Why do Republicans think polls can't underestimate Democrats?

[EJ24]

I can understand the Biden campaign thinking polls are wrong after 2022.

Take a look at the RealClearPolitics average of polls in the Nevada Senate race 2 years ago.

The last few read like this:

Laxalt (R) +5
Laxalt +3
Laxalt +2
Laxalt +6
Laxalt +5

RCP Average: Laxalt +3.4

Final Result
Cortez-Masto (D) +0.9


That's only one example, but I remember everybody expecting a red wave similar to 2010 or 2014, but it was nothing close to that, and most of the polling grossly overestimated the Trump-endorsed candidates.

Why do Republicans seemingly pretend this didn't happen?

[SnowLabrador]

Equating '24 to '22 doesn't make any sense given that Biden is actually on the ballot this time. He's the one dragging other Democrats down in polling.

[ProgressiveModerate]

“Because midterms are different and a bunch of low propensity Trump voters who don’t vote in midterms will show up in 2024”

There may be some truth to this but it’s too superficial for my liking

[EJ24]

“Because midterms are different and a bunch of low propensity Trump voters who don’t vote in midterms will show up in 2024”

There may be some truth to this but it’s too superficial for my liking

I'm willing to bet there's just as many low propensity anti-Trump voters, so I really don't think this is a good talking point.

[Progressive Pessimist]

The 2022 midterms are entirely why I am not as panicked as I would be.

It was very enlightening to see polls underestimate Democrats.

Is it certain to be the case this year? No, but it is a possibility. Perhaps even a strong possibility given how absurdly well Trump is doing according to polls.

[President Johnson]

I honestly think it's a good deal of wishful thinking. Democrats were "guilty" of this in 2020, too, and maybe are to a degree this year by pointing towards 2022.

As for this year, the things are pretty shaky and inconsistent, so from either side you find certain indicators and data that apparently back your preferred narrative.

[Arizona Iced Tea]

Weren't polls accurate in 2018 and then suddenly off again in 2020?

[Progressive Pessimist]

Weren't polls accurate in 2018 and then suddenly off again in 2020?

It doesn't necessarily mean that pattern will follow.

[Cyrusman]

It’s possible but it’s been a while since polling underestimated Dems in a significant manner in a presidential year. The only real examples are midterms such as 06, and 22.

Also, in what world were 2010 and 2014 not red waves?
Dems lost 63 house seats and I think 7 senate seats in 2010 and like 7-8 senate seats in 2014 while the GOP was winning like 247 house seats
If that’s not a wave, idk what is.

[LAKISYLVANIA]

They surely can, the issue is that Democrats are banking on the polls underestimating them which is generally not a good or reliable strategy, and this question can also be asked in reverse: Why do Democrats think polls can't underestimate Republicans or be accurate?

Besides, just look downballot. Democrats are crushing it there.

[Amenhotep Bakari-Sellers]

Due to fact Rs are trolling Ds that Trump is gonna win that's the J6 mindset that Ds can't win although Harris is a terrible fit for VP

[Mr. Smith]

They surely can, the issue is that Democrats are banking on the polls underestimating them which is generally not a good or reliable strategy, and this question can also be asked in reverse: Why do Democrats think polls can't underestimate Republicans or be accurate?

Besides, just look downballot. Democrats are crushing it there.

But what are the odds of a genuine split ticket?  

Similar things could have been said in 2016, but instead Kander couldn't quite do it, and Toomey got re-elected. Democrats were a shoo-in to take back the Senate it seemed, even as Hillary struggled to lock it all up.

Then Hillary ended up ahead of them.

[cherry mandarin]

Why do Republicans think polls can't underestimate Democrats?

Of course they can. Sometimes they have, and sometimes they do. But Trump has outperformed his polling in both of his campaigns, including a massive overperformance in his one race against his 2024 opponent, and Biden massively underperformed in his one race (which happened to be against his 2024 opponent).

The polls might favour either one of them. At this point, we can't tell for certain yet. But the history can sometimes be instructive, so if I had to guess, I'd say Trump is likelier to beat his polling than Biden is.

“Because midterms are different and a bunch of low propensity Trump voters who don’t vote in midterms will show up in 2024”

There may be some truth to this but it’s too superficial for my liking

What makes you say that?

I'm willing to bet there's just as many low-propensity anti-Trump voters, so I really don't think this is a good talking point.

The data suggests otherwise. Of course it's possible that Dems will have a turnout advantage with this group, it just seems unlikely. We know the composition of the 2022 electorate was more Dem-favourable than what we all expect the 2024 electorate to look like.

It doesn't necessarily mean that pattern will follow.

You're right, but it is a valid argument that it could happen again (although I do think it's a relatively weak one).

But what are the odds of a genuine split ticket?

Similar things could have been said in 2016, but instead Kander couldn't quite do it, and Toomey got re-elected. Democrats were a shoo-in to take back the Senate it seemed, even as Hillary struggled to lock it all up.

Then Hillary ended up ahead of them.

I think this example is pretty instructive because there were indeed fewer voters intending to split their tickets as the campaign went on, but this was the case because polling overestimated down-ballot Dem candidates rather than underestimating the Dem at the top of the ticket. Same thing could happen again this year, where Biden's unpopularity drag down Brown, Tester, and maybe some other incumbents too.

Of course the opposite could happen, but we have generally tended to witness fewer examples of "reverse coattails" than the traditional presidential coattails bleeding down from the top of the ticket.

[Ferguson97]

This doesn’t even really make sense even if you only look at presidential elections. A sample size of two is not really significant.

[S019]

Nevada is a pretty poor state to use as an example, given that it is notorious for really bad polling. As for 2022, when a prediction calls for 237 seats (as Sabato did), that isn't a number set in stone and there can be some fluctuation, the number landing at 222 just indicates a few competitive seats broke towards the Dems that weren't expected to (and a lot of these seats were basically pure coin flips that wouldn't have been rated in the first place, if Sabato didn't pick leans for all their tossups). Cook and IE's final predictions did have Republicans below 218 with tossups holding the balance (in elections like 2010+2014 at least one of them called Republican majorities, which was a strong sign that the median district would vote significantly Republican in those elections).

[LAKISYLVANIA]

They surely can, the issue is that Democrats are banking on the polls underestimating them which is generally not a good or reliable strategy, and this question can also be asked in reverse: Why do Democrats think polls can't underestimate Republicans or be accurate?

Besides, just look downballot. Democrats are crushing it there.

But what are the odds of a genuine split ticket?  

Similar things could have been said in 2016, but instead Kander couldn't quite do it, and Toomey got re-elected. Democrats were a shoo-in to take back the Senate it seemed, even as Hillary struggled to lock it all up.

Then Hillary ended up ahead of them.

I have no clue about the downballot races. I'm a bit in doubt there as well on how to interpret it and what is more likely to happen.

I guess one thing we need to keep in mind is that a lot of races have Democratic incumbents and that the Republican challengers in most cases barely have any name recognition and that the senate campaigns haven't really started at full intensity. I suspect a lot of those races - especially if Trump maintains this polling advantage - would start to tilt towards a narrative of more competitive races in the senate, because I have a hard time believing this amount of split voting as well.

I do think though 2016 in part was also because people were expecting Clinton to be elected and some would've voted downballot Republican to not give her too much power as both candidates were disliked. If the narrative would be that we likely would get president Trump, republicans might not have as much of an advantage there. And people likely would think twice before sending a Republican to the senate.

I suspect the races will tighten but that in the swing states Democrats likely would end up winning for the most part but by narrower margins than currently predicted, unless Trump maintains +5 leads in AZ and NV in which i think these seats would be more in danger. And MI is also risky since it is an open race. But Slotkin is favoured I believe even if Trump wins by a slim margin. I'm less optimistic about Ohio and Montana though. Politics are too polarized on a partisan level now, and this is unlike 2018 a presidential cycle.

I also suspect a lot of self-declared independents or third party voters might also favour the Democrat downballot.

[Amenhotep Bakari-Sellers]

As I said earlier the Maga argument is to think no one can beat Trump they wantta sew seeds of doubt in your head

[Dan the Roman]

I think it is perfectly possible polls are underestimating Biden and in fact i would probably bet money on it. However, the issue is not whether they are underestimating Biden but whether they are underestimating him by a degree almost unprecedented in history. This is not a situation where Biden is down 1-2 in Georgia, Arizona, Nevada, etc. It is one where the disagreement is whether he is down 3-4 or 9+.

So I absolutely don't believe Trump is up 13 in Nevada or 9 in Georgia. But I also think a lot of Democrats are in denial that just because there is something wrong with polls showing Trump headed for a landslide that this somehow is evidence Biden is ahead.

Biden is currently polling worse than Trump was in 2016.

And that Laxalt error from the top of this thread? Even if repeated with current polling it would still have Trump winning Georgia/Nevada/Arizona.

[oldkyhome]

“Because midterms are different and a bunch of low propensity Trump voters who don’t vote in midterms will show up in 2024”

There may be some truth to this but it’s too superficial for my liking

There's no truth to it. Republicans beat out Democrats by turnout in 2022. They won by persuasion. But there may be some truth to the notion that independents are more willing to vote for a generic Democrat than Joe Biden.

[AncestralDemocrat.]

“Because midterms are different and a bunch of low propensity Trump voters who don’t vote in midterms will show up in 2024”

There may be some truth to this but it’s too superficial for my liking

It's very superficial to suggest that the polls underestimated dems in a midterm.. thus they will be underestimated in a presidential election.

They're both inherently different.

Anyway, generic ballot polling literally didn't overestimate republicans at all in 2022.

[Amenhotep Bakari-Sellers]

“Because midterms are different and a bunch of low propensity Trump voters who don’t vote in midterms will show up in 2024”

There may be some truth to this but it’s too superficial for my liking

It's very superficial to suggest that the polls underestimated dems in a midterm.. thus they will be underestimated in a presidential election.

They're both inherently different.

Anyway, generic ballot polling literally didn't overestimate republicans at all in 2022.

Yeah it's different because the Rs lost so you know Biden Approval was at 45% Trafalgar had his APPROVAL AT 39% and they Pred a Red wave Da always land where the Approval ratings lie

[kwabbit]

“Because midterms are different and a bunch of low propensity Trump voters who don’t vote in midterms will show up in 2024”

There may be some truth to this but it’s too superficial for my liking

There's no truth to it. Republicans beat out Democrats by turnout in 2022. They won by persuasion. But there may be some truth to the notion that independents are more willing to vote for a generic Democrat than Joe Biden.

Both things are true, paradoxically. The electorate had good turnout for registered Republicans, but a disproportionate amount of those Republicans had voted for Joe Biden and were especially receptive to Democratic messaging. The midterm electorates in PA and WI, for example, were won by Joe Biden despite registered R turnout advantage. While Evers and Fetterman both had decent persuasion gains, they weren't huge Blue Wave levels implied by the registered turnout advantage.

[cherry mandarin]

A sample size of two is not really significant.

When it comes to Trump, it's all we have.

I do think though 2016 in part was also because people were expecting Clinton to be elected and some would've voted downballot Republican to not give her too much power as both candidates were disliked. If the narrative would be that we likely would get President Trump, Republicans might not have as much of an advantage there. And people likely would think twice before sending a Republican to the Senate.

With all the obstructionism in DC over the past decade, people are splitting their tickets less than ever, as they realize they need to give the president's party majorities in both chambers of Congress rubber stamp in order to get pretty much anything done.

This is not a situation where Biden is down 1-2 in Georgia, Arizona, Nevada, etc. It is one where the disagreement is whether he is down 3-4 or 9+.

And that Laxalt error from the top of this thread? Even if repeated with current polling it would still have Trump winning Georgia/Nevada/Arizona.

The polls showing Trump up 9 points in those states are outliers. Trump's leads in all three are closer to the 3-4 point figure you gave than to 9 points right now.

Furthermore, the national polling is a dead heat right now, with neither candidate up by more than a point or two currently. Sure we elect presidents using the Electoral College rather than the popular vote, but there are plenty of indicators to suggest Trump's EC advantage will be far narrower this year than it was in either 2020 or even 2016—that is, if it still favours him at all.

More importantly, the NV+AZ+GA combination alone won't be enough for Trump to win. Looking at various polling aggregators, Trump seems to lead by less than 2 points in the apparent tipping-point state, whether that is Wisconsin, Michigan, or Pennsylvania. This is far from being a wide lead by any means, much less a safe one.

Anyway, generic ballot polling literally didn't overestimate Republicans at all in 2022.

Yes but what matters in the presidential race isn't the result of the generic ballot or the popular vote, it's what happens in the key swing states. And by and large, Democrats overperformed their polling numbers in those states in 2022.

[David Hume]

They surely can, the issue is that Democrats are banking on the polls underestimating them which is generally not a good or reliable strategy, and this question can also be asked in reverse: Why do Democrats think polls can't underestimate Republicans or be accurate?

Besides, just look downballot. Democrats are crushing it there.

The Senate Dems are doing quite well in the polls. This makes me think that a lot of Dem leaner will go back to support Biden when the election is close.

[LAKISYLVANIA]

They surely can, the issue is that Democrats are banking on the polls underestimating them which is generally not a good or reliable strategy, and this question can also be asked in reverse: Why do Democrats think polls can't underestimate Republicans or be accurate?

Besides, just look downballot. Democrats are crushing it there.

The Senate Dems are doing quite well in the polls. This makes me think that a lot of Dem leaner will go back to support Biden when the election is close.

Too many undecideds in those polls to really suggest that. Again, barely anyone is probably familiar with who the GOP challenger will be or who he is in those states since basically all states have Democratic incumbents.

Wouldn't surprise me if most of those undecideds would eventually go to GOP.