising. Of course, Trump cares nothing about things where he is not the centre of attention, so more funds were switched to television advertising. Because Trump has managed to focus so much of the money on himself, there has been a lot less available for down ticket candidates. This may come back to haunt Trump.

Early voting is over 78 million votes now. This points towards a high voter turnout. High turnout will favour Harris rather than Trump. There is also 10% gap between women and men in early voting. High turnout of women will also favour Harris.

Finally, both campaigns have their own internal polling that we don't usually hear about. The body language from the campaigns points to the Trump campaign knowing they are in trouble; whereas the Harris campaign seems confident.

The path to victory appears easier for Harris than Trump. Trump must win both North Carolina and Pennsylvania and have other states fall his way to win. Hence his activities today. The Harris campaign seems to be confident of Wisconsin and Michigan, so all the effort today is on Pennsylvania. The fact that North Carolina is in play in this election is all down to mismanagement by the Republicans driven by Trump.

[dyqik]

A local poll today has Harris ahead by 23 in New Hampshire. Biden won it by 7 in 2020, and it has a serious Trump libertarian bent outside the cities.