President Kamala Harris

[Cameroni]

29 minutes ago, Fat is a type of crazy said:

I've seen you make this point before but I don't get it. If you ask a 1000 people - you will get Democrat, Republican, Other, Undecided, - the ratio will change over time changing the results. I don't disagree that polling is tricky - and mistakes have and will be made - but I don't see why a change in those ratios in say a swing state is not somewhat of a guide to who will win because  - your point - undecided's are not reflected in polls. 

 

First of all there is a  risk that a perfectly random sample of a given size may not reflect the characteristics of the population as a whole (known as “sampling error”). In fact, the group of people who participate in any given survey are virtually never an idealised random sub-set of the population that will actually turn out to vote. Instead, they can differ from the eventual mix in important ways, which collectively are known as “non-sampling error”.

 

Then polls are subject to the vagaries of voter turnout. Polls conducted among all adults will include the views of people who are ineligible or not registered to vote. Those limited to registered voters treat all respondents in this group as if they had an equal probability of showing up to vote, which they surely do not. And those that seek to filter out respondents unlikely to vote, or that grant more weight to the views of people who are more likely to show up, can get such calculations wrong. Although no two surveys are identical, ones that use a similar approach to predicting turnout are more likely to wind up with errors of a similar size and direction than are ones that handle it differently. In statistical terms, each of these different methods of turnout projection can produce a “bias”, which is likely to contaminate the results of all the pollsters that use it in a similar way.

 

The group of people pollsters can reach by using live telephone interviewers may have different voting intentions than those they can reach by automated phone calls, or via the internet. Individual pollsters may make methodological choices, such as weighting schemes, that consistently lead to more or less favourable results for a particular political party.

 

Ahead of time, it is impossible to know the direction or size of the bias that each of these characteristics may introduce.

 

Voters who have a soft but consistent preference for one of the two major parties often say they are undecided or planning to vote for a third party. This makes polls in the first half of the year a surprisingly weak predictor of final results. For example, in June 1988, George H.W. Bush trailed Michael Dukakis by 12 percentage points in polling averages (he went on to win by eight). Exactly four years later, Mr Bush led Bill Clinton by ten percentage points, and wound up losing by seven.

 

In more recent years, polling errors have been a bit smaller—but they can still be substantial. In 2000 Mr Bush’s son George W. saw his ten-point lead over Al Gore in the popular vote turn into a deficit during the final three months of the campaign. It took the Electoral College and a disputed 537-vote victory in Florida to save his presidential bid. And notoriously, Hillary Clinton led Donald Trump by around eight points in June, August and even October of 2016, before she barely squeaked out a two-point edge in the popular vote.

 

https://www.economist.com/interactive/us-2024-election/prediction-model/president/how-this-works

 

Ultimately, some people who claim to be undecided are not, they favour one or the other party. Some who claim they will vote for one or the other, do not, or are not eligible to vote.

 

The key point is however that undecided voters in the swing states will determine the election, which is not about who gets the most votes, Clinton got the most votes and lost, because of the way the US electoral system works.

 

Undecided voters do change over time, and polls cannot accurately predict how they will vote. Reality, ultimately, is more complex than science and mathematics will be able to model. We saw this with corona, and we saw it with past polling efforts.

[maesariang]

2 minutes ago, Cameroni said:

 

First of all there is a  risk that a perfectly random sample of a given size may not reflect the characteristics of the population as a whole (known as “sampling error”). In fact, the group of people who participate in any given survey are virtually never an idealised random sub-set of the population that will actually turn out to vote. Instead, they can differ from the eventual mix in important ways, which collectively are known as “non-sampling error”.

 

Then polls are subject to the vagaries of voter turnout. Polls conducted among all adults will include the views of people who are ineligible or not registered to vote. Those limited to registered voters treat all respondents in this group as if they had an equal probability of showing up to vote, which they surely do not. And those that seek to filter out respondents unlikely to vote, or that grant more weight to the views of people who are more likely to show up, can get such calculations wrong. Although no two surveys are identical, ones that use a similar approach to predicting turnout are more likely to wind up with errors of a similar size and direction than are ones that handle it differently. In statistical terms, each of these different methods of turnout projection can produce a “bias”, which is likely to contaminate the results of all the pollsters that use it in a similar way.

 

The group of people pollsters can reach by using live telephone interviewers may have different voting intentions than those they can reach by automated phone calls, or via the internet. Individual pollsters may make methodological choices, such as weighting schemes, that consistently lead to more or less favourable results for a particular political party.

 

Ahead of time, it is impossible to know the direction or size of the bias that each of these characteristics may introduce.

 

Voters who have a soft but consistent preference for one of the two major parties often say they are undecided or planning to vote for a third party. This makes polls in the first half of the year a surprisingly weak predictor of final results. For example, in June 1988, George H.W. Bush trailed Michael Dukakis by 12 percentage points in polling averages (he went on to win by eight). Exactly four years later, Mr Bush led Bill Clinton by ten percentage points, and wound up losing by seven.

 

In more recent years, polling errors have been a bit smaller—but they can still be substantial. In 2000 Mr Bush’s son George W. saw his ten-point lead over Al Gore in the popular vote turn into a deficit during the final three months of the campaign. It took the Electoral College and a disputed 537-vote victory in Florida to save his presidential bid. And notoriously, Hillary Clinton led Donald Trump by around eight points in June, August and even October of 2016, before she barely squeaked out a two-point edge in the popular vote.

 

https://www.economist.com/interactive/us-2024-election/prediction-model/president/how-this-works

 

Ultimately, some people who claim to be undecided are not, they favour one or the other party. Some who claim they will vote for one or the other, do not, or are not eligible to vote.

 

The key point is however that undecided voters in the swing states will determine the election, which is not about who gets the most votes, Clinton got the most votes and lost, because of the way the US electoral system works.

 

Undecided voters do change over time, and polls cannot accurately predict how they will vote. Reality, ultimately, is more complex than science and mathematics will be able to model. We saw this with corona, and we saw it with past polling efforts.

Great polls. Voter turnout is important as is debates. Polls are a rough guide. For example one poll had margins going from 1 point to 10 points only a month apart.

[Cameroni]

2 minutes ago, maesariang said:

Great polls. Voter turnout is important as is debates. Polls are a rough guide. For example one poll had margins going from 1 point to 10 points only a month apart.

Yes, polls are "somewhat" of a guide. With a strong emphasis on "somewhat". Can they accurately predict how undecided voters in swing states will vote? No. Can they accurately predict how many undecided voters there will be in swing states? No.

 

So "somewhat" yes. But accurate? Not necessarily. 

 

Thus it's a bit foolish to get so excited about polls.

[Dan747]

Republican former President Donald Trump currently holds a 268-226 lead over Harris in the latest NPR Electoral Map of the most competitive states (see map above), just short of the 270 needed to win the presidency. Trump leads in the "Blue Wall" states of Pennsylvania, Wisconsin and Michigan within the margin of error.

[Irish star]

15 minutes ago, maesariang said:

They can't talk inflation or unemployment. So it is "cult" nonsense.

 

 

We can talk Stock market, lowest unemployment Ever , Abortion Rights , overthrowing Roe-Wade , we can discuss Republicans in Jail and percentages, we can discuss Putin Assets , Trump , MTG most of the republican roster , We can discuss Christofascist in the Republican Party , 2025  , Kim who he owes 20 million dollars , we can discuss Logan Act , but Biden Harris passed so many Bills ,  how much of his term did he play Golf , you can google Biden Accomplishments it’s easy versus Trump debacle presidency   

[Irish star]

2 minutes ago, Dan747 said:

Republican former President Donald Trump currently holds a 268-226 lead over Harris in the latest NPR Electoral Map of the most competitive states (see map above), just short of the 270 needed to win the presidency. Trump leads in the "Blue Wall" states of Pennsylvania, Wisconsin and Michigan within the margin of error.

This is Blatant Lie Geezus Christ 

[Irish star]

9 minutes ago, maesariang said:

Great polls. Voter turnout is important as is debates. Polls are a rough guide. For example one poll had margins going from 1 point to 10 points only a month apart.

Old News Wake up 

[Cameroni]

57 minutes ago, Fat is a type of crazy said:

Maybe - with months to go who knows - but his point seems to be that undecided's are somehow invisible in polls or something.

 

Yes, indeed undecided voters are invisible in polls to some degree because polls cannot accurately predict how many undecided there will be, if they will vote, or how they will vote.

 

Some people who claim to be undecided in polls are not. Some who claim they will vote do not. In the end reality is too complex for mathematical modelling to be accurate.

[maesariang]

5 minutes ago, Dan747 said:

Republican former President Donald Trump currently holds a 268-226 lead over Harris in the latest NPR Electoral Map of the most competitive states (see map above), just short of the 270 needed to win the presidency. Trump leads in the "Blue Wall" states of Pennsylvania, Wisconsin and Michigan within the margin of error.

Link?

[TedG]

20 minutes ago, Irish star said:

Clueless if you believe this 

Do you have any evidence to disprove it?

[maesariang]

Just now, Cameroni said:

 

Yes, indeed undecided voters are invisible in polls to some degree because polls cannot accurately predict how many undecided there will be, if they will vote, or how they will vote.

 

Some people who claim to be undecided in polls are not. Some who claim they will vote do not. In the end reality is too complex for mathematical modelling to be accurate.

Some decided voters may not turn up to vote.

[maesariang]

4 minutes ago, Irish star said:

We can talk Stock market, lowest unemployment Ever , Abortion Rights , overthrowing Roe-Wade , we can discuss Republicans in Jail and percentages, we can discuss Putin Assets , Trump , MTG most of the republican roster , We can discuss Christofascist in the Republican Party , 2025  , Kim who he owes 20 million dollars , we can discuss Logan Act , but Biden Harris passed so many Bills ,  how much of his term did he play Golf , you can google Biden Accomplishments it’s easy versus Trump debacle presidency   

Nice rant. Impressive. 

[maesariang]

3 minutes ago, Irish star said:

Old News Wake up 

Calm down.

[TedG]

16 minutes ago, Cameroni said:

 

First of all there is a  risk that a perfectly random sample of a given size may not reflect the characteristics of the population as a whole (known as “sampling error”). In fact, the group of people who participate in any given survey are virtually never an idealised random sub-set of the population that will actually turn out to vote. Instead, they can differ from the eventual mix in important ways, which collectively are known as “non-sampling error”.

 

 

 

One of my favorite college courses was sampling theory. Nice post.  

[TedG]

8 minutes ago, Irish star said:

We can talk Stock market, lowest unemployment Ever , Abortion Rights , overthrowing Roe-Wade , we can discuss Republicans in Jail and percentages, we can discuss Putin Assets , Trump , MTG most of the republican roster , We can discuss Christofascist in the Republican Party , 2025  , Kim who he owes 20 million dollars , we can discuss Logan Act , but Biden Harris passed so many Bills ,  how much of his term did he play Golf , you can google Biden Accomplishments it’s easy versus Trump debacle presidency   

Would you be able to talk about a topic in depth? 

[Irish star]

5 minutes ago, TedG said:

Would you be able to talk about a topic in depth? 

Just talking points and I receive a one liner , I’ve lost IQ points trying to reason with you , your lower than “Gutter Oil “ Google it it’s quite interesting what’s Going on in China especially if you travel 

Edited August 23 by Irish star

[Dan747]

[maesariang]

1 minute ago, Dan747 said:

In 3.5 years she would still be blaming Trump. 

[luckymitchell]

1 hour ago, Will B Good said:

 

 

...but just in case we can still use the last plan....you know the one with the voting machines, postal votes and stuffing ballot boxes during the night shift?......just to make doubly sure.

Yes, just like last time. using a over exaggerated flu as disguise to enable Ballot harvesting via mail in ballots, dead people Voting, over nigjt massive jumps on tallys, no ID needed either, because that's racist as those pesky black people don't have ID's.. oh no, silly me, that was the safetest securist election EVA 

[Irish star]

3 minutes ago, luckymitchell said:

Yes, just like last time. using a over exaggerated flu as disguise to enable Ballot harvesting via mail in ballots, dead people Voting, over nigjt massive jumps on tallys, no ID needed either, because that's racist as those pesky black people don't have ID's.. oh no, silly me, that was the safetest securist election EVA 

It must be Happy Hour ffs 

[luckymitchell]

24 minutes ago, Irish star said:

Just talking points and I receive a one liner , I’ve lost IQ points trying to reason with you , your lower than “Gutter Oil “ Google it it’s quite interesting what’s Going on in China especially if you travel 

You cannot lose what you don't have!

[Irish star]

1 hour ago, maesariang said:

They can't talk inflation or unemployment. So it is "cult" nonsense.

 

 

[luckymitchell]

1 minute ago, Irish star said:

It must be Happy Hour ffs 

Ive already used that joke, you need to get your own

[Fat is a type of crazy]

28 minutes ago, Cameroni said:

 

First of all there is a  risk that a perfectly random sample of a given size may not reflect the characteristics of the population as a whole (known as “sampling error”). In fact, the group of people who participate in any given survey are virtually never an idealised random sub-set of the population that will actually turn out to vote. Instead, they can differ from the eventual mix in important ways, which collectively are known as “non-sampling error”.

 

Then polls are subject to the vagaries of voter turnout. Polls conducted among all adults will include the views of people who are ineligible or not registered to vote. Those limited to registered voters treat all respondents in this group as if they had an equal probability of showing up to vote, which they surely do not. And those that seek to filter out respondents unlikely to vote, or that grant more weight to the views of people who are more likely to show up, can get such calculations wrong. Although no two surveys are identical, ones that use a similar approach to predicting turnout are more likely to wind up with errors of a similar size and direction than are ones that handle it differently. In statistical terms, each of these different methods of turnout projection can produce a “bias”, which is likely to contaminate the results of all the pollsters that use it in a similar way.

 

The group of people pollsters can reach by using live telephone interviewers may have different voting intentions than those they can reach by automated phone calls, or via the internet. Individual pollsters may make methodological choices, such as weighting schemes, that consistently lead to more or less favourable results for a particular political party.

 

Ahead of time, it is impossible to know the direction or size of the bias that each of these characteristics may introduce.

 

Voters who have a soft but consistent preference for one of the two major parties often say they are undecided or planning to vote for a third party. This makes polls in the first half of the year a surprisingly weak predictor of final results. For example, in June 1988, George H.W. Bush trailed Michael Dukakis by 12 percentage points in polling averages (he went on to win by eight). Exactly four years later, Mr Bush led Bill Clinton by ten percentage points, and wound up losing by seven.

 

In more recent years, polling errors have been a bit smaller—but they can still be substantial. In 2000 Mr Bush’s son George W. saw his ten-point lead over Al Gore in the popular vote turn into a deficit during the final three months of the campaign. It took the Electoral College and a disputed 537-vote victory in Florida to save his presidential bid. And notoriously, Hillary Clinton led Donald Trump by around eight points in June, August and even October of 2016, before she barely squeaked out a two-point edge in the popular vote.

 

https://www.economist.com/interactive/us-2024-election/prediction-model/president/how-this-works

 

Ultimately, some people who claim to be undecided are not, they favour one or the other party. Some who claim they will vote for one or the other, do not, or are not eligible to vote.

 

The key point is however that undecided voters in the swing states will determine the election, which is not about who gets the most votes, Clinton got the most votes and lost, because of the way the US electoral system works.

 

Undecided voters do change over time, and polls cannot accurately predict how they will vote. Reality, ultimately, is more complex than science and mathematics will be able to model. We saw this with corona, and we saw it with past polling efforts.

Thanks for the detailed explanation and it is understood there are many variables as you have outlined and it is difficult to predict which way they will go as the election gets closer. I guess my point was that you seemed to be specific about the undecideds as being invisible in polling and that that was a serious impediment to the worth of polls. It seems though that if it is a reasonable sample and a well run poll - and say we see Harris 4 points ahead this week compared to 4 points behind last week or vice versa - it seems it should still be be a  fairly good indication of a trend  and a valid guide to what's happening out there i.e. undecideds would not in particular be randomly unpredictable IN THEIR POLLING over time  - even if their actual votes may be a little less easy to predict on the day. So polling trends - based on all participants in the poll - would be valuable and meaningful even if final votes by undecideds  are not totally predictable for the reasons you have noted.  It may be we are on the same page in a sense. 

[Cameroni]

13 minutes ago, Fat is a type of crazy said:

Thanks for the detailed explanation and it is understood there are many variables as you have outlined and it is difficult to predict which way they will go as the election gets closer. I guess my point was that you seemed to be specific about the undecideds as being invisible in polling and that that was a serious impediment to the worth of polls. It seems though that if it is a reasonable sample and a well run poll - and say we see Harris 4 points ahead this week compared to 4 points behind last week or vice versa - it seems it should still be be a  fairly good indication of a trend  and a valid guide to what's happening out there i.e. undecideds would not in particular be randomly unpredictable IN THEIR POLLING over time  - even if their actual votes may be a little less easy to predict on the day. So polling trends - based on all participants in the poll - would be valuable and meaningful even if final votes by undecideds  are not totally predictable for the reasons you have noted.  It may be we are on the same page in a sense. 

Yes, looks like we are, and the reason I highlighted the undecided voters in swing states was merely because pollsters have highlighted that they believe the swing states and in particular the undecided voters in those states will decide the election. As this is a close election, at this point in time that seemed a reasonable argument to me. Given the importance of the undecided voters in swing states, the difficulty of predicting how they would vote struck me and hence how hard pollsters would have it to predict their numbers or votes accurately, given how some undecided are actually voters who have decided, people who will not turn up to vote and so on. Having looked at Democrat pollsters in particular they certainly do not claim to know the outcome of the election, and I just found it funny how some people here thought polls are Nostradamus like certain predictions. I don't think they are.

[Irish star]

20 minutes ago, luckymitchell said:

You cannot lose what you don't have!

Mitch  what’s  so Lucky about ya , Brain Worms , Mentally Challenged behavior, , KDS , Cringe Fringe, Cmon tell us Why Mitch is so Lucky, I need to hear this , since your arguments are turning Juvenile at best 

[maesariang]

Polling trends are like a dog crossing a road. On this day Biden was 7.8% ahead. Before election 7.2% ahead. Final result 4.5%. So the trend dipped 0.6% and was still 2.7% too high. Harris is only 1.5% ahead. If the error rate is the same Trump wins an election held tomorrow by 1.2%.

Edited August 23 by maesariang

[metisdead]

A post with a derogatory trolling comment has been removed.

[Cameroni]

5 minutes ago, maesariang said:

Polling trends are like a dog crossing a road. On this day Biden was 7.8% ahead. Before election 7.2% ahead. Final result 4.5%. So the trend dipped 0.6% and was still 2.7% too high. 

 

Indeed, to all intents and purposes polls are a snapshot in time always. They will try to amend their data to reflect new  trends but this may or may not succeed in reality.

 

I'm a forex trader and we use a variety of tools to predict how the market will act. However, more often than not the market surprises almost everyone, despite the best info and tools we use to predict what it will do. Reality has too many variables.

 

It's true, past events form future trends...often, but not always. There can be major suprises.

Edited August 23 by Cameroni

[maesariang]

4 minutes ago, Cameroni said:

 

Indeed, to all intents and purposes polls are a snapshot in time always. They will try to amend their data to reflect new  trends but this may or may not succeed in reality.

 

I'm a forex trader and we use a variety of tools to predict how the market will act. However, more often than not the market surprises almost everyone, despite the best info and tools we use to predict what it will do. Reality has too many variables.

 

It's true, past events form future trends...often, but not always. There can be major suprises.

Some large traders go short to stop out the weak holders then buy back in.