Were polls more accurate before COVID

[Motorcity]

https://projects.fivethirtyeight.com/polls/president-general/wisconsin/

If you look at polls throughout 2019 and up until COVID hit, the generic Democrat only led Trump by a point or two in most swing states. The generic Democrat and Trump were dead even in states like Wisconsin with Trump leading a few polls. Mind you Trump led most Florida polls before COVID.

A lot of people assume that Trump supporters don't pick up the phone. Or Democrats were more likely to be home during the pandemic to do polls.

Considering the final results were much closer to pre-pandemic polling, can we say pre-pandemic polling was much more accurate of the national mood?

2016 was a bust for polling, but 2018 was spot on. This makes me think that the changes made did make polling more accurate from 2017-2019, its just that the pandemic canceled out any changes pollsters did.

[Del Tachi]

No, the polls were always wrong and it's probably the case that Trump would have been easily reelected had the election been in 2019 or early 2020. 

[Tekken_Guy]

Wasn’t that the case in other races as well?

In the Senate, Daines, Ernst, and Graham were all considered pretty safe. Susan Collins was thought to have a good chance to hold on. Thom Tillis was considered more likely than not to win. And John James was well within reach of victory.

In the House, Joe Cunningham, Abby Finkenauer, Ben McAdams, and Anthony Brindisi were all believed to be highly vulnerable. And all but a couple GOP-held seats were expected to stay Republican.

[Roll Roons]

Wasn’t that the case in other races as well?

In the Senate, Daines, Ernst, and Graham were all considered pretty safe. Susan Collins was thought to have a good chance to hold on. Thom Tillis was considered more likely than not to win. And John James was well within reach of victory.

In the House, Joe Cunningham, Abby Finkenauer, Ben McAdams, and Anthony Brindisi were all believed to be highly vulnerable. And all but a couple GOP-held seats were expected to stay Republican.

That was more because Bullock hadn't entered the race yet. When he got in (which happened to be right around the time COVID started to explode), a lot of people instantly moved it to a tossup.

[Mr.Phips]

Wasn’t that the case in other races as well?

In the Senate, Daines, Ernst, and Graham were all considered pretty safe. Susan Collins was thought to have a good chance to hold on. Thom Tillis was considered more likely than not to win. And John James was well within reach of victory.

In the House, Joe Cunningham, Abby Finkenauer, Ben McAdams, and Anthony Brindisi were all believed to be highly vulnerable. And all but a couple GOP-held seats were expected to stay Republican.

Prior to COVID, Dems were not given a good chance to take the Senate.

[DaleCooper]

Wasn’t that the case in other races as well?

In the Senate, Daines, Ernst, and Graham were all considered pretty safe. Susan Collins was thought to have a good chance to hold on. Thom Tillis was considered more likely than not to win. And John James was well within reach of victory.

In the House, Joe Cunningham, Abby Finkenauer, Ben McAdams, and Anthony Brindisi were all believed to be highly vulnerable. And all but a couple GOP-held seats were expected to stay Republican.

Prior to COVID, Dems were not given a good chance to take the Senate.

A big part of the reason the Dems were so underrated in the Senate compared to the House is because the forecasting models dramatically underestimated how competitive Georgia was at all levels, especially compared to other states.

[Tekken_Guy]

Wasn’t that the case in other races as well?

In the Senate, Daines, Ernst, and Graham were all considered pretty safe. Susan Collins was thought to have a good chance to hold on. Thom Tillis was considered more likely than not to win. And John James was well within reach of victory.

In the House, Joe Cunningham, Abby Finkenauer, Ben McAdams, and Anthony Brindisi were all believed to be highly vulnerable. And all but a couple GOP-held seats were expected to stay Republican.

Prior to COVID, Dems were not given a good chance to take the Senate.

They only won it because of the GA runoffs.

[Motorcity]

Wasn’t that the case in other races as well?

In the Senate, Daines, Ernst, and Graham were all considered pretty safe. Susan Collins was thought to have a good chance to hold on. Thom Tillis was considered more likely than not to win. And John James was well within reach of victory.

In the House, Joe Cunningham, Abby Finkenauer, Ben McAdams, and Anthony Brindisi were all believed to be highly vulnerable. And all but a couple GOP-held seats were expected to stay Republican.

Yeah

Polling from 2018-2019 seemed to be accurate at the congressional level but seemed to get worst in 2020 because of the pandemic

[Podgy the Bear]

Wasn’t that the case in other races as well?

In the Senate, Daines, Ernst, and Graham were all considered pretty safe. Susan Collins was thought to have a good chance to hold on. Thom Tillis was considered more likely than not to win. And John James was well within reach of victory.

In the House, Joe Cunningham, Abby Finkenauer, Ben McAdams, and Anthony Brindisi were all believed to be highly vulnerable. And all but a couple GOP-held seats were expected to stay Republican.

Prior to COVID, Dems were not given a good chance to take the Senate.

They only won it because of the GA runoffs.

And we have Trump to thank for that.

[Pericles]

I find it hard to believe that Covid had no impact on public opinion. It was such a huge crisis, and Trump was seen by a polling error proof majority as having responded poorly. Persuasion isn't that dead. So Trump must have won a hypothetical election without Covid. It is possible that he got a swing to him based on some other factor like the riots or the Supreme Court but he didn't clearly win those issues either. It's also easy to see him winning in early 2020 when things were going pretty well for normal people.

[Pericles]

Wasn’t that the case in other races as well?

In the Senate, Daines, Ernst, and Graham were all considered pretty safe. Susan Collins was thought to have a good chance to hold on. Thom Tillis was considered more likely than not to win. And John James was well within reach of victory.

In the House, Joe Cunningham, Abby Finkenauer, Ben McAdams, and Anthony Brindisi were all believed to be highly vulnerable. And all but a couple GOP-held seats were expected to stay Republican.

Prior to COVID, Dems were not given a good chance to take the Senate.

They only won it because of the GA runoffs.

And we have Trump to thank for that.

Mitch McConnell too for not supporting $2,000 checks even though Trump did. Trump raising the checks issue and so giving Ossoff and Warnock a great message was also part of the problem. When the discussion shifted to the checks, the polling shifted by just enough that it may have made up the Ossoff margin.

[Chips]

No, the polls were always wrong and it's probably the case that Trump would have been easily reelected had the election been in 2019 or early 2020. 

I agree honestly.

[TML]

I think Trump’s presence on the ballot was definitely going to throw a wrench into polls vs. actual results to some degree or another, as it now seems that many Trump supporters distrusted polling firms and thus refused to respond to them entirely. This was evidenced by some places being much more Republican than polls had forecasted (e.g. Northern Maine was about 10 points more Republican than polls had indicated in both 2016 and 2020, and Trump was on the ballot for both occasions).

[Sir Mohamed]

Polls more than 6 months away from an election are hardly accurate anyway and should generally taken with a huge grain of salt. If the election was held in late 2019, Trump may have won (the EC), but that question can't be answered since in late 2019 or early 2020 there was neither a challenger nominated nor did too many voters actually pay attention to the election. Majority of people don't follow politics and elections as closely we do.

[Ferguson97]

This is a bit of a weird question considering that covid was like the main thing about the election

[Samof94]

I think Trump’s presence on the ballot was definitely going to throw a wrench into polls vs. actual results to some degree or another, as it now seems that many Trump supporters distrusted polling firms and thus refused to respond to them entirely. This was evidenced by some places being much more Republican than polls had forecasted (e.g. Northern Maine was about 10 points more Republican than polls had indicated in both 2016 and 2020, and Trump was on the ballot for both occasions).

The “shy Trump voter”. They obviously weren’t shy but didn’t trust polling.

[ultraviolet]

Wasn’t that the case in other races as well?

In the Senate, Daines, Ernst, and Graham were all considered pretty safe. Susan Collins was thought to have a good chance to hold on. Thom Tillis was considered more likely than not to win. And John James was well within reach of victory.

In the House, Joe Cunningham, Abby Finkenauer, Ben McAdams, and Anthony Brindisi were all believed to be highly vulnerable. And all but a couple GOP-held seats were expected to stay Republican.

Prior to COVID, Dems were not given a good chance to take the Senate.

They only won it because of the GA runoffs.

And we have Trump to thank for that.

Mitch McConnell too for not supporting $2,000 checks even though Trump did. Trump raising the checks issue and so giving Ossoff and Warnock a great message was also part of the problem. When the discussion shifted to the checks, the polling shifted by just enough that it may have made up the Ossoff margin.

I am convinced that McConnell blocking the checks is what lost Trump the election. Oh well

[○∙◄☻¥tπ[╪AV┼cVê└]

There was always a shy Trump voter effect.