New York Times: Late deciders are swinging Republican by +9 points…

Revolver News ^ | November 03,2026 | Staff

[Red Badger]

Well, well, well. Wouldn’t you know it?

After months of listening to political experts and the media tell Americans that Republicans were headed for an absolute bloodbath in the midterms, the New York Times has suddenly discovered something rather interesting.

Late-deciding voters are breaking toward Republicans by NINE POINTS in seven critical Senate races.

And this little revelation comes just as Americans are heading to the polls.

How convenient, right?

For much of this election cycle, the message coming from the polling world has been downright miserable for Republicans. Dems were supposedly surging, the GOP was in serious trouble, and one gloomy forecast after another painted a picture of a political disaster waiting to unfold.

But we’ve been watching this polling circus for a long time at Revolver, and we’ve become increasingly suspicious of the way these numbers are packaged and sold to the American public.

Because somewhere along the way, political polling started looking like a very powerful way to influence it.

Think about the impact of repeatedly telling people their guy is doomed. What does that do to enthusiasm, donations, and the overall mood surrounding an election?

It kills it.

And when those numbers come from prestigious polling institutions, they’re treated like the political Ten Commandments and blasted everywhere.

But as we get closer to the actual voting day, things sometimes start quietly changing.

Which brings us to this story.

The New York Times has released its final round of polling, and buried inside all those rather depressing numbers for Republicans is a development that’s getting a lot of attention.

Apparently, the people who’ve been making up their minds most recently are leaning Republican.

JUST IN: Late deciders are swinging Republican by +9 points – NYT

— Ames (@VivaLaAmes11) November 2, 2026

The Times partnered with Siena University to survey voters in seven highly competitive Senate races, including Michigan, Texas, Alaska, and Ohio.

These are some of the contests that could determine which party controls the Senate after Tuesday.

And while the overall results still paint a favorable picture for Dems, the researchers discovered something interesting when they examined voters who had just made up their minds in the past month.

That group was breaking toward Republican candidates.

In a big way.

Newsmax:

Republicans are winning over late-deciding voters by 9 percentage points in seven battleground Senate races, according to a new poll suggesting the GOP is gaining momentum ahead of Tuesday’s midterm elections.

The New York Times/Siena University survey, released Sunday, found that voters who made their choices within the past month favored Republican candidates, even as Democrats maintained leads in six of the seven contests.

“There are still some undecided voters, and voters who decided only in the last month favored Republicans by a nine-point margin,” the Times reported.

After all the dire predictions, there’s apparently a sizable group of voters making their decisions later in the game who aren’t following the gloomy Republican script we’ve been hearing about.

And that brings us to one of our biggest complaints about political polling.

In many cases, they’ve been weaponized. But this isn’t something new, is it? Who can forget this doozy from the New York Times back in 2016:

.

And that’s really the problem with the polling culture we’ve created.

People are being told to treat these polls as gospel. As if they’re telling us what’s going to happen, not what a very particular group of respondents told a very particular group of researchers at a very particular moment.

And once the media starts building a political narrative around the numbers, the narrative quickly becomes bigger than the poll.

And we all remember the 2016 presidential election, when virtually every important person in the political and media world was convinced Hillary Clinton was about to become America’s first female president.

The over-confidence was actually weird and scary.

And the New York Times was right there, publishing its presidential forecast with an eye-popping probability of victory for Clinton.

Why even bother to vote, guys. Hillary had this one in the bag, according to every poll on planet earth.

Three weeks later, Donald Trump was elected president.

Even so, Americans spent that election season being told the outcome was a foregone conclusion delivered from Mount Sinai.

And after most of the polls were proven wrong, we were told that it was some kind of fluke error that had to do with cell phones and landlines, and it’d be fixed.

Tech Psych:

First of all, the polling agencies do not and probably cannot sample a representative sample. They typically poll those who have landline, as opposed to cellular, telephone service. This skews to older people, as many if not most young adults do not have landlines. However, if they poll only those who have cell phones, this will skew to younger people, and older voters will be lost. The same is true with Survey Monkey polls, as this will skew to younger voters because younger people are more comfortable with computers than older people. This method also skews to more urban and suburban people and away from rural people.

Second, not only are any of these methods questionable, there is also the problem of who answers the poll. For example, I have a landline, but I never answer it unless it is from someone I know. If it is a pollster, I will not answer it. So what kind of people answer a pollster? We don’t know, and we don’t know how representative these people are. Secondarily, pollsters call multiple people at once, and when one person answers the phone, all of the other calls are dropped, so again, what kind of people are answering the poll, and how representative are they?

But was polling ever really broken, or is this exactly what it’s always been? A political weapon, largely wielded by the left, to undermine Republican candidates, dry up their fundraising, and convince voters there’s no point in showing up because the election is already lost?

Something to think about.

We’ll see how things shake out on Tuesday, and something tells us we’ll have plenty to talk about afterward, from the polls and vote counting to surprise upsets and voter turnout.

In the meantime, whatever else you’ve got going on today, make sure you get out there and cast your vote.

---

TOPICS: Business/Economy; Front Page News; Government; Politics/Elections; US: New York
KEYWORDS: psyop

[z3n]

This falls in line with what I predicted. *fingers crossed*

Months and months of non-stop fake-polls to try to influence the election, and at the last minute they have to do damage control to salvage some credibility.

[gundog]

I guess “late deciders” suits the NYT better than “folks trying to beat the cheat.”

[maddog55]

This just means the democrat ballot drop needs to be 10%.

[OttawaFreeper]

Looking at an earlier thread today, I noticed that the economic numbers looked actually quite good, so I can not imagine a “it’s the economy, stupid” type of scenario playing out on a significant scale.

[Bon of Babble]

The public should have learned from these major pollsters in 2016 and 2024 how wrong they were.

Their polls showed Hillary winning in 2016 and Harris taking at least five of the seven swing states in 2024.

Trump won in 2016 and took all of the swing states in 2024.

[kaktuskid]

It’s all about the price of gas, diesel or food. Or fertilizer in the Midwest.

[personalaccts]

To little to late in NC we have a mountain to climb

[bobcat62]

The recently deceased will be swinging Democrat by 10 points.

[ProtectOurFreedom]

"... there’s apparently a sizable group of voters making their decisions later in the game"

I think that is an assumption without facts in evidence. Is there REALLY such a thing as an "undecided" voter these days? You are either in favor of communism or Freedom & Liberty. You are either in favor of forced conversion to Islam or freedom to choose your own faith.

[MinorityRepublican]

Months and months of non-stop fake-polls to try to influence the election, and at the last minute they have to do damage control to salvage some credibility.

We'll find out soon but I think they motivated Republicans to show up today to vote lol.

[webheart]

I certainly don’t rush to get my ballot in early. I don’t trust that it will be properly handled if it gets to be put in a box and stacked in a corner somewhere.

[webheart]

I certainly don’t rush to get my ballot in early. I don’t trust that it will be properly handled if it gets to be put in a box and stacked in a corner somewhere.

[ProtectOurFreedom]

Another assumption: "The pollsters typically poll those who have landline, as opposed to cellular, telephone service. This skews to older people, as many if not most young adults do not have landlines."

It's an old trope: grandma answering the call from the pollster:

Does ANYBODY really use a landline any more? Or did they keep just in case the mobile service goes south? We ditched our landline years ago. We almost never used it for voice calls, either. It was mainly for the security system to make calls, but we converted that to the Internet over a decade ago.

[Bob434]

Months and months of non-stop fake-polls...

AND

Fox News’s constant doom and gloom trying to dissuade republicans from voting

[Fury]

WHAT SPECIFIC races is the +9 points referring to?

Good grief.

[Jamestown1630]

I don’t know how anyone can be a ‘late decider’. There’s never been an election where the differences between the two sides were so stark.

[oil_dude]

WOW, a random act of TRUTH form the NYT???? What’s next, dogs and cats living together?

[1Old Pro]

I don’t know how anyone can be a ‘late decider’.

I hope that people (Stevens voters) in Michigan who would have voted Sayad are now voting Rogers due to Stevens endorsement. That's a late decision perhaps - we need 1-2% of the vote here to switch to help Rogers. That would be a 2-4% vote swing.

[spintreebob]

78% of US adults live in a wireless phone only household.
1 or 2% live in a landline only household.
21% live in a household with both types of phones.

It costs extra for a polling firm to take the extra step to determine if a phone is landline or cell...and if a household is only one or the other. These polling firms are cheap and do not spend extra money for such nonsense.

They do not know the nature of the phone they are calling.
More likely, “for demographic reasons”, they know much more about the person who might answer the phone than the phone itself.

But how many times is a phone shared by 2 or 5 persons in the household? Some polling firms will poll only the person they intended to poll. Most polling firms will talk to anyone who answers the phone.

Regardless of how the poll is conducted, “normal” people want to give the correct answer...which is the answer wanted by the pollster. The questions leading up to the money question set the context and greatly influence the answer to the money question.

I’ve taken in person polls where I purposely phrase the context and the money question differently. I can get a big swing one way or the other based solely on 1) the context set by previous questions. and 2) the wording of the money question.