The peculiarities of the Wisconsin GOP primary

[uti2]

The fact that Trump got murdered so hard in the WOW area of Wisconsin alone while winning every major urban area in other parts of the Midwest shows the differences between Wisconsin and other Midwestern states and why WI can't be used as a template. Pretty much every WI Political Activist and Political consultant understands this dynamic. Voting patterns in this state weren't typical:

http://www.politico.com/story/2016/03/why-wisconsins-never-trump-movement-is-different-221407




Just compare Trump's performance in Milwuakee to his performance in Chicago or Detroit, it's not even a comparison.

Coordinated anti-Trump voting was a phenomenon specific to WOW, not other parts of the urban Midwest. Trump barely received ~20% of the vote in WOW, with Cruz getting ~60%. Those are lopsided voting patterns that are specific to a certain region and didn't appear as a pattern in any other state.


From Scott Walker's campaign manager, even he flat out stated this:

http://www.usnews.com/news/blogs/run-2016/2016/03/30/the-talkers-may-torpedo-trump-in-wisconsin

"Trump would win Wisconsin absent talk radio."

That's a pretty bold and direct claim. This is the guy who worked on Walker's election and re-election in the state. Don't know how much more clear it can get, when even Scott Walker's own political machine admits this.

[(Still) muted by Kalwejt until March 31]

Sioux County, IA and Ottawa County, MI are similar cases. Very socially conservative, but contemptuous of DJT.

[RI]

The WOW counties voted very similarly to the Columbus area in Ohio (only for Cruz rather than Kasich, of course).

[Vosem]

I wanted to point out that I live in Columbus, which if anything voted even harder against Trump. Anyway, while the exact regional breakdowns in different Midwestern states were, well, different, IL/MI/OH/WI all basically had the same result (39% Trump in Illinois, 36% Trump in Michigan and Ohio, and 35% in Wisconsin). The idea that Wisconsin was some bizarre outlier isn't born out by the actual election results. This isn't even about polling, where you can cherry-pick: we have actual results.

Sioux County, IA and Ottawa County, MI are similar cases. Very socially conservative, but contemptuous of DJT.

Northwest Arkansas and Southwest Missouri fit this pattern too; Fort Bend would've if not for the general Trump surge happening by that time. Socially conservative Midwesterners were much more viscerally anti-Trump than socially conservative Southerners, who actually mostly gave Trump a plurality of their votes.

[100% pro-life no matter what]

I wanted to point out that I live in Columbus, which if anything voted even harder against Trump. Anyway, while the exact regional breakdowns in different Midwestern states were, well, different, IL/MI/OH/WI all basically had the same result (39% Trump in Illinois, 36% Trump in Michigan and Ohio, and 35% in Wisconsin). The idea that Wisconsin was some bizarre outlier isn't born out by the actual election results. This isn't even about polling, where you can cherry-pick: we have actual results.

Sioux County, IA and Ottawa County, MI are similar cases. Very socially conservative, but contemptuous of DJT.

Northwest Arkansas and Southwest Missouri fit this pattern too; Fort Bend would've if not for the general Trump surge happening by that time. Socially conservative Midwesterners were much more viscerally anti-Trump than socially conservative Southerners, who actually mostly gave Trump a plurality of their votes.

Trump's percentage wasn't that high in Southern states that he won like Arkansas, Tennessee, Kentucky, Louisiana, Georgia, and South Carolina.

[Vosem]

I wanted to point out that I live in Columbus, which if anything voted even harder against Trump. Anyway, while the exact regional breakdowns in different Midwestern states were, well, different, IL/MI/OH/WI all basically had the same result (39% Trump in Illinois, 36% Trump in Michigan and Ohio, and 35% in Wisconsin). The idea that Wisconsin was some bizarre outlier isn't born out by the actual election results. This isn't even about polling, where you can cherry-pick: we have actual results.

Sioux County, IA and Ottawa County, MI are similar cases. Very socially conservative, but contemptuous of DJT.

Northwest Arkansas and Southwest Missouri fit this pattern too; Fort Bend would've if not for the general Trump surge happening by that time. Socially conservative Midwesterners were much more viscerally anti-Trump than socially conservative Southerners, who actually mostly gave Trump a plurality of their votes.

Trump's percentage wasn't that high in Southern states that he won like Arkansas, Tennessee, Kentucky, Louisiana, Georgia, and South Carolina.

Well, Kentucky is a caucus, so it's not really directly comparable to the others -- while Trump did pretty poorly, and would've lost to a unified alternative, in a primary he would've destroyed the opposition (the same is true for Maine). As for Arkansas, the northwestern section of the state has long voted more like (a very socially conservative part of) the rural Midwest than the rural South; Trump came in third, behind Rubio and Cruz, in Washington and Benton counties, for instance. TN/LA/GA/SC reflect a more general Southern pattern, where Trump was stronger than in the Midwest (41 in Louisiana, and 39 in Tennessee and Georgia; South Carolina, as an early state, received way more attention and campaigning than its neighbors and so voted somewhat differently); the more "Deep South" a state was, the stronger Trump was there (he was at 43 in MS, and 47 in AL).

Of these states, AR would've definitely voted for any unified oppositionist candidate, SC could've voted for a stronger Rubio, LA or TN could've voted for a stronger Cruz, and GA could've voted for either if Trump was at an unfortunate moment in the campaign. MS and AL were basically safe for Trump from the beginning of the primary season.

[uti2]

I wanted to point out that I live in Columbus, which if anything voted even harder against Trump. Anyway, while the exact regional breakdowns in different Midwestern states were, well, different, IL/MI/OH/WI all basically had the same result (39% Trump in Illinois, 36% Trump in Michigan and Ohio, and 35% in Wisconsin). The idea that Wisconsin was some bizarre outlier isn't born out by the actual election results. This isn't even about polling, where you can cherry-pick: we have actual results.

Sioux County, IA and Ottawa County, MI are similar cases. Very socially conservative, but contemptuous of DJT.

Northwest Arkansas and Southwest Missouri fit this pattern too; Fort Bend would've if not for the general Trump surge happening by that time. Socially conservative Midwesterners were much more viscerally anti-Trump than socially conservative Southerners, who actually mostly gave Trump a plurality of their votes.

OH, which had a popular incumbent governor in the running where the state voted for their Governor. This pattern you saw in WI simply doesn't appear anywhere else in the region. Trump got 20-25% in the WOW area v. Cruz who had 60%, that was an overwhelmingly consolidation against Trump specifically concentrated in that region.

[Vosem]

I wanted to point out that I live in Columbus, which if anything voted even harder against Trump. Anyway, while the exact regional breakdowns in different Midwestern states were, well, different, IL/MI/OH/WI all basically had the same result (39% Trump in Illinois, 36% Trump in Michigan and Ohio, and 35% in Wisconsin). The idea that Wisconsin was some bizarre outlier isn't born out by the actual election results. This isn't even about polling, where you can cherry-pick: we have actual results.

Sioux County, IA and Ottawa County, MI are similar cases. Very socially conservative, but contemptuous of DJT.

Northwest Arkansas and Southwest Missouri fit this pattern too; Fort Bend would've if not for the general Trump surge happening by that time. Socially conservative Midwesterners were much more viscerally anti-Trump than socially conservative Southerners, who actually mostly gave Trump a plurality of their votes.

OH, which had a popular incumbent governor in the running where the state voted for their Governor. This pattern you saw in WI simply doesn't appear anywhere else in the region. Trump got 20-25% in the WOW area v. Cruz who had 60%, that was an overwhelmingly consolidation against Trump specifically concentrated in that region.

The WOW area was not any more anti-Trump than the Columbus area, and ascribing this to the popular Governor is disingenuous because Ted Cruz, who had no connection to Wisconsin, achieved the same thing. Wisconsin also had a greater than typical number of rural counties which gave an outright majority of the vote to Trump -- 13 compared to Michigan's 6, and Wisconsin's counties are a larger fraction of the state's population.

Explain why it was that the differences between Trump's percentages were less than 2% in Wisconsin, Michigan, and Ohio, and barely outside of that range in Illinois.

[uti2]

Of these states, AR would've definitely voted for any unified oppositionist candidate, SC could've voted for a stronger Rubio, LA or TN could've voted for a stronger Cruz, and GA could've voted for either if Trump was at an unfortunate moment in the campaign. MS and AL were basically safe for Trump from the beginning of the primary season.

Rubio did as 'strong' as he could've vis-a-vis Cruz and Trump in SC, pretty much those who were set on him, already voted for him. Jeb was always going to stay in SC, and all of Jeb's establishment voters had pretty much already abandoned him, the rump of voters that were left, many of them, simply didn't like rubio. A lot of those Cruz voters were Huckabee/Santorum voters who were dead-set on Cruz as it was. Ron Paul voters mostly split between Trump and Cruz.

Rubio was never in a position where he polled in the 20s in NH, he was always in the teens, that's why the margins between rubio and the other candidates were always close there, that's about as much momentum as he possibly could've gotten, which he received anyway in SC with Haley, etc. coming out for him.

What happened in SC, was Rubio viciously attacking Cruz in alliance with Trump, in the whole 'Lyin Ted' act.

Trump didn't care that those attacks were damaging him too, he just wanted to prevent the consolidation behind Cruz, he wouldn't have even needed to do that otherwise.

And when Rubio was forced to attack later on after Jeb was out, his numbers also fell.

[uti2]

I wanted to point out that I live in Columbus, which if anything voted even harder against Trump. Anyway, while the exact regional breakdowns in different Midwestern states were, well, different, IL/MI/OH/WI all basically had the same result (39% Trump in Illinois, 36% Trump in Michigan and Ohio, and 35% in Wisconsin). The idea that Wisconsin was some bizarre outlier isn't born out by the actual election results. This isn't even about polling, where you can cherry-pick: we have actual results.

Sioux County, IA and Ottawa County, MI are similar cases. Very socially conservative, but contemptuous of DJT.

Northwest Arkansas and Southwest Missouri fit this pattern too; Fort Bend would've if not for the general Trump surge happening by that time. Socially conservative Midwesterners were much more viscerally anti-Trump than socially conservative Southerners, who actually mostly gave Trump a plurality of their votes.

OH, which had a popular incumbent governor in the running where the state voted for their Governor. This pattern you saw in WI simply doesn't appear anywhere else in the region. Trump got 20-25% in the WOW area v. Cruz who had 60%, that was an overwhelmingly consolidation against Trump specifically concentrated in that region.

The WOW area was not any more anti-Trump than the Columbus area, and ascribing this to the popular Governor is disingenuous because Ted Cruz, who had no connection to Wisconsin, achieved the same thing. Wisconsin also had a greater than typical number of rural counties which gave an outright majority of the vote to Trump -- 13 compared to Michigan's 6, and Wisconsin's counties are a larger fraction of the state's population.

Explain why it was that the differences between Trump's percentages were less than 2% in Wisconsin, Michigan, and Ohio, and barely outside of that range in Illinois.

As was explained, WOW was a region that specifically consolidated against Trump, where Trump got blown out 60-20, this simply didn't happen anywhere else. For OH, this is like calling Cruz's win in TX, disingenuous. Kasich's win in OH, was a closer parallel to Cruz's win in TX.

[uti2]

Go look at 2008, Huckabee won 60% of the vote in Arkansas, whereas he only received 43-41 v. Mccain in Louisiana. Having a home state advantage makes a large difference. Mccain also received 60% of the vote in AZ in the 2000 primary

[Vosem]

Of these states, AR would've definitely voted for any unified oppositionist candidate, SC could've voted for a stronger Rubio, LA or TN could've voted for a stronger Cruz, and GA could've voted for either if Trump was at an unfortunate moment in the campaign. MS and AL were basically safe for Trump from the beginning of the primary season.

Rubio did as 'strong' as he could've vis-a-vis Cruz and Trump in SC, pretty much those who were set on him, already voted for him. Jeb was always going to stay in SC, and all of Jeb's establishment voters had pretty much already abandoned him, the rump of voters that were left, many of them, simply didn't like rubio.

This isn't the case; already in SC, a large fraction of floating anti-Trump voters backed Cruz (who rose from the low teens in early SC polling to finish in the low 20s), and virtually all of Kasich's voters in the South had Rubio as a second choice. It's true that many of Bush's voters didn't (many of them actually went over to Trump), but this wouldn't have been enough to prevent the Rubio victory.

A lot of those Cruz voters were Huckabee/Santorum voters who were dead-set on Cruz as it was. Ron Paul voters mostly split between Trump and Cruz.

I haven't seen any actual studies, and I'd be interested in seeing them, but in my experience Ron Paul voters over a certain age (35-40) tended to go over to Trump, while those under that age tended to become floating anti-Trump voters, and mostly backed Rubio towards the beginning of the primary season.

Rubio was never in a position where he polled in the 20s in NH, he was always in the teens, that's why the margins between rubio and the other candidates were always close there,

There was a large population of undecided voters in NH that broke to Kasich at the very end after having cycled through many different candidates (they supported Fiorina in September-October, when she was in the mid-teens). During Rubio's post-Iowa surge, they certainly backed him, and it's a bit conservative to suggest he would only have made low twenties; considering he was also taking support directly from Trump during the surge, he may well have done better and come within single digits of the Orange Man himself.

that's about as much momentum as he possibly could've gotten, which he received anyway in SC with Haley, etc. coming out for him.

Haley endorsing him helped him, but at the very end Cruz was still in second and a large number of floating anti-Trump voters in SC voted for Cruz.

What happened in SC, was Rubio viciously attacking Cruz in alliance with Trump, in the whole 'Lyin Ted' act.

Trump didn't care that those attacks were damaging him too, he just wanted to prevent the consolidation behind Cruz, he wouldn't have even needed to do that otherwise.

And when Rubio was forced to attack later on after Jeb was out, his numbers also fell.

Sure, but this is a little beside the point. There was a moment when Rubio had consolidated all the floating anti-Trump voters and was also taking some from Trump himself. Had he sustained that, and without Christie there's no reason to suspect he wouldn't have, he would've been nominated easily. Here's (roughly) how the voters in the national primary broke down:

30% Hardcore Trumpists. Supported Trump from the beginning to the end.
5% Shiny Object Lovers. Mostly supported Trump throughout the primary season, but went over to Rubio during his surge post-Iowa.
5% Undecideds. These guys were legitimately uncertain throughout the primary season. Mostly ended up scattering among various anti-Trumps (especially in early Midwestern contests virtually none went for Trump), but broke hard for Trump starting with New York.
10% The Faint of Heart. Anti-Trump voters, most of whom started with Rubio and went over to Cruz, who stopped voting once it became clear the options were either a Trump win or a contested convention, around New York. Prevented the contested convention.
20% Against Trump All Costs. Floating anti-Trump voters at their most pure; started with Rubio, broke for Kasich in a few states (NH/VT/MA/OH), started crossing over to Cruz in SC and were fully consolidated behind him by the WI primary. About a third of these guys did not support Trump in the general election.
15% Fundie Base. These guys backed Cruz from the very beginning (Iowa) to the very end (Indiana). All went over to Trump by the end of the general election.
15% NeverTrump and NeverCruz. Very prominent in the Northeast; were split between Rubio and Kasich towards the beginning (though most backed Rubio during his surge); eventually coalesced behind Kasich. A majority, maybe two-thirds or more, of these guys, didn't support Trump in the general election.

Against Trump and Cruz, Rubio had a national ceiling over 50%. I doubt he would've made it (I tend to think the Shiny Object voters would've returned to Trump quickly, for instance), but it was doable.

I wanted to point out that I live in Columbus, which if anything voted even harder against Trump. Anyway, while the exact regional breakdowns in different Midwestern states were, well, different, IL/MI/OH/WI all basically had the same result (39% Trump in Illinois, 36% Trump in Michigan and Ohio, and 35% in Wisconsin). The idea that Wisconsin was some bizarre outlier isn't born out by the actual election results. This isn't even about polling, where you can cherry-pick: we have actual results.

Sioux County, IA and Ottawa County, MI are similar cases. Very socially conservative, but contemptuous of DJT.

Northwest Arkansas and Southwest Missouri fit this pattern too; Fort Bend would've if not for the general Trump surge happening by that time. Socially conservative Midwesterners were much more viscerally anti-Trump than socially conservative Southerners, who actually mostly gave Trump a plurality of their votes.

OH, which had a popular incumbent governor in the running where the state voted for their Governor. This pattern you saw in WI simply doesn't appear anywhere else in the region. Trump got 20-25% in the WOW area v. Cruz who had 60%, that was an overwhelmingly consolidation against Trump specifically concentrated in that region.

The WOW area was not any more anti-Trump than the Columbus area, and ascribing this to the popular Governor is disingenuous because Ted Cruz, who had no connection to Wisconsin, achieved the same thing. Wisconsin also had a greater than typical number of rural counties which gave an outright majority of the vote to Trump -- 13 compared to Michigan's 6, and Wisconsin's counties are a larger fraction of the state's population.

Explain why it was that the differences between Trump's percentages were less than 2% in Wisconsin, Michigan, and Ohio, and barely outside of that range in Illinois.

As was explained, WOW was a region that specifically consolidated against Trump, where Trump got blown out 60-20, this simply didn't happen anywhere else. For OH, this is like calling Cruz's win in TX, disingenuous. Kasich's win in OH, was a closer parallel to Cruz's win in TX.

Why is it then that Michigan's statewide result was so similar to Wisconsin's? Stop dodging the result. Nobody had any favorite-son effect in Michigan.

[Vosem]

Go look at 2008, Huckabee won 60% of the vote in Arkansas, whereas he only received 43-41 v. Mccain in Louisiana. Having a home state advantage makes a large difference. Mccain also received 60% of the vote in AZ in the 2000 primary

I'm not denying this. But I don't think it negates the fact that the areas in Ohio that voted against Trump were demographically similar to ones outside of Ohio that also voted against Trump. Or that the difference between Trump's result in Ohio and his result in Michigan was -- wait -- just 0.60%. The difference in Kasich's result was 22.74%. There's your home-state effect. But it's pretty clear that levels for Trump weren't affected by that virtually anywhere. In Texas he did very similarly as in Oklahoma, while Cruz was double-digits weaker in Oklahoma than Texas.

[uti2]

Of these states, AR would've definitely voted for any unified oppositionist candidate, SC could've voted for a stronger Rubio, LA or TN could've voted for a stronger Cruz, and GA could've voted for either if Trump was at an unfortunate moment in the campaign. MS and AL were basically safe for Trump from the beginning of the primary season.

Rubio did as 'strong' as he could've vis-a-vis Cruz and Trump in SC, pretty much those who were set on him, already voted for him. Jeb was always going to stay in SC, and all of Jeb's establishment voters had pretty much already abandoned him, the rump of voters that were left, many of them, simply didn't like rubio.

This isn't the case; already in SC, a large fraction of floating anti-Trump voters backed Cruz (who rose from the low teens in early SC polling to finish in the low 20s), and virtually all of Kasich's voters in the South had Rubio as a second choice. It's true that many of Bush's voters didn't (many of them actually went over to Trump), but this wouldn't have been enough to prevent the Rubio victory.

A lot of those Cruz voters were Huckabee/Santorum voters who were dead-set on Cruz as it was. Ron Paul voters mostly split between Trump and Cruz.

I haven't seen any actual studies, and I'd be interested in seeing them, but in my experience Ron Paul voters over a certain age (35-40) tended to go over to Trump, while those under that age tended to become floating anti-Trump voters, and mostly backed Rubio towards the beginning of the primary season.

Rubio was never in a position where he polled in the 20s in NH, he was always in the teens, that's why the margins between rubio and the other candidates were always close there,

There was a large population of undecided voters in NH that broke to Kasich at the very end after having cycled through many different candidates (they supported Fiorina in September-October, when she was in the mid-teens). During Rubio's post-Iowa surge, they certainly backed him, and it's a bit conservative to suggest he would only have made low twenties; considering he was also taking support directly from Trump during the surge, he may well have done better and come within single digits of the Orange Man himself.

that's about as much momentum as he possibly could've gotten, which he received anyway in SC with Haley, etc. coming out for him.

Haley endorsing him helped him, but at the very end Cruz was still in second and a large number of floating anti-Trump voters in SC voted for Cruz.

What happened in SC, was Rubio viciously attacking Cruz in alliance with Trump, in the whole 'Lyin Ted' act.

Trump didn't care that those attacks were damaging him too, he just wanted to prevent the consolidation behind Cruz, he wouldn't have even needed to do that otherwise.

And when Rubio was forced to attack later on after Jeb was out, his numbers also fell.

Sure, but this is a little beside the point. There was a moment when Rubio had consolidated all the floating anti-Trump voters and was also taking some from Trump himself. Had he sustained that, and without Christie there's no reason to suspect he wouldn't have, he would've been nominated easily. Here's (roughly) how the voters in the national primary broke down:

30% Hardcore Trumpists. Supported Trump from the beginning to the end.
5% Shiny Object Lovers. Mostly supported Trump throughout the primary season, but went over to Rubio during his surge post-Iowa.
5% Undecideds. These guys were legitimately uncertain throughout the primary season. Mostly ended up scattering among various anti-Trumps (especially in early Midwestern contests virtually none went for Trump), but broke hard for Trump starting with New York.
10% The Faint of Heart. Anti-Trump voters, most of whom started with Rubio and went over to Cruz, who stopped voting once it became clear the options were either a Trump win or a contested convention, around New York. Prevented the contested convention.
20% Against Trump All Costs. Floating anti-Trump voters at their most pure; started with Rubio, broke for Kasich in a few states (NH/VT/MA/OH), started crossing over to Cruz in SC and were fully consolidated behind him by the WI primary. About a third of these guys did not support Trump in the general election.
15% Fundie Base. These guys backed Cruz from the very beginning to the very end. All went over to Trump by the end.
15% NeverTrump and NeverCruz. Very prominent in the Northeast; were split between Rubio and Kasich towards the beginning (though most backed Rubio during his surge); eventually coalesced behind Kasich. A majority, maybe two-thirds or more, of these guys, didn't support Trump in the general election.

Against Trump and Cruz, Rubio had a national ceiling over 50%. I doubt he would've made it (I tend to think the Shiny Object voters would've returned to Trump quickly, for instance), but it was doable.

I wanted to point out that I live in Columbus, which if anything voted even harder against Trump. Anyway, while the exact regional breakdowns in different Midwestern states were, well, different, IL/MI/OH/WI all basically had the same result (39% Trump in Illinois, 36% Trump in Michigan and Ohio, and 35% in Wisconsin). The idea that Wisconsin was some bizarre outlier isn't born out by the actual election results. This isn't even about polling, where you can cherry-pick: we have actual results.

Sioux County, IA and Ottawa County, MI are similar cases. Very socially conservative, but contemptuous of DJT.

Northwest Arkansas and Southwest Missouri fit this pattern too; Fort Bend would've if not for the general Trump surge happening by that time. Socially conservative Midwesterners were much more viscerally anti-Trump than socially conservative Southerners, who actually mostly gave Trump a plurality of their votes.

OH, which had a popular incumbent governor in the running where the state voted for their Governor. This pattern you saw in WI simply doesn't appear anywhere else in the region. Trump got 20-25% in the WOW area v. Cruz who had 60%, that was an overwhelmingly consolidation against Trump specifically concentrated in that region.

The WOW area was not any more anti-Trump than the Columbus area, and ascribing this to the popular Governor is disingenuous because Ted Cruz, who had no connection to Wisconsin, achieved the same thing. Wisconsin also had a greater than typical number of rural counties which gave an outright majority of the vote to Trump -- 13 compared to Michigan's 6, and Wisconsin's counties are a larger fraction of the state's population.

Explain why it was that the differences between Trump's percentages were less than 2% in Wisconsin, Michigan, and Ohio, and barely outside of that range in Illinois.

As was explained, WOW was a region that specifically consolidated against Trump, where Trump got blown out 60-20, this simply didn't happen anywhere else. For OH, this is like calling Cruz's win in TX, disingenuous. Kasich's win in OH, was a closer parallel to Cruz's win in TX.

Why is it then that Michigan's statewide result was so similar to Wisconsin's? Stop dodging the result. Nobody had any favorite-son effect in Michigan.

Where is the source for all Kasich voters in the south having that as a second choice?

Why would Trump need to attack Cruz and lower his numbers if that was the case?, Rubio would've gotten momentum to keep his numbers competitive with Cruz anyway.

Those voters in IA were upset with Trump skipping the debate, so you keep talking about a hypothetical with Christie, but what if Trump had attended that IA debate, and then won IA? Where would rubio be after that? Trump was currently attacking Cruz at the time, and rubio really wasn't attacked so he skirted by as another option in IA, that's also exactly what happened in SC. When rubio was forced into a position where he also had to attack, after SC, his numbers also went down. But there is no data to suggest that rubio was in the 20s in NH, which is a primary, not a caucus, and numbers are generally less volatile.  



Trump got hammered in the WOW area of WI, there was momentum going for Trump overall, but it didn't matter if the WOW area opposed him. The midwest generally went which ever way momentum went at any given time, it was a Trump-neutral area.

Go look at 2008, Huckabee won 60% of the vote in Arkansas, whereas he only received 43-41 v. Mccain in Louisiana. Having a home state advantage makes a large difference. Mccain also received 60% of the vote in AZ in the 2000 primary

I'm not denying this. But I don't think it negates the fact that the areas in Ohio that voted against Trump were demographically similar to ones outside of Ohio that also voted against Trump. Or that the difference between Trump's result in Ohio and his result in Michigan was -- wait -- just 0.60%. The difference in Kasich's result was 22.74%. There's your home-state effect. But it's pretty clear that levels for Trump weren't affected by that virtually anywhere. In Texas he did very similarly as in Oklahoma, while Cruz was double-digits weaker in Oklahoma than Texas.

Why did 80% of Arkansas vote against Mccain in 2008?

[uti2]

It pretty much comes down to this, the actual data when you break down second choice rankings for the candidates simply doesn't show that.

https://www.theatlantic.com/politics/archive/2016/02/the-ceiling-buster/470919/


The rest, when you calculate the math based on the second choice preferences pretty much gets you the same result with Trump.

It also doesn't work when you look at the primary results:

Look at the March 15 results, Trump pretty much got 39% in IL, 41% in MO and 40% in NC, that's all pretty proportional. There is no evidence there to suggest that the midwest was anti-trump.

http://www.nytimes.com/elections/2016/results/2016-03-15

The only outlying state there was Kasich with his home state advantage.

The fact that Kasich, Mccain, Huckabee, and Cruz all did well in their home states fits a pattern.

By the way, the biggest hole in your theory, if Kasich voters were purely anti-trump voters and 100% of kasich's voters would've gone to rubio in the south, why didn't they show any movement in that direction to test your theory in FL?

[uti2]

Romney was also polling at ~35% until his final few weeks.

http://www.realclearpolitics.com/epolls/2012/president/us/republican_presidential_nomination-1452.html

The biggest factor that can't be ignored is this. The Establishment actually is conservative, the way the establishment candidate wins is by generally being an establishment/moderate hybrid candidate, taking camps from both voters establishment and moderates, which are 2 different categories. Romney was the quintessential establishment candidate, fiscally conservative, but socially somewhat moderate.

Rubio was not a moderate, and moderate voters understood this. With socially liberal, pro-gay/pro-choice republicans, Trump was leading Rubio by similar margins as Cruz. This is why when you calculate the net preferences it shows what it does.

A ton of those Kasich voters weren't establishment, they were socially liberal moderates, and that's why many of them would go to Trump in a split:

https://www.thenation.com/article/how-prochoice-republicans-are-helping-donald-trump/

[100% pro-life no matter what]

Here's (roughly) how the voters in the national primary broke down:

30% Hardcore Trumpists. Supported Trump from the beginning to the end.
5% Shiny Object Lovers. Mostly supported Trump throughout the primary season, but went over to Rubio during his surge post-Iowa.
5% Undecideds. These guys were legitimately uncertain throughout the primary season. Mostly ended up scattering among various anti-Trumps (especially in early Midwestern contests virtually none went for Trump), but broke hard for Trump starting with New York.
10% The Faint of Heart. Anti-Trump voters, most of whom started with Rubio and went over to Cruz, who stopped voting once it became clear the options were either a Trump win or a contested convention, around New York. Prevented the contested convention.
20% Against Trump All Costs. Floating anti-Trump voters at their most pure; started with Rubio, broke for Kasich in a few states (NH/VT/MA/OH), started crossing over to Cruz in SC and were fully consolidated behind him by the WI primary. About a third of these guys did not support Trump in the general election.
15% Fundie Base. These guys backed Cruz from the very beginning (Iowa) to the very end (Indiana). All went over to Trump by the end of the general election.
15% NeverTrump and NeverCruz. Very prominent in the Northeast; were split between Rubio and Kasich towards the beginning (though most backed Rubio during his surge); eventually coalesced behind Kasich. A majority, maybe two-thirds or more, of these guys, didn't support Trump in the general election.

Against Trump and Cruz, Rubio had a national ceiling over 50%. I doubt he would've made it (I tend to think the Shiny Object voters would've returned to Trump quickly, for instance), but it was doable.

Interesting group descriptions, though I would say that the percentages of those who wound up not supporting Trump may be a little high.  Most of the Rubio-Cruz (anti-Trump) group wound up behind him in November, and I would bet that over half of the hyper moderate (don't nominate Trump or Cruz) group wound up coming home to Trump over the last week.

[ElectionsGuy]

The few differences between Wisconsin and other midwest states was that Wisconsin didn't have Rubio in the running and the Cruz vote was more concentrated than in other states, and the Kasich vote was down.

Talk radio and Scott Walker just boosted Cruz to a higher level than he would have been at otherwise.

[uti2]

Here's (roughly) how the voters in the national primary broke down:

30% Hardcore Trumpists. Supported Trump from the beginning to the end.
5% Shiny Object Lovers. Mostly supported Trump throughout the primary season, but went over to Rubio during his surge post-Iowa.
5% Undecideds. These guys were legitimately uncertain throughout the primary season. Mostly ended up scattering among various anti-Trumps (especially in early Midwestern contests virtually none went for Trump), but broke hard for Trump starting with New York.
10% The Faint of Heart. Anti-Trump voters, most of whom started with Rubio and went over to Cruz, who stopped voting once it became clear the options were either a Trump win or a contested convention, around New York. Prevented the contested convention.
20% Against Trump All Costs. Floating anti-Trump voters at their most pure; started with Rubio, broke for Kasich in a few states (NH/VT/MA/OH), started crossing over to Cruz in SC and were fully consolidated behind him by the WI primary. About a third of these guys did not support Trump in the general election.
15% Fundie Base. These guys backed Cruz from the very beginning (Iowa) to the very end (Indiana). All went over to Trump by the end of the general election.
15% NeverTrump and NeverCruz. Very prominent in the Northeast; were split between Rubio and Kasich towards the beginning (though most backed Rubio during his surge); eventually coalesced behind Kasich. A majority, maybe two-thirds or more, of these guys, didn't support Trump in the general election.

Against Trump and Cruz, Rubio had a national ceiling over 50%. I doubt he would've made it (I tend to think the Shiny Object voters would've returned to Trump quickly, for instance), but it was doable.

Interesting group descriptions, though I would say that the percentages of those who wound up not supporting Trump may be a little high.  Most of the Rubio-Cruz (anti-Trump) group wound up behind him in November, and I would bet that over half of the hyper moderate (don't nominate Trump or Cruz) group wound up coming home to Trump over the last week.

There is an overestimation in the numbers of what percentage of voters were inherently anti-Trump. Most Republicans had a favorable view of Trump overall consistently throughout the primary. WI was one of the few states were they had a negative view of him from the beginning, this trend was mainly limited to a few western states west of the Mississippi, but east of the west coast/nv/az.

This was talked about as a unique phenomenon in the local WI press, even in 2015:

http://www.pressreader.com/usa/milwaukee-journal-sentinel/20151220/281496455241223

That's why UT was so opposed to Trump, they're the quintessential traditionally Western state.



https://www.nytimes.com/2015/12/31/upshot/donald-trumps-strongest-supporters-a-certain-kind-of-democrat.html

The few differences between Wisconsin and other midwest states was that Wisconsin didn't have Rubio in the running and the Cruz vote was more concentrated than in other states, and the Kasich vote was down.

Talk radio and Scott Walker just boosted Cruz to a higher level than he would have been at otherwise.

WI was always uniquely anti-Trump and where Trump had one of the lowest favorables of republicans v. Trump's national numbers and other midwestern states. The reason for that is WOW/Walker/Talk radio. Everyone in the local WI media discussed this:

http://www.pressreader.com/usa/milwaukee-journal-sentinel/20151220/281496455241223

[TJ in Oregon]

The Wisconsin Republicans have maintained more of an upscale educated conservative demographic than the Republicans have in most other states. Trump walked unwittingly into a buzz saw the week he spent here during the primary; people were coming out of the woodwork to vote against him (Look at the primary turnout!). Had the country voted on the same day during the primaries, Wisconsin would have been one of Trump's worst states in whole country.