NYT inadvertently indicates that Biden is probably pulling away

[babylonsister]

https://www.dailykos.com/stories/2024/5/18/2241402/-NYT-inadvertently-indicates-that-Biden-is-probably-pulling-away

NYT inadvertently indicates that Biden is probably pulling away
Saturday, May 18, 2024 at 4:59:23p EDT
J Ash Bowie

The New York Times must have access to some data showing that Biden is starting to pull away (or perhaps that Trump is falling behind), because their Saturday online “front page” is working hard to drag Biden down. Here are the “above the fold” stories up right now:

Headline: In His Beloved Philadelphia, Biden Faces Wariness From Black Voters
Subheader: Even in the president’s favorite political stomping ground, his standing has slipped with Democrats who will be vital to a repeat victory, interviews with nearly two dozen Black voters showed.

Headline: Biden Flipped Georgia in 2020. This Year Could Be Different.
Subheader: His narrow win there in 2020 was seen as a sign of Georgia’s emergence as a battleground state. But in 2024, President Biden faces a changed landscape there.

And the top Opinion piece is: Why the Manhattan Trial Is Probably Helping Trump
By the odious Ross Douthat.

Sulzberger must be getting nervous….

[UPDATE]
Sunday morning now, and the top Opinion piece is from Ezra Klein, Seven Theories for Why Biden Is Losing (and What He Should Do About It). The NYT is definitely nervous that this race is turning into a boring and non-profitable landslide.

[MutantAndProud]

1. Next up we'll hear about

The ‘consequences’ of certain Biden policies being enacted, which is totally not a NY mob threat in code…

[TwilightZone]

2. That doesn't make any sense.

As people are always pointing out, the NYT has been posting articles like that for months. It's not a sign that Biden is suddenly pulling away -- it's the status quo.

What has changed is that the number of people predicting a Biden landslide based on no evidence (yet) seems to be increasing. One poster claimed Biden is going to get 70-75% of the popular vote, which is laughable. I think he ultimately wins, but I also think this race is going to be much closer than many presume. What should happen and what will happen are often not the same thing, particularly when Trump is involved.

There's a fine line between optimism and overconfidence, and some are veering toward the latter. Shades of 2016.

[Lovie777]

3. Nope........................

But what it is, my opinion, about the extreme poor quality of the current USA polls, Democrats are keeping their eyes on the prize.

[elocs]

12. 2016. Too many Democrats just don't get it. Save the deep thinking and just vote for the party nominee.

Here in Wisconsin in '16, too many Democrats were so overconfident that Hillary would win that Jill Stein of the Green Party received more votes than Trump's margin of victory and I'm certain that most of those votes came from Democrats who felt free to vote for Stein since Hillary had the election in the bag.

[Lonestarblue]

4. And today it's Biden having trouble in traditionally blue Nevada.

[mopinko]

5. ezra klien shd keep his wet dreams to himself.

jayzuz i’m tired of that asshat.

[everyonematters]

6. When the NYT came out with it's poll, the pollster was Siena.

Emerson came out with polls on swing states about a week earlier with similar results. We have a lot of work ahead of us if we are going to reelect Biden. Conspiracy theories aren't going to help any.

[getagrip_already]

8. And yet, primary polls missed by outside the margin of error numbers

It seems actual voters don't agree with the polls.

The voters must be defective because we keep hearing here that the polls are accurate.

It's those darned voters. We should do away with them and just elect politicians using polls published by the nyt.

[everyonematters]

9. The primary polls aren't Republican vs Democrat.

In 2020 most of the reputable swing state polls showed Biden winning. He won them but by lower margins because they underestimated the Trump vote. Now they are showing Trump winning.

[getagrip_already]

10. So what?

They are polls conducted under the same criteria, and they are way off.

The election polls follow a pattern if you watch them long enough. When the election draws close, the polls narrow to the point that the results can't help but fall within the margin of error unless something truly unexpected happens.

They can't be wrong by design. Pollsters see to it.

[Ruby Soho]

15. Primary polls weren't universally way off.

There are two types of incorrect polling:

1. Margin-based but the poll leader still wins.

2. The candidate leading in the polls loses.

They're vastly different.

The primary polls generally still predicted the basic outcome of the race: the leading candidate still won.

I don't recall any primary poll that had Trump winning but Haley actually prevailed.

Just look at Iowa.

The final average of polls had Trump winning 52.5 (he won 51%), Haley 18.8 (she won 19.1%), DeSantis 15.7 (he won 21.3%).

The biggest outlier was DeSantis, who improved on his average by about five points.

But he didn't win. Trump basically hit his polling average.

What about New Hampshire?

The average of polls had Trump winning 55.8 (he won 54.4), Haley was at 36.5 (she won 43.2) and DeSantis at 8 (but he got .7).

So at face value, you could say Haley overperformed by a decent amount. But during the polling, DeSantis dropped out. It's clear that his voters shifted to Haley more than Trump. Regardless, the polling was correct on Trump's total. It predicted Trump would win and Trump won.

Finally, let's look at South Carolina.

The final average of polls had Trump at 60.8 and he won 59.8. Haley polled at 37.5 and she won 39.5.

Now that it was more or less a two-person race, the numbers start to line up with the actual results.

Trump did basically one-point worse and Haley two-points better for a net difference of three-points. That's typically what the average MOE will be in a poll (+/- 3 or so).

I would not call any of these results way off.

To me, way off is the polls predicting Trump to win New Hampshire and he actually loses.

That's happened. It happened in 2008 with Obama. He was the polling favorite vs Clinton and she shocked the polls.

Haley didn't shock in Iowa or New Hampshire or South Carolina, which is her own home state.

Finally, in regards to your final assessment that pollsters generally adjust their polling and make the results closer toward the end of the campaign, do you have evidence of this?

Sure didn't happen in 2020.

That's what's so confusing to me. It's only been four years. Did the media just decide four years ago they liked Biden so they didn't cook their polls?

The race didn't tighten according to the polls in 2020. Not by much anyway.

On May 20, 2020, Biden's average lead was 4.9 points.

On June 20, 9.5 points.

On July 20, 8.6 points.

On August 20, 7.4 points.

On September 20, 6.5 points.

On October 20, 8.9 points.

On election day? 7.2 points.

So, his lead grew a bit in the summer from roughly 5 points in May to nearly nine points in both July and October. But overall, the margin of his lead stayed between 5 and 9 points. That's pretty consistent for polling.

But again, were the pollsters manipulating the polls in favor of Biden?

Funny how I don't recall many suggesting that four years ago.

It's the same pollsters. What's changed?

And are you saying all the pollsters all colluding to bring down Biden?

Because there's essentially two claims in your post:

Polling is just inherently not accurate anymore.

Pollsters make up the data to drive the narrative. That's what you're suggesting with your last comment.

So, which is it and did they just make up polls in 2020?

Also: what does Emerson College, I'd say a pretty liberal college out of Boston, gain by making up their polls?

Because their polls show the same thing as Morning Consult and NYT and others that apparently fudge their numbers.

I don't see it.

I think the polls are showing us exactly what this race will be: very close with two unpopular candidates. And that's typically what the polls show. Especially in Wisconsin, Pennsylvania and Michigan.

[FBaggins]

13. Margin of error is a statistical measure that does not include all forms of error

Primaries receive far less polling, have far higher undecided populations, and are much harder to pin down on the "likelihood" question.

[Ruby Soho]

16. Especially when candidates suspend their campaigns between primaries...

...which happened between Iowa and New Hampshire with DeSantis. You're going to struggle catching the shifts in polling in the short period of time between primaries.

[FakeNoose]

7. It will be a landslide and NYT will be on the losing side

I'm so sick of this shit. Presidential polls need to be banned immediately.

Polls only reinforce the bubble because people pay attention to polls (real or fake) that confirm what they want to hear. Newspapers only use them to make money. Polls are basically worthless.

[RobinA]

11. In Fairness

the Philadelphia article started in the Philadelphia Inquirer sometime around last weekend.

[FBaggins]

14. So bad news is actually evidence of good news that they're hiding?

Now I'm worried.

Quick question. The Harvard Harris poll four years ago this week showed Biden up by six... and now shows Trump up by six. Has their bias changed in some way?

[Ruby Soho]

17. I'm not a poll truther but the Harvard-Harris poll has been a major outlier for months now.

I don't know if their bias changed but I wouldn't be surprised if their adjustments did. That poll has consistently been very favorable to Trump this campaign season and pretty much out of step with most national polls.

The most recent national polls:

FOX: Trump +1
Yahoo: Tied
Morning Consult: Trump +1
Reuters: Tied
I&I/TIPP: Biden +2
NPR/PBS: Biden +2
CNN: Trump +6
Quinnipiac: Tied

That CNN poll is the only one that aligns with Harvard's and that poll is a month old.

Every other poll has either Trump up narrowly, Biden up narrowly or the race tied.

I'm inclined to believe that's more accurate than the Harvard poll.

I'm also suspicious of anything that is connected to Mark Penn, as the Harvard poll is.

[FBaggins]

19. It was something of an outlier four years ago too

Quinnipiac, Monmouth, and ABC/WPost had us up by double digits

There are always outliers in both directions. There are always polls that seem to have odd samples with "too many democrats/republicans". But the totality of the polls tell a consistent story of a several-point shift against us compared to polling four years ago (e.g., Morning Consult's T+1 from this morning compares to B+8 in the poll closest to May four years ago). And those polls were off in our favor by three points.

Anyone who wants to spin that as bias on the part of the polling firms has to explain why they didn't help Trump win last time.

[Ruby Soho]

20. I think it's simple.

Mark Penn has clearly shifted further and further to the right over the years. This isn't some conspiracy, either. It's literally written in his Wikipedia bio. His statements, especially over the last four years, are far more pro-Trump than what he was putting out previously.

The dude tweeted just the other day that Garland is the most partisan attorney general in modern history. He clearly is biased in a way that absolutely could impact the polling he's linked to.

And it's clear, by just following Penn, he's gone further and further to the right over the last four years. That alone raises questions about his ability to produce polling that isn't biased.

Penn's entire twitter feed is spent attacking Biden and his administration.

Finally, I'll just leave this here:

Link to tweet

[FBaggins]

21. I'm didn't realize that Mark Penn controlled all of the polling firms

Devilishly clever that one

[LiberalFighter]

18. The Manhattan trial is helping TSF? Wow

Using coke? Has a worm in his brain?

[SocialDemocrat61]

22. Doubt it will be a landslide

ala 72 or 84. And the best campaigns take nothing for granted. Better to work like we’re a few points behind than assume that we’re ahead.