Trump and Biden are deadlocked in six key 2020 election states, CNBC/Change Research poll finds

[Chomsky]

President Donald Trump and former Vice President Joe Biden are locked in a tight race in six states that will shape who wins the White House in November, according to a new CNBC/Change Research survey.

The Republican incumbent and apparent Democratic nominee are virtually tied in the battlegrounds of Arizona, Florida, Michigan, North Carolina, Pennsylvania and Wisconsin, according to the States of Play poll of swing states released Wednesday. Trump edges Biden by a 48% to 47% margin across those states, the poll found.

The poll, which surveyed 5,787 likely voters in the six states on April 17 and 18, has a margin of error is plus or minus 1.3 percentage points.

Source: Trump and Biden are deadlocked in six key 2020 election states, CNBC/Change Research poll finds

Well look at this, and yes I'm a bit surprised.

Let's not forget this is a poll of 'likely voters'. Of the three general national poll types ('American's', 'registered voters', 'likely voters'), it is thought to be the most accurate of the three in predicting election results.

A 'likely voters' poll should better describe the upcoming election than 'approval' polls. So while I've generally been seeing the approval polls slightly favoring Biden, it's the 'likely voters' that count. And it's the swing states that count the most.

This poll sets back my thinking a bit, in my assumption that Biden was looking likely to win MI, and probable to win WI.

But it's only one poll, and we'll have to see if we get confirming (likely voter) polls. If so, we still have a real horse race on our hands!

 

[ecofarm]

Without a breakdown by state, it could be considerable leads in some and considerably behind in others.

 

[Chomsky]

Without a breakdown by state, it could be considerable leads in some and considerably behind in others.

I took this below, to mean all six of the mentioned states were within the margin of error:

The Republican incumbent and apparent Democratic nominee are virtually tied in the battlegrounds of Arizona, Florida, Michigan, North Carolina, Pennsylvania and Wisconsin, according to the States of Play poll of swing states released Wednesday.

But the aggregate polling results returned put Trump up a single point (though still 'tied' withing the margin of error).

Trump edges Biden by a 48% to 47% margin across those states, the poll found.

 

[HumblePi]

Source: Trump and Biden are deadlocked in six key 2020 election states, CNBC/Change Research poll finds

Well look at this, and yes I'm a bit surprised.

Let's not forget this is a poll of 'likely voters'. Of the three general national poll types ('American's', 'registered voters', 'likely voters'), it is thought to be the most accurate of the three in predicting election results.

A 'likely voters' poll should better describe the upcoming election than 'approval' polls. So while I've generally been seeing the approval polls slightly favoring Biden, it's the 'likely voters' that count. And it's the swing states that count the most.

This poll sets back my thinking a bit, in my assumption that Biden was looking likely to win MI, and probable to win WI.

But it's only one poll, and we'll have to see if we get confirming (likely voter) polls. If so, we still have a real horse race on our hands!

Trump is probably the first president in history that won an election by a sliver. He lost the popular vote, he won the electoral college because he won a handful of states by the narrowest of margins. He saw in that victory that America loved him. Most presidents would see they had work to do especially when you factor in that his wining margin were people that were voting against Hillary Clinton not for him.

Typically this crisis would be a moment for any president to reach out broadly, bring everyone together, not play to the basis of the base. Not use his megaphone to only his loyal supporters but use it to calm people and reassure them. He has failed miserably. He goes through his old playbook. Make MAGA hat wearers happy, play to Fox, and not think about those voters who didn't vote for him and only voted for him because of the alternative.

Without Trump pulling some sort of trick or miracle out of his fat ass, he's not going to win the election again.

 

[ecofarm]

I took this below, to mean all six of the mentioned states were within the margin of error:



But the aggregate polling results returned put Trump up a single point (though still 'tied' withing the margin of error).

I read that to mean the six, total, are said percentage. Yours makes more sense but I'm not sure what they did.

 

[Mycroft]

Source: Trump and Biden are deadlocked in six key 2020 election states, CNBC/Change Research poll finds

Well look at this, and yes I'm a bit surprised.

Let's not forget this is a poll of 'likely voters'. Of the three general national poll types ('American's', 'registered voters', 'likely voters'), it is thought to be the most accurate of the three in predicting election results.

A 'likely voters' poll should better describe the upcoming election than 'approval' polls. So while I've generally been seeing the approval polls slightly favoring Biden, it's the 'likely voters' that count. And it's the swing states that count the most.

This poll sets back my thinking a bit, in my assumption that Biden was looking likely to win MI, and probable to win WI.

But it's only one poll, and we'll have to see if we get confirming (likely voter) polls. If so, we still have a real horse race on our hands!

Strike one: Online, opt-in poll.

Strike two: Low percentage of people sampled from the battleground states.



Strike three: Sampling by Party affiliation is skewed.



Flagrant foul: Excessive over sampling from high COVID-19 states and counties.




POLL REJECTED!!!

 

[Moonglow]

Source: Trump and Biden are deadlocked in six key 2020 election states, CNBC/Change Research poll finds

Well look at this, and yes I'm a bit surprised.

Let's not forget this is a poll of 'likely voters'. Of the three general national poll types ('American's', 'registered voters', 'likely voters'), it is thought to be the most accurate of the three in predicting election results.

A 'likely voters' poll should better describe the upcoming election than 'approval' polls. So while I've generally been seeing the approval polls slightly favoring Biden, it's the 'likely voters' that count. And it's the swing states that count the most.

This poll sets back my thinking a bit, in my assumption that Biden was looking likely to win MI, and probable to win WI.

But it's only one poll, and we'll have to see if we get confirming (likely voter) polls. If so, we still have a real horse race on our hands!

If they were neck and neck in the race then Trump would win what with his massive chins..

 

[Casper]

Source: Trump and Biden are deadlocked in six key 2020 election states, CNBC/Change Research poll finds

Well look at this, and yes I'm a bit surprised.

Let's not forget this is a poll of 'likely voters'. Of the three general national poll types ('American's', 'registered voters', 'likely voters'), it is thought to be the most accurate of the three in predicting election results.

A 'likely voters' poll should better describe the upcoming election than 'approval' polls. So while I've generally been seeing the approval polls slightly favoring Biden, it's the 'likely voters' that count. And it's the swing states that count the most.

This poll sets back my thinking a bit, in my assumption that Biden was looking likely to win MI, and probable to win WI.

But it's only one poll, and we'll have to see if we get confirming (likely voter) polls. If so, we still have a real horse race on our hands!

Wait for October, we have a lot of time for much to happen. My guess is many people will voting Against a candidate verses voting for one. Living History as it unfolds.

 

[Chomsky]

Trump is probably the first president in history that won an election by a sliver. He lost the popular vote, he won the electoral college because he won a handful of states by the narrowest of margins. He saw in that victory that America loved him. Most presidents would see they had work to do especially when you factor in that his wining margin were people that were voting against Hillary Clinton not for him.

Typically this crisis would be a moment for any president to reach out broadly, bring everyone together, not play to the basis of the base. Not use his megaphone to only his loyal supporters but use it to calm people and reassure them. He has failed miserably. He goes through his old playbook. Make MAGA hat wearers happy, play to Fox, and not think about those voters who didn't vote for him and only voted for him because of the alternative.

Without Trump pulling some sort of trick or miracle out of his fat ass, he's not going to win the election again.

No, I don't think the bolded is a foregone conclusion at all. Trump, the GOP, Fox News, Limbaugh, etc., have done a masterful job of creating a fake reality in this country.

I just had a reply here this morning, where a poster shot down my impeccable & esteemed MSN source, calling it 'fake news', and instead he referred me to a media blog run by the Heritage Foundation as his evidence proving me 'wrong'.

We can't underestimate the power of the Trump media cabal, and it's influence on our citizenry. Remember, Trump is in power. His media machine doesn't have to prevail in evidence. It only needs to provide "doubt" & "license". This is how propaganda works in the hands of those in power.

 

[Chomsky]

I read that to mean the six, total, are said percentage. Yours makes more sense but I'm not sure what they did.

Agreed.

 

[Chomsky]

Strike one: Online, opt-in poll.

Strike two: Low percentage of people sampled from the battleground states.

View attachment 67278625

Strike three: Sampling by Party affiliation is skewed.

View attachment 67278624

Flagrant foul: Excessive over sampling from high COVID-19 states and counties.

View attachment 67278626


POLL REJECTED!!!

Did you just make the argument that the earlier polls showing Biden leading are more accurate?

Anyway, there's nothing wrong with the sampling spectrum as long as the poll does proper normalization (which we've previously discussed). But regardless, even without normalization the sampling is reasonably representative.

As to your claim of the lower sampling percentages in the key states in relation to the broader poll, that matters not, since the data analyzed & presented in the article is specific to those states discussed. Ditto for your Covid claim.

With all respect, besides your concern with the poll being 'online opt-in' (I was not aware), I don't see the validity of your claims. However, may I ask where you found this detailed data?

 

[Chomsky]

If they were neck and neck in the race then Trump would win what with his massive chins..

:mrgreen:

 

[Chomsky]

Wait for October, we have a lot of time for much to happen. My guess is many people will voting Against a candidate verses voting for one. Living History as it unfolds.

Part of what to watch for will be the number of undecided. Undecideds tend to vote against the incumbent. In 2016 Trump got 2/3 of those undecided going into election day. While HRC technically wasn't an incumbent, she essentially ran as and was perceived as one.

 

[dcsports]

Trump is probably the first president in history that won an election by a sliver. He lost the popular vote, he won the electoral college because he won a handful of states by the narrowest of margins. He saw in that victory that America loved him. Most presidents would see they had work to do especially when you factor in that his wining margin were people that were voting against Hillary Clinton not for him.

Typically this crisis would be a moment for any president to reach out broadly, bring everyone together, not play to the basis of the base. Not use his megaphone to only his loyal supporters but use it to calm people and reassure them. He has failed miserably. He goes through his old playbook. Make MAGA hat wearers happy, play to Fox, and not think about those voters who didn't vote for him and only voted for him because of the alternative.

Without Trump pulling some sort of trick or miracle out of his fat ass, he's not going to win the election again.

This was hardly the first unusually close win. Al Gore might disagree with you, for starters. There's also Bill Clinton's first presidency, where no one received the majority, and arguably he would have lost if not for Ross Perot siphoning off votes.

You can blame Trump, or say that he 'lucked' into the presidency, but he did win. All Democrats had to do to win was field a good candidate, and they couldn't find one. Biden would have been a far better choice four years ago, but I don't know that he's the answer now. Certainly, not even Democrats are excited about him.

 

[Helix]

seems like a pretty clear choice between insufferable, dangerously incompetent authoritarian early dementia guy vs regular early dementia guy. way to go, two party system. either way, i know which early dementia guy i'm voting for.

 

[dcsports]

Part of what to watch for will be the number of undecided. Undecideds tend to vote against the incumbent. In 2016 Trump got 2/3 of those undecided going into election day. While HRC technically wasn't an incumbent, she essentially ran as and was perceived as one.

I think this is going to boil down to how the next 6 months go. If we are open for business, back to 'normal', and the economy has rebounded, Trump will win. If we're still in lockdown, or in a recession, Biden will. I'm not sure anything else matters.

 

[Mycroft]

Did you just make the argument that the earlier polls showing Biden leading are more accurate?

Anyway, there's nothing wrong with the sampling spectrum as long as the poll does proper normalization (which we've previously discussed). But regardless, even without normalization the sampling is reasonably representative.

As to your claim of the lower sampling percentages in the key states in relation to the broader poll, that matters not, since the data analyzed & presented in the article is specific to those states discussed. Ditto for your Covid claim.

With all respect, besides your concern with the poll being 'online opt-in' (I was not aware), I don't see the validity of your claims. However, may I ask where you found this detailed data?

I got this data from links within the OP's link.

btw, what makes you think the poll did "proper normalization"?

For a poll to claim their results reflect opinions in various states, it would be to their advantage to poll high numbers of people in those states. Eight percent of the people polled being from Arizona means an extremely low number of Arizonans were polled. That reduces the accuracy of the poll.

 

[Ikari]

Source: Trump and Biden are deadlocked in six key 2020 election states, CNBC/Change Research poll finds

Well look at this, and yes I'm a bit surprised.

Let's not forget this is a poll of 'likely voters'. Of the three general national poll types ('American's', 'registered voters', 'likely voters'), it is thought to be the most accurate of the three in predicting election results.

A 'likely voters' poll should better describe the upcoming election than 'approval' polls. So while I've generally been seeing the approval polls slightly favoring Biden, it's the 'likely voters' that count. And it's the swing states that count the most.

This poll sets back my thinking a bit, in my assumption that Biden was looking likely to win MI, and probable to win WI.

But it's only one poll, and we'll have to see if we get confirming (likely voter) polls. If so, we still have a real horse race on our hands!

We'll find out as we get closer to November. But I have stated for some time now that this was the Democrats' election to lose, and they may have found a way to do that with Biden.

 

[HumblePi]

This was hardly the first unusually close win. Al Gore might disagree with you, for starters. There's also Bill Clinton's first presidency, where no one received the majority, and arguably he would have lost if not for Ross Perot siphoning off votes.

You can blame Trump, or say that he 'lucked' into the presidency, but he did win. All Democrats had to do to win was field a good candidate, and they couldn't find one. Biden would have been a far better choice four years ago, but I don't know that he's the answer now. Certainly, not even Democrats are excited about him.

Gore won that election. Florida ****ed it all up for him and the presidency was decided by the S.C. Not the same as Trump's narrow win, not at all. Trump will not be reelected and when he gets desperate enough with the knowledge that he can't win, he'll do something insane to cancel it or at least postpone it indefinitely.

 

[GDViking]

I think this is going to boil down to how the next 6 months go. If we are open for business, back to 'normal', and the economy has rebounded, Trump will win. If we're still in lockdown, or in a recession, Biden will. I'm not sure anything else matters.

Biden will have the Obama's, and whoever he jooses as a running mate campaing hard for him.

I don't buy the dementia thing, however Biden has always been a clumsy speaker so all he needs to do is step back let the better speakers speak for him and let trump continue to be the incompitent moron he is...

 

[Chomsky]

I got this data from links within the OP's link.

Thanks.

btw, what makes you think the poll did "proper normalization"?

It's generally applied by quality pollsters, but I can't find anything specific in this poll.

However, I did find this:

Our Dynamic Online Sampling establishes and continuously rebalances advertising targets across region, age, gender, race, and partisanship to dynamically deliver large samples that accurately reflect the demographics of a population.

This appears to be pre-sampling targeting, rather than post-sampling normalization. So in this specific case with this poll, I am going to give you this one.

However in the future with large & established polls thought to be of high quality, I'm going to assume they use normalization even if their normalization algorithms are not publicly available. Most pollsters consider their algorithms proprietary I.P., and as such they do not make them public. To disqualify a poll on that technicality alone, strikes me as unwarranted.

For a poll to claim their results reflect opinions in various states, it would be to their advantage to poll high numbers of people in those states. Eight percent of the people polled being from Arizona means an extremely low number of Arizonans were polled. That reduces the accuracy of the poll.

Well, they're claiming almost 6K sampled, so AZ's 8% would be in the upper 400 range. Not a lot, but still reasonable enough I think to get into the ballpark.

However, the actually number of respondents is low (IMO). So I think I'm going to give you the nod here too, at least in terms of saying the poll could be more accurate with a higher number of respondents. I do think it's good enough to give us a general window on where we're at though, as long as we realize that window may have some width.

 

[GDViking]

We'll find out as we get closer to November. But I have stated for some time now that this was the Democrats' election to lose, and they may have found a way to do that with Biden.

I disagree, Biden was and is the candidate that trump most feared, with a strong running mate I believe he has this election hands down.

That is provided the election is on the up and up.

Trump has never won a fair election and never will...

 

[dcsports]

Biden will have the Obama's, and whoever he jooses as a running mate campaing hard for him.

I don't buy the dementia thing, however Biden has always been a clumsy speaker so all he needs to do is step back let the better speakers speak for him and let trump continue to be the incompitent moron he is...

I don't buy the dementia thing either. It's extremely difficult to speak crisply off the cuff when every word is recorded and analyzed. Even Obama, who was an intelligent man and a great speech reader, could sound like a bumbling idiot when he went off script.

I do think Biden is going to have to be very visible, and make some great speeches, to have a chance of winning. Hillary tried the strategy you mentioned, and failed miserably. It's the old adage that people tend to vote for the person they think is most presidential. You can't appear presidential if people don't see you. The Democrat candidates have barely been seen for the past 6 months. Biden is going to have to find a way to get exposure.

 

[dcsports]

Thanks.

It's generally applied by quality pollsters, but I can't find anything specific in this poll.

However, I did find this:

This appears to be pre-sampling targeting, rather than post-sampling normalization. So in this specific case with this poll, I am going to give you this one.

However in the future with large & established polls thought to be of high quality, I'm going to assume they use normalization even if their normalization algorithms are not publicly available. Most pollsters consider their algorithms proprietary I.P., and as such they do not make them public. To disqualify a poll on that technicality alone, strikes me as unwarranted.

Well, they're claiming almost 6K sampled, so AZ's 8% would be in the upper 400 range. Not a lot, but still reasonable enough I think to get into the ballpark.

However, the actually number of respondents is low (IMO). So I think I'm going to give you the nod here too, at least in terms of saying the poll could be more accurate with a higher number of respondents. I do think it's good enough to give us a general window on where we're at though, as long as we realize that window may have some width.

These polls always have numbers that are too low. They are also very dependent on sampling issues. If X% of the people have a landline, and Y% of those are in your database, and Z% of those people answer the phone, and then you factor in a response rate...

The media tends to use polls incorrectly anyway. They do have some usefulness for campaigns in measuring the effectiveness of their efforts, and putting together a campaign strategy. They are useless in predicting the outcome this far out.

 

[Plasmaball]

On the one hand this isnt great for trump.
On the other hand it's not bad for trump. He has a lot of time to swing this one way or the other.
The same goes for Biden. But i I think this plays to Biden better right now perception wise. But its razor thin