From an Emerson poll a few days ago:
“Voters who made up their mind in the last month or week break for Harris, 60% to 36%,” Kimball said. “The three percent of voters who said they could still change their mind currently favor Harris, 48% to 43%.”
This gibes with my belief that most people who haven't already committed to Trump are likely to vote for Harris when they finally make a decision. We'll see.
I sure hope you are right!
🤞🤞🤞🤞🤞. The bigger the margin of her win the harder it will be for
them to cheat.
Good point. However, the enthusiasm factor is still a major element in the election outcome. In 2020 the polls way underestimated Trump voters. Millions more showed up to vote for Trump. If we are lucky, the enthusiasm factor this time will work more in Harris's favor.
Also, I heard a polling expert say that pollsters are “re-weighting” polls to try to correct for errors in 2016 and 2020 polls. I take this to mean that they are upping estimates for Trump votes since he overperformed in the past. If this is true, Harris does not need to win by say 4 points in the popular vote to have a good chance of winning. I sure hope this is true.
There’s enthusiasm, and there is grim determination, as in, “We’re not going back!” And there is being motivated by both sentiments.
yeah, there's a ton of re-weighting and other tweeks supposedly happening at all the pollsters, but it's all backward looking in a polling environment that keeps changing so i'm not at all confident it's increasing accuracy vs decreasing it. we'll never know until the election is over.
If accuracy is decreasing such that they are underestimating Harris – I'm good with that!
from your lips to god's ears. problem is, as Nate Silver pointed out today, "Empirically, there’s basically no correlation in polling error from one cycle to the next one" and maybe they're overestimating harris. who the heck knows. so i'll be hopefully until my hopes are crushed.
Nate Cohn has an article today about "what went wrong" in 2020. One of the explanations - the one he seems to favor is COVID. Because Liberals took the pandemic restricions more seriously, they were more likely to be home when polling firms called in 2020. That led to Trump voters being undercounted.
I've worked doing telephone surveys. You don't call a number just once. If you try and there's no answer, it gets tried again, several times. It minimizes the chances that you just call someone at the wrong time.
Because Liberals took the pandemic restricions more seriously, they were more likely to be home when polling firms called in 2020.
That's as good an explanation as any.
I take some comfort from the strong evidence that suggests 2016's and 2020's polling errors were entirely unrelated, and that, moreover, there's not really any history of systematic polling error favoring one party (or the other) over many cycles. In other words, it seems to me simple probability indicates the GOP isn't terribly likely to benefit from favorable polling imprecision three presidential elections in a row, and certainly it has been known to go the other way (in 2012, for instance, the bulk of the polling ended up underestimating Obama's popular vote total).
This whole genre of navel-gazing about how the polls were so wrong in 2016 and 2020 is ridiculous. The errors were within normal polling errors. Trump won in 2016 by virtue of a few thousand votes spread across three states. If we were a normal democracy and chose based on actual votes, he would have lost.
He's also lost every election since then: 2018, 2020, 2022 because he's a known quantity and while there's a core of about 47% who like him (a horrifying enough thought) that's not enough to win and he has no chance of convincing the 53% who are adamantly opposed to him.
In 2016, the national polls were within normal error ranges. The state level polling, especially Wisconsin, was a disaster.
In 2020, neither performed very well.
There's something happening here.
In '16 the polls looked good but there were people like Michael Moore and Chris Arnade saying if you looked at the ground level, it was not good. Trump had more support than the polls showed and than a lot of people realized.
I think the opposite is happening in '24. I've seen many reports of voters in unlikely places showing strong support for Kamala. It's anecdotal, not "data," but it's also not just an isolated case or two. It's people finding out they're not alone and that something is changing in the country, and for the better. It may be the great story of our times, though you won't hear about it from the powers dedicated to the idea that we are a hopelessly and evenly divided country.
This short video from a woman in red America is a great example:
https://x.com/maddenifico/status/1848532726443360409
Another thing.
I have seen news stories and media posts on people voting for the Dem candidate for the first time.
I may have missed it, but I have not seen any reports of voters who were for Clinton in '16, for Biden in '20, and have now swung to Trump for the '24 election.
Harry Enten has pointed out that Harris is doing better than Clinton or Biden with non-college educated voters, a key demographic in Pennsylvania, Michigan, and Wisconsin. Course she's doing worse among Blacks and Hispanics so who knows how things will turn out?
From your keyboard to God's ears....
Another another thing.
What the hell is happening at the L.A. Times? In this of all elections the paper cannot find any candidate for president worth endorsing? Maybe letting the megarich buy up newspapers is not the best way to save journalism, or democracy.
The owner is a billionaire from South Africa. Need I say more?
I’d like to see a poll with crosstabs for ‘Republican women whose SO’s are/aren’t within earshot…’
And for voters of all genders, I hope entering their polling places is a sobering experience. I just read a comment or two in the NYT today — one from a claimed Harris voter! — saying how ‘fun’, ‘entertaining’, ‘likable’ the crazy old orange guy is. Dammit, people, the winner of this one doesn’t get a rose or a sash, they get the nuclear codes and the Executive branch.
I have not seen the press out searching for the true Harris supporters.
We've been subjected to years of this with Trump supporters; trying to understand them, identify them (it ain't hard, just look for the hats), and sympathize with them.
I have yet to see a single piece focused on liberal voters in the same way. And that's a shame. Maybe after the blue tsunami that will change.
Thanks, this gives me some hope!
I just spoke to someone who lives in a more rural area and she said there are very few Trump signs compared to the past. I'm cautiously optimistic.
I hope you are right. I mean, I *really* hope you are right. But since we're doing anecdata: The rural area where I live will have no effect on the final outcome -- it's a very red part of New York state -- but I have been seeing a lot more Trump signs than I remember from 2020. Hopefully that's not representative of a shift in attitude in rural parts of PA, WI, etc.
One thing that's notable in the polling is that Trump was increased his support in blue states, especially in New York and California. This is why Harris probably doesn't need as big of a national popular vote margin to win as Clinton and Biden did.
I, too, just drove through a small, conservative town in Michigan and all I saw were Harris signs. Not a single Trump sign.
It would help if she could get a word in edgewise. Press is wall-to-wall coverage of Trump's insanity and they wonder why he's winning.
A lot to criticize on how the press covers Trump. But he's not winning.
That sounds good, but the key question is: what was the partisanship breakdown of those late deciders? If most of them identify as Dems or leaners, this may or may not be good news.
I don't follow your argument. The polling suggests: (1) a pretty close race (2) a likely Harris lead in the two point range—which may not a be a margin large enough to win the Electoral College, and, (3) a modest number of voters—perhaps 5%—still up for grabs.
If Trump wins the bulk of these he likely wins the election. If Harris wins the bulk of these she likely wins the election.
At this point party registration doesn't matter. All that matters is being first past the post in enough swing states.
I’m in State College, PA (a blue island with a 3-mile radius, surrounded by deep red) and I don’t see any less enthusiasm for Trump outside of town than the last two elections. Hopefully, a couple of percent of 2020-Trumpers will switch their vote, but have been afraid to put up Harris signs.
What’s disconcerting (at least to me) is the more overt support for Trump this time from some factions of PSU Students…namely the frats: Bro culture. Not sure how widespread and deep that sort of support runs, but it concerns me.
When was the last election that a comparable number of highly placed members of one major party supported the candidate of the other? It’s difficult to imagine a candidate so unfit as to cause that, without also supposing that quite a few of the rank-and-file are not also doing the same, or considering it, or deciding to vote third-party/write-in, or not vote the top line at all.
Bexar County TX (San Antonio) first day of voting set a new record by 33%. This is one of the blue dots in TX. Not sure what this means nationally but a lot of democrats are voting here in bright red Texas!
Great! I'm in Tx too. One thing, I think (hope!) Ted Cruz might be going down. Hoping Allred overperforms compared to polls. O’Rouke overperformed polls by 4.4 points in 2018.
If Ted Cruz loses, we're past blue-wave and into blue-tsunami territory.
If he loses you can be sure he'll jump on the "it was rigged/they cheated" band wagon. He's another piece of shit that I assume won't concede or not till he has no other choice.
I still think the biggest news in this election is going unreported; though my evidence is anecdotal.
Every single person I know has lost a loved one to the alter of Trump. Grandmothers who (for the 3rd election in a row) have granddaughters who will not talk to them. Young mothers who are estranged from their fathers, and reluctant to have their children around them for the way Daddy runs off at the mouth.
Trump is not just tearing the nation apart, he's tearing generations of families apart. And I hope, 3rd time around, there's a lot of old farts reconsidering that.
This is absolutely true, and you'd think this would be fodder for a good story, but if there has been one, I haven't seen it. Unfortunately, I don't think anyone is going to reconsider their views as a result.
I hope you're wrong; I know a few grandmothers who are 1) estranged from their families over Trump, 2) pro-life, and 3) longing for reconciliation. So I expect some of these older women to switch or simply not vote for president.
Not voting for Trump solves a lot of their problems; even if a vote for Harris feels a bridge too far.
I have two brothers-in-laws, and one sister-in-law, who are Trump supporters. I have one old friend I keep in touch with who is a Trump supporter. To avoid hard feelings, my solution is to simply refuse to talk politics with them, beyond acknowledging we are on opposite sides. I know there is no way I can change their minds.
The thing about so many Trump supporters is that they think they are entitled to rant and rave and everyone will simply agree with them. My one brother-in-law started doing that with my wife (his sister), and she just calmly said she doesn't support Trump. He still got mad at her, and he may still be. But they are at least still on okay terms.
What Drum is reporting is also what I would expect. But it does not seem to be being reflected in the polls which are drifting slightly towards Trump. I am still hoping that Drum's take is right. It makes a lot of sense. But it doesn't seem to be supported by the data.
I believe that newly registered voters do not show up in the Likely Voter model. There's lots of new young registered voters that aren't showing up in the polling.
Yeah, this really isn't Kevin's area of expertise. For example, counter what Kevin wrote a few days ago, here's Nate Silver on why favorability doesn't matter:
https://www.natesilver.net/p/likability-isnt-enough
Way back in 2020:
To reiterate, the campaign, lacking cash all season long, has outsourced the ground game to SuperPACs. Those SuperPACs have outsourced to contractors. Those contractors have outsourced (hired) flaky individuals who haven't done much of the work they were claiming to have performed. Vendors have been fired. Fired individuals have moseyed on over to a different vendor.
Last election it was the GOP's goal to match the Democratic Party's vaunted ground game; this year they're outsourcing it to people like Charlie Kirk and Elon Musk. Have these folks ever been involved in building a ground game? Nope. Do they at least have a working app to track efforts to contact voters? Nope.
If Musk weren't desperate, he wouldn't be walking a legal line offering to give people who sign a pledge to support the 1st and 2nd Amendments a chance to win a daily $1M giveaway.
I read in the Guardian that canvassers are using a GPS spoofing app to pretend they have knocked on doors when they haven't. Gee, who could have guessed that when grifters outsource essential activities to other grifters, this kind of thing might happen?
Reminds me of the people playing Pokemon Go who used GPS spoofing to travel around the world to catch regionally-limited Pokemon and access spots they could not otherwise access, right up until Niantic finally caught on and started banning them.
Early voting is smashing records, and by all accounts turnout is going to be very high this year. Consider that 2020 was tied (with 1960) for the highest turnout as percentage of eligible voters since at least 1932. My belief is that the NUMBER of votes Trump got in 2020 is his ceiling. R's turnout regularly and it's unlikely he's going to convince more than a token number of Biden voters to cross over. If Harris can match the turnout from 2020, she should win handily.
I think it will come down to whether Arab-Americans in Michigan drink their election beer from a Stein. If Dr. Jill gets more than a smattering of votes from them, Harris will lose the state and the election. North Carolina could replace Michigan, but I just don't see it. The African-American population is a bit too much smaller than in Georgia.
Here in central New Mexico, the League of Women Voters has had to order an extra 4,000 copies of its printed Voter Guide, n addition to the 28,000 already distributed. The Bernalillo County clerk has reported record numbers of early voters, and the Voter Guides have been highly coveted.
I just returned from voting. The line was long, but moved fast considering that the ballot contained 10 races, 20 judicial retention questions, 11 bonds, 2 mill levies, and 6 amendments to either the state Constitution or the city Charter. The bilingually printed ballot was HUGE, but I heard no complaints. People were cheerful going in and more so coming out!
I hope Kevin is right, but I don't share his optimism. Besides, the real problems are the blatantly undemocratic features of our quasi-democracy: the Electoral College, the undemocratic representation in the Senare (which influences the selection of judges and justices), and gerrymandering. All these features can and have been used to subvert democracy. Even if Trump is thwarted this time, he has shown the way for more determined autocrats that will follow to exploit the vulnerabilities in our Constitution to highjack democracy.
The features that you mention are certainly a problem. But I think far less so than wealth & powers ability to control information & indoctrination/socialization infrastructure (media & education). A process that has existed in all societies for millennia, has been greatly amplified by technology, and will continue to be by datamining, AI, & surveillance (virtual & physical).
Let us first resolve the problems that have festered since the eighteenth century. Solving them might even help resolve the ones that you mentioned.
I hope Kevin is right, but I don't share his optimism.
Nor I. I'm not pessimistic in the "I think she's going to lose" sense. She may win! I'm pessimistic in the "her odds look similar to a coin toss." (At least as of now...it's obviously possible we could yet have one, last momentum shift—hopefully in her favor).
Just checked 538 and wow, the race has definitely tightened in the past week or so. Virtually all the swing states are showing as dead even or with a 1 or 2 point Trump advantage. Kamala's national advantage is less than 2%.
I guess you can take solace in the fact that Trump's unfavorable is 52.2% and Kamala's is 47.2. Maybe that's how the election goes? Or in the fact that when you add his 46.4 to her 48.1 you only get 94.5. So maybe most of that 5.5% goes to Kamala and she get's 52% of the popular vote.
But that's a lot of cherry picking. Good lord. I was really hoping we'd be in a much better place right now.
Don't worry. Alan Lichtman has assured us Harris will win.
/s
R polls flooding the public, diluting the quality of the signal. People aren't flipping back and forth. People are tired of answering polls.
The only shifts are from undecided people who are tuning into politics just before the election, and as noted, most have decided on Harris.
The rest is up to getting out the vote.
WaPo suggests early voting is going much better for Trump than the last two cycles:
https://www.washingtonpost.com/politics/2024/10/22/early-voting-republicans-democrats-trump-harris/