Indiana SurveyUSA: Trump +7

[new_patomic]

I don't believe the poll was weighted by region, so I'm not sure why it's relevant.

"Adult respondents were weighted to the most recent US Census targets for Indiana for gender, age, race, education, and home-ownership."

Their 'regions' are just another sub-sample for people to make comparisons with, nothing more.

[Ferguson97]

This is a 12 point shift from 2016 (56-37 Trump).

Clinton won the NPV by 3 points, so this would be consistent with a Biden+15 national lead.

[StateBoiler]

I don't believe the poll was weighted by region, so I'm not sure why it's relevant.

"Adult respondents were weighted to the most recent US Census targets for Indiana for gender, age, race, education, and home-ownership."

Their regions are just another sub-sample for people to make comparisons with.

You can't make a comparison combining the districts of Pete Visclosky, Jackie Walorski, and Jim Banks together.

Also, 527 people. So for all of a region, we're talking approximately 132 people for that whole area.

[ProgressiveModerate]

Likely R. It would be epic if IN flipped back after no one said it was possible, though I think that's extremely unlikely at this point.

[kireev]

2016 vote is Trump 48%, Clinton 31%.  This is about right and it's still a major shirt.

[Darthpi – Anti-Florida Activist]

B-b-but muh “state polls aren’t consistent with a double digit national Biden lead” 😭

Biden's strength with olds seems to be doing a bit of depolarisation. There seems to be a curious phenomenon where he's getting extremely strong polling in red states like MT, MO, IN, even WV, while the battleground states are moving towards him slightly less decisively from the pre-debate leads.

I compared the polling averages in various states with a hypothetical Biden +9 map based on a uniform national swing, and I think what we're seeing in Indiana and Missouri is a combination of what you're suggesting here along with Biden also overperforming generally across the Midwest as well, including in the battleground states.

[DaleCooper]

If Biden is at 42% in Indiana then Trump isn't competitive in the Rust Belt.

[StateBoiler]

2016 vote is Trump 48%, Clinton 31%.  This is about right and it's still a major shirt.

That was not the 2016 vote. That only adds up to 79 and Gary Johnson did not have 21% support then. Is it seriously that difficult for people to look up information on here?*

Trump 1,557,286 57.16%
Clinton 1,033,126 37.92%
Johnson 133,993 4.92%

*The majority of the people of this board have been determined to be less than 28 years old.

[Co-Chair Bagel23]

SUSA was off in Indiana by 6 points in 2018.  Regardless, a bad poll here for Mr. Trump!

Even if Trump wins Indiana by 13 points, he is utterly f#cked. If Indiana swings from +19 Trump to +13 Trump, he is losing Pennsylvania, Michigan, and Wisconsin, and thus is f#cked.

[StateBoiler]

SUSA was off in Indiana by 6 points in 2018.  Regardless, a bad poll here for Mr. Trump!

Even if Trump wins Indiana by 13 points, he is utterly f#cked. If Indiana swings from +19 Trump to +13 Trump, he is losing Pennsylvania, Michigan, and Wisconsin, and thus is f#cked.

Why are people on here acting like this matters? Biden is winning! We all know this! I've known it since March.

[Holmes]

2016 vote is Trump 48%, Clinton 31%.  This is about right and it's still a major shirt.

That was not the 2016 vote. That only adds up to 79 and Gary Johnson did not have 21% support then. Is it seriously that difficult for people to look up information on here?*

Trump 1,557,286 57.16%
Clinton 1,033,126 37.92%
Johnson 133,993 4.92%

*The majority of the people of this board have been determined to be less than 28 years old.

I think it’s 48-31 among poll responders. Some people who were polled didn’t vote in 2016 or just didn’t say.

[kireev]

2016 vote is Trump 48%, Clinton 31%.  This is about right and it's still a major shirt.

That was not the 2016 vote. That only adds up to 79 and Gary Johnson did not have 21% support then. Is it seriously that difficult for people to look up information on here?*

Trump 1,557,286 57.16%
Clinton 1,033,126 37.92%
Johnson 133,993 4.92%

*The majority of the people of this board have been determined to be less than 28 years old.

21% is "other" plus people who did not vote in 2016, which does seem about right.

[Buzz]

SUSA was off in Indiana by 6 points in 2018.  Regardless, a bad poll here for Mr. Trump!

Even if Trump wins Indiana by 13 points, he is utterly f#cked. If Indiana swings from +19 Trump to +13 Trump, he is losing Pennsylvania, Michigan, and Wisconsin, and thus is f#cked.

We already knew he was losing those states though... it’s just comes down to how close will the election be?

[StateBoiler]

2016 vote is Trump 48%, Clinton 31%.  This is about right and it's still a major shirt.

That was not the 2016 vote. That only adds up to 79 and Gary Johnson did not have 21% support then. Is it seriously that difficult for people to look up information on here?*

Trump 1,557,286 57.16%
Clinton 1,033,126 37.92%
Johnson 133,993 4.92%

*The majority of the people of this board have been determined to be less than 28 years old.

21% is "other" plus people who did not vote in 2016, which does seem about right.

Other in 2016 was 5% counting write-ins. So you're telling me 16% or thereabouts of people in this poll did not vote in 2016 when the number of people voting in the state in 2016 was higher than 2008? I'll grant you people ages 18 to 22 and some people that did not vote. You're not getting up to 16% with that crowd.

I just want the critical application of mathematical skill. A person can give you a more or less correct general result while still being sh**t at their job of the methodology getting to the end point. This pollster in one tweet said if Biden improved 4 points in performance, it meant the Democrat in the Attorney General race could win, the same Democrat this same poll said is losing by 12 points. And this was all done off the back of CROWDFUNDING!, so don't sit here and tell me there's an automatically assumed aspect of quality to this.

[Asta]

Trump will probably win states like IN, MO, WV by few points more than the polls say. That's not to say he's not in trouble but there is tendency for blue voters in red states and red voters in blue states to be less enthusiastic to vote, knowing their candidate is going to lose in the state.

That's why swing states tend to have higher voter participation rate.

[Skye]

StateBoiler ease up pal. It's just a poll. No need to get so worked up.

[kireev]

2016 vote is Trump 48%, Clinton 31%.  This is about right and it's still a major shirt.

That was not the 2016 vote. That only adds up to 79 and Gary Johnson did not have 21% support then. Is it seriously that difficult for people to look up information on here?*

Trump 1,557,286 57.16%
Clinton 1,033,126 37.92%
Johnson 133,993 4.92%

*The majority of the people of this board have been determined to be less than 28 years old.

21% is "other" plus people who did not vote in 2016, which does seem about right.

Other in 2016 was 5% counting write-ins. So you're telling me 16% or thereabouts of people in this poll did not vote in 2016 when the number of people voting in the state in 2016 was higher than 2008?

I just want the critical application of mathematical skill. A person can give you a more or less correct general result while still being sh**t at their job of the methodology getting to the end point. This pollster in one tweet said if Biden improved 4 points in performance, it meant the Democrat in the Attorney General race could win, the same Democrat this same poll said is losing by 12 points. And this was all done off the back of "CROWDFUNDING!"

There is absolutely nothing wrong with the 16% new voters number. 18-22-year olds start voting, and the pool of 2016 is smaller because old people are dying. In addition, a small percent of the 2016 voters will not vote in 2020 election. So we probably need a turnout 10% higher than in 2016 to get to the 16% new voters. Is it really hard to imagine a 10% higher turnout in 2020 than in 2016? I really don't think so.

[ON Progressive]

I’d like to know how exactly having regional definitions you don’t like makes it a bad poll, because that seems incredibly stupid to me.

[Penn_Quaker_Girl]

I’d like to know how exactly having regional definitions you don’t like makes it a bad poll, because that seems incredibly stupid to me.

I'm still trying to figure out why being under the age of 28 is a bad thing. 

[demoman1596]

2016 vote is Trump 48%, Clinton 31%.  This is about right and it's still a major shirt.

That was not the 2016 vote. That only adds up to 79 and Gary Johnson did not have 21% support then. Is it seriously that difficult for people to look up information on here?*

Trump 1,557,286 57.16%
Clinton 1,033,126 37.92%
Johnson 133,993 4.92%

*The majority of the people of this board have been determined to be less than 28 years old.

21% is "other" plus people who did not vote in 2016, which does seem about right.

Other in 2016 was 5% counting write-ins. So you're telling me 16% or thereabouts of people in this poll did not vote in 2016 when the number of people voting in the state in 2016 was higher than 2008? I'll grant you people ages 18 to 22 and some people that did not vote. You're not getting up to 16% with that crowd.

I just want the critical application of mathematical skill. A person can give you a more or less correct general result while still being sh**t at their job of the methodology getting to the end point. This pollster in one tweet said if Biden improved 4 points in performance, it meant the Democrat in the Attorney General race could win, the same Democrat this same poll said is losing by 12 points. And this was all done off the back of CROWDFUNDING!, so don't sit here and tell me there's an automatically assumed aspect of quality to this.

For a person harping on "critical application of... skill," you don't seem to be making very good mathematical or logical arguments here.  I mean, I get pissed off about stuff like that, too, and you're totally allowed to feel however you want to feel, but your posts might be coming off differently than you expect.

[TiltsAreUnderrated]

B-b-but muh “state polls aren’t consistent with a double digit national Biden lead” 😭

Biden's strength with olds seems to be doing a bit of depolarisation. There seems to be a curious phenomenon where he's getting extremely strong polling in red states like MT, MO, IN, even WV, while the battleground states are moving towards him slightly less decisively from the pre-debate leads.

Louder for those in the back: polling in deep red/blue states tends to underestimate the dominant party.

MoE: 4.5%

[Sbane]

I expect something like 55-43 Trump in the end. If Biden can get that margin down beneath 10, he's got the election locked up.

Yup, I agree. Even a 12 point win in Indiana for Trump would imply a close race in Ohio and not so particularly close in Michigan, Pennsylvania and Wisconsin.

[Hammy]

SUSA was off in Indiana by 6 points in 2018.  Regardless, a bad poll here for Mr. Trump!

FWIW the topline numbers were pretty garbage at 41-40. That's just begging for a massive polling error.

[South Dakota Democrat]

This is a 12 point shift from 2016 (56-37 Trump).

Clinton won the NPV by 3 points, so this would be consistent with a Biden+15 national lead.

2 points

[StateBoiler]

StateBoiler ease up pal. It's just a poll. No need to get so worked up.

It's a bullsh**t poll, and people make decisions and assumptions on stuff that's shoddily put together and wrong. It's a guy on Twitter that crowdfunded a poll, which is Red Flag #1, and then when he publishes the poll immediately points out he's partisan and hopeful about other races swinging Democrats' way, Red Flag #2. You have zero legitimacy at that point as far as being a neutral fair arbiter conducting the poll, even worse than saying Rainwater is at 24%. Throw on top of it his regions he assign does not show an in-depth knowledge of the political regions of Indiana at all, he might as well have lumped Indianapolis in with Switzerland County. That ranks at a level lower in respectability than campaign internal polling.

Let's look at this polling in detail, say black voters:

Biden 75, Trump 21
Myers 57, Holcomb 37, Rainwater 3
Weinzapfel 59, Rokita 25

This should be huge effing news if anyone thinks black voters have swung Republican to this degree to where in the race for Attorney General where most random voters can't even name who's running and there's no Trump bias involved, they're going to get 1 out of 4. (And Holcomb even is at 3 of 8.) Did Curtis Hill, a black Republican that won Attorney General then, get 1 out of 4 black voters in 2016? Do people on this thread defending this poll believe black voters have become more Republican the past 4 years? Sure doesn't match the national media narrative does it?

Hey, I'm in favor of polling, but it's like journalism. Anything shoddy should be thrown into a trashcan. This is a trashcan poll.