Are dems punching above weight in Iowa?

[freepcrusher]

Here in the Hawkeye state it's been a rough outing for dems ever since Obama became president in 2008. But is it possible that the Dems here are actually doing relatively well since there's no chicago or twin cities?

Looking at precincts over 90 percent white, trump won 64 percent of the vote in Iowa. But here are his numbers in other states I've looked at on DRA

AL 86 percent
GA 81 percent
AR 79 percent
TN 79 percent
VA 75 percent
WV 74 percent
KY 74 percent
MO 74 percent
IN 72 percent
NC 70 percent
OH 69 percent
SC 68 percent
IL 68 percent
MD 68 percent

The only difference is that in Iowa, said precincts are 40 percent of the population. If it was 20 or 30 percent like it is in Indiana or Ohio, the state would probably elect democrats pretty regularly

[WalterWhite]

This is the classic North vs. South divide. White Iowans are less conservative than white Southerners, so they vote less Republican than white Southerners.

The 1980s Farm Crisis also contributed to this. The mostly white Driftless Region shifted dramatically to the left, causing a drastic leftward shift statewide. Even nowadays, the Driftless Region votes to the left of neighboring rural areas.

[lfromnj]

Wait SC and IL are the same? I mean  it makes sense because most suburban precincts should have enough Asians and other races to get whites below 90 but wow.

[I spent the winter writing songs about getting better]

Admittedly it's a statistical tie with the margin of error and all that and that the DRA numbers are actually kind of funky to begin with but notable that Trump possibly did better amongst white voters in Virginia than in West Virginia.

EDIT: But per the exit poll Trump won 54% of white voters in Virginia. I just think the DRA numbers are real off.

[Sol]

Admittedly it's a statistical tie with the margin of error and all that and that the DRA numbers are actually kind of funky to begin with but notable that Trump possibly did better amongst white voters in Virginia than in West Virginia.

EDIT: But per the exit poll Trump won 54% of white voters in Virginia. I just think the DRA numbers are real off.

DRA numbers for the state as a whole are probably more accurate than exit polls, since data smudging and the absentee ballot vagueness of VA disappear in the aggregate.

[I spent the winter writing songs about getting better]

Admittedly it's a statistical tie with the margin of error and all that and that the DRA numbers are actually kind of funky to begin with but notable that Trump possibly did better amongst white voters in Virginia than in West Virginia.

EDIT: But per the exit poll Trump won 54% of white voters in Virginia. I just think the DRA numbers are real off.

DRA numbers for the state as a whole are probably more accurate than exit polls, since data smudging and the absentee ballot vagueness of VA disappear in the aggregate.

I'm not seeing how it's mathematically possible Biden did only 4 points better amongst whites in Virginia than he did in Tennessee.

[ProgressiveModerate]

Part of it is def just IA being a northern and relatively secular state, and containing a large part of the driftless region.

And while yes, IA lacks any giant metro like Chicago or MSP, it's rurals are also slightly less dense than many of it's midwestern neighbors. Des Moines or Cedar Rapids def aren't enough to carry the state for Dems, but in Iowa context they're substantial enough to make a seriously impact

Another reason I suspect IA might be closer federally than Demographics might suggest is because it's the first primary state. I know that may sound stupid, but seriously, IA generally has like really solid turnout especially given it's not super high educational attainment or anything. I think this in large part can be attributed to all the activation of people around the Iowa cacuses.



On 2020 Pres numbers, IA doesn't really stand out as being notably less R than it's neighbors; only exception is the MO-IA border but that's equally a question about why Missouri rurals are so red. Also, one could actually argue that northwestern IA stands out as being abnormally R.

[SevenEleven]

Admittedly it's a statistical tie with the margin of error and all that and that the DRA numbers are actually kind of funky to begin with but notable that Trump possibly did better amongst white voters in Virginia than in West Virginia.

EDIT: But per the exit poll Trump won 54% of white voters in Virginia. I just think the DRA numbers are real off.

DRA numbers for the state as a whole are probably more accurate than exit polls, since data smudging and the absentee ballot vagueness of VA disappear in the aggregate.

I'm not seeing how it's mathematically possible Biden did only 4 points better amongst whites in Virginia than he did in Tennessee.

The precincts where Biden wins a higher percentage of whites aren't 90% white.

Virginia being 60% white vs Iowa's 86% is significant.

[Tintrlvr]

Admittedly it's a statistical tie with the margin of error and all that and that the DRA numbers are actually kind of funky to begin with but notable that Trump possibly did better amongst white voters in Virginia than in West Virginia.

EDIT: But per the exit poll Trump won 54% of white voters in Virginia. I just think the DRA numbers are real off.

DRA numbers for the state as a whole are probably more accurate than exit polls, since data smudging and the absentee ballot vagueness of VA disappear in the aggregate.

I'm not seeing how it's mathematically possible Biden did only 4 points better amongst whites in Virginia than he did in Tennessee.

The precincts where Biden wins a higher percentage of whites aren't 90% white.

Virginia being 60% white vs Iowa's 86% is significant.

Right. In Iowa there are lots of urban precincts that are 90% white. In Virginia there are no 90% white precincts that are urban, even in relatively segregated "old Southern" cities like Norfolk, Richmond or Roanoke. So looking at 90% white precincts in Virginia means sampling exclusively white people in rural areas in contrast to a much more (though not entirely) representative sample of white people in Iowa. Same goes for Tennessee as well, to a lesser degree.

[ProgressiveModerate]

Admittedly it's a statistical tie with the margin of error and all that and that the DRA numbers are actually kind of funky to begin with but notable that Trump possibly did better amongst white voters in Virginia than in West Virginia.

EDIT: But per the exit poll Trump won 54% of white voters in Virginia. I just think the DRA numbers are real off.

DRA numbers for the state as a whole are probably more accurate than exit polls, since data smudging and the absentee ballot vagueness of VA disappear in the aggregate.

I'm not seeing how it's mathematically possible Biden did only 4 points better amongst whites in Virginia than he did in Tennessee.

The precincts where Biden wins a higher percentage of whites aren't 90% white.

Virginia being 60% white vs Iowa's 86% is significant.

Right. In Iowa there are lots of urban precincts that are 90% white. In Virginia there are no 90% white precincts that are urban, even in relatively segregated "old Southern" cities like Norfolk, Richmond or Roanoke. So looking at 90% white precincts in Virginia means sampling exclusively white people in rural areas in contrast to a much more (though not entirely) representative sample of white people in Iowa. Same goes for Tennessee as well, to a lesser degree.

The closest I can get in VA is a few very white precincts to the west of downtown Richmond and a few Virginia Beach precincts. Def suggests suburban VA whites still lean R, but not by much, and a few precincts close to downtown Richmond def have whites that vote D.

[Sol]

Why are y'all measuring this by white vote in 90% white precincts?

[Aurelius2]

FWIW I'm pretty sure DRA significantly overestimates nonwhite voters in white-majority areas for some reason. Their Native American numbers, in particular, are drastically higher than the Census numbers in many states.

[Aurelius2]

Why are y'all measuring this by white vote in 90% white precincts?

Yeah, if you took this approach to the Jewish vote you'd assume they vote 90%+ R.

[freepcrusher]

Admittedly it's a statistical tie with the margin of error and all that and that the DRA numbers are actually kind of funky to begin with but notable that Trump possibly did better amongst white voters in Virginia than in West Virginia.

EDIT: But per the exit poll Trump won 54% of white voters in Virginia. I just think the DRA numbers are real off.

https://davesredistricting.org/maps#viewmap::46f48b1e-ac1e-4f92-87e9-369ba3378ffd

[I spent the winter writing songs about getting better]

Admittedly it's a statistical tie with the margin of error and all that and that the DRA numbers are actually kind of funky to begin with but notable that Trump possibly did better amongst white voters in Virginia than in West Virginia.

EDIT: But per the exit poll Trump won 54% of white voters in Virginia. I just think the DRA numbers are real off.

DRA numbers for the state as a whole are probably more accurate than exit polls, since data smudging and the absentee ballot vagueness of VA disappear in the aggregate.

Just did some quick math and indeed the VRA numbers are impossible.

Virginia was about 58% white alone and 62% white self reported last census. If we take the lower number and don't regard higher whire turnout 0.58×0.75=over 43%, and Trump received 44% in VA. That means he would have to received almost no votes whatsoever from the non-white voters. 1/42=2.3% so accounting for third parties Biden would still have to have won over 95% of the non-white vote!

[Unelectable Bystander]

Admittedly it's a statistical tie with the margin of error and all that and that the DRA numbers are actually kind of funky to begin with but notable that Trump possibly did better amongst white voters in Virginia than in West Virginia.

EDIT: But per the exit poll Trump won 54% of white voters in Virginia. I just think the DRA numbers are real off.

DRA numbers for the state as a whole are probably more accurate than exit polls, since data smudging and the absentee ballot vagueness of VA disappear in the aggregate.

Just did some quick math and indeed the VRA numbers are impossible.

Virginia was about 58% white alone and 62% white self reported last census. If we take the lower number and don't regard higher whire turnout 0.58×0.75=over 43%, and Trump received 44% in VA. That means he would have to received almost no votes whatsoever from the non-white voters. 1/42=2.3% so accounting for third parties Biden would still have to have won over 95% of the non-white vote!

That’s not what this statistic is saying. It is excluding whites in precincts where they make up less than 90%. In Virginia that is equivalent to saying “Trump received 75 of the vote in the whitest rural areas in the western part of the state and maybe a few NOVA exurbs.”

[Sol]

Admittedly it's a statistical tie with the margin of error and all that and that the DRA numbers are actually kind of funky to begin with but notable that Trump possibly did better amongst white voters in Virginia than in West Virginia.

EDIT: But per the exit poll Trump won 54% of white voters in Virginia. I just think the DRA numbers are real off.

DRA numbers for the state as a whole are probably more accurate than exit polls, since data smudging and the absentee ballot vagueness of VA disappear in the aggregate.

Just did some quick math and indeed the VRA numbers are impossible.

Virginia was about 58% white alone and 62% white self reported last census. If we take the lower number and don't regard higher whire turnout 0.58×0.75=over 43%, and Trump received 44% in VA. That means he would have to received almost no votes whatsoever from the non-white voters. 1/42=2.3% so accounting for third parties Biden would still have to have won over 95% of the non-white vote!

Yeah I would not take the stats of the OP as an accurate measure of the statewide vote. Doing some back of the envelope calculations, the statewide vote is around ~37% for Biden, which is basically in line with Adam Griffin's estimates in 2012. I also assumed around 95% of Black voters for Biden, so if your model of Black voters in VA has them voting a little to the right of that the white vote for Biden goes up a bit.

[I spent the winter writing songs about getting better]

Admittedly it's a statistical tie with the margin of error and all that and that the DRA numbers are actually kind of funky to begin with but notable that Trump possibly did better amongst white voters in Virginia than in West Virginia.

EDIT: But per the exit poll Trump won 54% of white voters in Virginia. I just think the DRA numbers are real off.

DRA numbers for the state as a whole are probably more accurate than exit polls, since data smudging and the absentee ballot vagueness of VA disappear in the aggregate.

Just did some quick math and indeed the VRA numbers are impossible.

Virginia was about 58% white alone and 62% white self reported last census. If we take the lower number and don't regard higher whire turnout 0.58×0.75=over 43%, and Trump received 44% in VA. That means he would have to received almost no votes whatsoever from the non-white voters. 1/42=2.3% so accounting for third parties Biden would still have to have won over 95% of the non-white vote!

Yeah I would not take the stats of the OP as an accurate measure of the statewide vote. Doing some back of the envelope calculations, the statewide vote is around ~37% for Biden, which is basically in line with Adam Griffin's estimates in 2012. I also assumed around 95% of Black voters for Biden, so if your model of Black voters in VA has them voting a little to the right of that the white vote for Biden goes up a bit.

Virginia is only a little over 20% black though. The rest is mostly Hispanic and Asian, who actually together almost make up almost as much as the black percentage, and obviously aren't going to be >90% for Biden.

[Sol]

Admittedly it's a statistical tie with the margin of error and all that and that the DRA numbers are actually kind of funky to begin with but notable that Trump possibly did better amongst white voters in Virginia than in West Virginia.

EDIT: But per the exit poll Trump won 54% of white voters in Virginia. I just think the DRA numbers are real off.

DRA numbers for the state as a whole are probably more accurate than exit polls, since data smudging and the absentee ballot vagueness of VA disappear in the aggregate.

Just did some quick math and indeed the VRA numbers are impossible.

Virginia was about 58% white alone and 62% white self reported last census. If we take the lower number and don't regard higher whire turnout 0.58×0.75=over 43%, and Trump received 44% in VA. That means he would have to received almost no votes whatsoever from the non-white voters. 1/42=2.3% so accounting for third parties Biden would still have to have won over 95% of the non-white vote!

Yeah I would not take the stats of the OP as an accurate measure of the statewide vote. Doing some back of the envelope calculations, the statewide vote is around ~37% for Biden, which is basically in line with Adam Griffin's estimates in 2012. I also assumed around 95% of Black voters for Biden, so if your model of Black voters in VA has them voting a little to the right of that the white vote for Biden goes up a bit.

Virginia is only a little over 20% black though. The rest is mostly Hispanic and Asian, who actually together almost make up almost as much as the black percentage, and obviously aren't going to be >90% for Biden.

I used DRA's CVAP 2020 data -- a lot of those Latinos and Asians aren't citizens.

Also it should be noted that I assumed roughly even turnout which is probably also wrong; with that in mind white vote percentage for Biden is probably around 40% or a few % points about which seems to make sense to me.

[freepcrusher]

Admittedly it's a statistical tie with the margin of error and all that and that the DRA numbers are actually kind of funky to begin with but notable that Trump possibly did better amongst white voters in Virginia than in West Virginia.

EDIT: But per the exit poll Trump won 54% of white voters in Virginia. I just think the DRA numbers are real off.

DRA numbers for the state as a whole are probably more accurate than exit polls, since data smudging and the absentee ballot vagueness of VA disappear in the aggregate.

Just did some quick math and indeed the VRA numbers are impossible.

Virginia was about 58% white alone and 62% white self reported last census. If we take the lower number and don't regard higher whire turnout 0.58×0.75=over 43%, and Trump received 44% in VA. That means he would have to received almost no votes whatsoever from the non-white voters. 1/42=2.3% so accounting for third parties Biden would still have to have won over 95% of the non-white vote!

Yeah I would not take the stats of the OP as an accurate measure of the statewide vote. Doing some back of the envelope calculations, the statewide vote is around ~37% for Biden, which is basically in line with Adam Griffin's estimates in 2012. I also assumed around 95% of Black voters for Biden, so if your model of Black voters in VA has them voting a little to the right of that the white vote for Biden goes up a bit.

Virginia is only a little over 20% black though. The rest is mostly Hispanic and Asian, who actually together almost make up almost as much as the black percentage, and obviously aren't going to be >90% for Biden.

Do you not listen very well? Did you not hear where I'm saying this applies only to 90 percent or more white precincts?

[I spent the winter writing songs about getting better]

Admittedly it's a statistical tie with the margin of error and all that and that the DRA numbers are actually kind of funky to begin with but notable that Trump possibly did better amongst white voters in Virginia than in West Virginia.

EDIT: But per the exit poll Trump won 54% of white voters in Virginia. I just think the DRA numbers are real off.

DRA numbers for the state as a whole are probably more accurate than exit polls, since data smudging and the absentee ballot vagueness of VA disappear in the aggregate.

Just did some quick math and indeed the VRA numbers are impossible.

Virginia was about 58% white alone and 62% white self reported last census. If we take the lower number and don't regard higher whire turnout 0.58×0.75=over 43%, and Trump received 44% in VA. That means he would have to received almost no votes whatsoever from the non-white voters. 1/42=2.3% so accounting for third parties Biden would still have to have won over 95% of the non-white vote!

Yeah I would not take the stats of the OP as an accurate measure of the statewide vote. Doing some back of the envelope calculations, the statewide vote is around ~37% for Biden, which is basically in line with Adam Griffin's estimates in 2012. I also assumed around 95% of Black voters for Biden, so if your model of Black voters in VA has them voting a little to the right of that the white vote for Biden goes up a bit.

Virginia is only a little over 20% black though. The rest is mostly Hispanic and Asian, who actually together almost make up almost as much as the black percentage, and obviously aren't going to be >90% for Biden.

Do you not listen very well? Did you not hear where I'm saying this applies only to 90 percent or more white precincts?

That's kind of a pointless number to use.

[freepcrusher]

Admittedly it's a statistical tie with the margin of error and all that and that the DRA numbers are actually kind of funky to begin with but notable that Trump possibly did better amongst white voters in Virginia than in West Virginia.

EDIT: But per the exit poll Trump won 54% of white voters in Virginia. I just think the DRA numbers are real off.

DRA numbers for the state as a whole are probably more accurate than exit polls, since data smudging and the absentee ballot vagueness of VA disappear in the aggregate.

Just did some quick math and indeed the VRA numbers are impossible.

Virginia was about 58% white alone and 62% white self reported last census. If we take the lower number and don't regard higher whire turnout 0.58×0.75=over 43%, and Trump received 44% in VA. That means he would have to received almost no votes whatsoever from the non-white voters. 1/42=2.3% so accounting for third parties Biden would still have to have won over 95% of the non-white vote!

Yeah I would not take the stats of the OP as an accurate measure of the statewide vote. Doing some back of the envelope calculations, the statewide vote is around ~37% for Biden, which is basically in line with Adam Griffin's estimates in 2012. I also assumed around 95% of Black voters for Biden, so if your model of Black voters in VA has them voting a little to the right of that the white vote for Biden goes up a bit.

Virginia is only a little over 20% black though. The rest is mostly Hispanic and Asian, who actually together almost make up almost as much as the black percentage, and obviously aren't going to be >90% for Biden.

Do you not listen very well? Did you not hear where I'm saying this applies only to 90 percent or more white precincts?

That's kind of a pointless number to use.

I'd argue it's illustrative. Like why are whites, when they have the place to themselves, so much more Republican?

[ProgressiveModerate]

Admittedly it's a statistical tie with the margin of error and all that and that the DRA numbers are actually kind of funky to begin with but notable that Trump possibly did better amongst white voters in Virginia than in West Virginia.

EDIT: But per the exit poll Trump won 54% of white voters in Virginia. I just think the DRA numbers are real off.

DRA numbers for the state as a whole are probably more accurate than exit polls, since data smudging and the absentee ballot vagueness of VA disappear in the aggregate.

Just did some quick math and indeed the VRA numbers are impossible.

Virginia was about 58% white alone and 62% white self reported last census. If we take the lower number and don't regard higher whire turnout 0.58×0.75=over 43%, and Trump received 44% in VA. That means he would have to received almost no votes whatsoever from the non-white voters. 1/42=2.3% so accounting for third parties Biden would still have to have won over 95% of the non-white vote!

Yeah I would not take the stats of the OP as an accurate measure of the statewide vote. Doing some back of the envelope calculations, the statewide vote is around ~37% for Biden, which is basically in line with Adam Griffin's estimates in 2012. I also assumed around 95% of Black voters for Biden, so if your model of Black voters in VA has them voting a little to the right of that the white vote for Biden goes up a bit.

Virginia is only a little over 20% black though. The rest is mostly Hispanic and Asian, who actually together almost make up almost as much as the black percentage, and obviously aren't going to be >90% for Biden.

Do you not listen very well? Did you not hear where I'm saying this applies only to 90 percent or more white precincts?

That's kind of a pointless number to use.

I'd argue it's illustrative. Like why are whites, when they have the place to themselves, so much more Republican?

Ye it's not unless, but def don't extrapolate too much.

Generally speaking, I think folks who self-sort into all-white communities are naturally going to be more conservative than those who move to nearby racially diverse neighborhoods.

[freepcrusher]

Admittedly it's a statistical tie with the margin of error and all that and that the DRA numbers are actually kind of funky to begin with but notable that Trump possibly did better amongst white voters in Virginia than in West Virginia.

EDIT: But per the exit poll Trump won 54% of white voters in Virginia. I just think the DRA numbers are real off.

DRA numbers for the state as a whole are probably more accurate than exit polls, since data smudging and the absentee ballot vagueness of VA disappear in the aggregate.

Just did some quick math and indeed the VRA numbers are impossible.

Virginia was about 58% white alone and 62% white self reported last census. If we take the lower number and don't regard higher whire turnout 0.58×0.75=over 43%, and Trump received 44% in VA. That means he would have to received almost no votes whatsoever from the non-white voters. 1/42=2.3% so accounting for third parties Biden would still have to have won over 95% of the non-white vote!

Yeah I would not take the stats of the OP as an accurate measure of the statewide vote. Doing some back of the envelope calculations, the statewide vote is around ~37% for Biden, which is basically in line with Adam Griffin's estimates in 2012. I also assumed around 95% of Black voters for Biden, so if your model of Black voters in VA has them voting a little to the right of that the white vote for Biden goes up a bit.

Virginia is only a little over 20% black though. The rest is mostly Hispanic and Asian, who actually together almost make up almost as much as the black percentage, and obviously aren't going to be >90% for Biden.

Do you not listen very well? Did you not hear where I'm saying this applies only to 90 percent or more white precincts?

That's kind of a pointless number to use.

I'd argue it's illustrative. Like why are whites, when they have the place to themselves, so much more Republican?

Ye it's not unless, but def don't extrapolate too much.

Generally speaking, I think folks who self-sort into all-white communities are naturally going to be more conservative than those who move to nearby racially diverse neighborhoods.

which is odd since it used to be that whites in diverse areas (like memphis or detroit) were much more prone to racial backlash than whites in places like Nebraska or the Dakotas. Like look at the Wallace numbers in places like downriver Detroit.

[freepcrusher]

Some other 2020 numbers I got

PA 66 percent trump
MI 63 percent
NJ 58 percent
NY 58 percent
CT 48 percent
NH 48 percent
RI 45 percent
MA 41 percent
VT 37 percent

You could argue that in some ways looking at 90 percent+ white precincts in Virginia vs West Virginia is much more an apples to apples comparison. A white living in an all white area in VA is going to have much more in common with a white in wv than a white living in oakton.

I also think I've refuted my own argument. Iowa whites are probably not unusually less Republican and the reason I thought that was because I was looking at Appalachian states first.

If Michigan is only at 63 percent, I'd guess Minnesota and Wisconsin is at high 50s/low 60s. Illinois is actually higher than Iowa but that's because a lot of the all white areas of IL are south of the 40.5 parallel (IA/MO border). So it's less Germanic Lutheran and more scots Irish.

I would say that the divisions I am seeing now are east west (between New England and New York as well as ia/mn and Nebraska/dakotas) and the more traditional north south.