What Polling Got Wrong.

Univision has an article on perhaps why there are some differences in the regular exit poll and the Latino decisions poll.

I think that is very important, Latinos/Hispanics that only speak English and do not have friends or Family that are immigrants, tend to vote more Republican than those of us that do have immigrant family and use Spanish in our lives.

Latino decisions underestimates those Hispanics, now perhaps it is because they want o capture more of the Latino community that still in touch with "Latinoness" over those that have become more assimilated.

http://www.univision.com/univision-news/politics/so-how-did-latinos-actually-vote-quien-sabe

 

What was the tipping point state? And what was Trump's percentage margin in that state?

It appears that Hillary will win the PV by a point and a half. I'm curious how bad the disconnect between PV and EV could have been.

2000 was essentially a tie in both. That's not what happened in 2016.

 

ok, not that I want to assign homework to you AB, but compared to Obama, is there a big difference?

 

That's not what I was asking. What was the state that Trump over 270, and what was his percentage win there? Let's say he won that state by 1.5%. That means Hillary could have won the PV by 2.9% (or so) and lost the EC.

One of the effects of Trump is that his election is such a humongous story that nobody is talking about the PV/EC split, or hardly anyone. At least compared to what we'd be seeing (IMHO) if Kasich had been the winner in the same manner as Trump.

 

Here is some interesting info regarding this:

https://en.wikipedia.org/wiki/United_States_presidential_election,_2016

States where the margin of victory was under 1% (50 electoral votes; 46 won by Trump, 4 by Clinton):

1. Michigan, 0.27%
2. New Hampshire, 0.37%
3. Wisconsin, 0.81%
4. Pennsylvania, 0.96%

That is barely 100k votes, or about .157% of her total. Hillary just got the votes in the wrong places, sort off.

 

@American Brummie can you translate this to English for me.

The economist does agree with you, that Republicans are dying off, but Trump did find a new batch of dying voters to get elected.

https://www.economist.com/news/unit...es-predict-trumpward-swings-illness-indicator

 

"Unsurprisingly, the single best predictor identified so far of the change from 2012 to 2016 in the share of each county’s eligible voters that voted Republican—in other words, the swing from Mitt Romney to Mr Trump—is the percentage of potential voters who are non-college whites."

If a county had more non-college whites that never voted, the more likely Trump got more votes in that county?

 

@American Brummie you may be interested on this, how Jared Kushner won Trump the presidency.

 

Which makes me wonder how many voters polled who responded that they would definitely vote, found that they had been purged from the voter rolls in those key states? IOW is the MOE here not really a polling error but more of a respondents' November surprise?

Supposedly there were something like a half million purged from the Wisconsin rolls that no one was really sure weren't valid and present voters-- what happened to those who were? Did they file provisionals, and if so were they counted? Or were they just turned away, and went home and ate their livers with some fava beans and a nice chianti?

 

This is fascinating. It's a representation of counties by their economy. (Keep in mind that it's based on where people WORK not where they LIVE.) Red=Trump, blue=Hillary




http://www.vox.com/new-money/2016/11/23/13715276/clinton-trump-booming-economy

The article goes on to say that Gore won counties with 54% of the economy, Hillary won 64%. That seems like a very, very significant change in 16 years. (Hillary is running about a point better than Gore, so all else being equal, his 54% would be her 55%.)

Also, of the approximately 3000 counties in the US, Gore won 659. Hillary won 493. So you don't have to do the math, she won 75% of the number of counties Gore won, but she's doing ever-so-slightly better.

Here's a very, very interesting graphic



In the Clinton recovery, rural areas did quite well. (NAFTA and farm export related?) In the Bush recovery, really small population areas did poorly, the rest are a mixed bag. But in the Obama recovery! Oh my word.

For those of you who want to make the "it's the economy, stupid" case, here are the arrows for your quiver.

 

We are a party of just the populated urban centers, that is our bubble.

Anyone outside large cities is a racist for not thinking like we do.

I know very simplistic on my part, but how can the DNC at is current form listen and engage some of those rural "deplorables" ?


Edit: by we I mean you, I am supposed to be green party

 

Ironically those graphs suggest that if Democrats just stay patient, eventually they can win just on the backs of these big urban centers. And not just the coastal cities like LA and NY. Those numbers show why Clinton performed well in Texas and Georgia ... mostly thanks to Houston metro, Dallas metro and Atlanta metro. Phoenix metro looks like the last major urban stronghold for GOP, and its days are probably numbered. The long term trend still doesn't look good for GOP. In many ways you can interpret this election as the last major white working class revolt. These small red counties in the Midwest are depopulating and will not be able to carry elections in the future. It'll be interesting to see what Dems do in the future. The math implies that they'll be fine in the long run if they double down on the diverse urban vote. Whether that's a moral or ethical thing to do is unclear. I still prefer a 50 state strategy where everyone has a stake, but who knows if that's really possible in this hyper polarized environment.

 

It is a very good and sound argument, if you ignore who they have just elected to steer the economy. In the same page as the article you posted, you can find this other piece:

http://www.vox.com/policy-and-politics/2016/11/22/13641654/paul-ryan-trump-poverty-safety-net

I’ve been working on these issues since 1972,” Robert Greenstein, the founder and president of the Center on Budget and Policy Priorities and Washington’s leading advocate for poor and low-income Americans, says. “This is by far the gravest threat to the safety net, and to low-income people, that I’ve seen in my close to half a century of working on these issues. I think there’s a potential in the first seven months, by the August recess, for Congress to pass policies that do more to increase poverty and hardship and widen inequality than we’ve seen in half a century.”


Ryan’s proposals would repudiate the federal government’s 50-year guarantee of medical care and food to America’s poorest residents, a promise generated by Lyndon B. Johnson when he made food stamps permanent in 1964 and created Medicaid in 1965. The expectation has not always been met, especially for childless adults, whom Medicaid largely did not cover until the Affordable Care Act, and whom non-expansion states still don’t offer coverage to. But for the poorest families with children, those two programs were there, providing at least modest assistance in desperate times.

If Paul Ryan and his allies in the House enact their domestic agenda, this promise will fall apart. The Affordable Care Act will be repealed, but the damage will hardly end there. Medicaid will see its funding gutted, federal guarantees of coverage and access removed, and its status as an entitlement upon which poor Americans can depend destroyed. The Supplemental Nutritional Assistance Program (SNAP, or “food stamps”) will be slashed and turned over to states, which will likely use the money as a slush fund for other endeavors.​

Why do poor and middle class working people keep voting for politicians that basically want them to have to respite? Yeah, messaging is part of the equation, but you know there are other elements..

 

The bolded assumes that districts won't be shaped to be, ummm, favorably disposed to such depopulating areas at the expense of urban centers

 

In my not particularly urban bubble I have regular contact (as in, exchanging words if not actually conversing in person, face to face) with Catholics, Muslims, Jews, atheists and protestants, as well as nones. Mostly heterosexual, but some gay and lesbian folks, and one guy I've worked with in the padt has a son who is likely going to go transgendered in the future. Mostly white people, but also African American, Indian, Pakistani, various asian and various middle eastern and even, if I keep my class a bit long, a Kurdish guy who teaches the class after mine. Most of these people don't care what I do on my own time, and I return the favor. In this group, there are two people I know for a fact voted for Trump. Probably more.

Outside large cities, though, at least the part where I grew up, it's not like this. In fact, that is where people live in bubbles where they rarely encounter anyone different from them. Throw a steady stream of pre-selected social media news streams ... and you have people who aren't that good at dealing with a diverse population. Though to be fair, they ARE better at it than the average Saudi I've ever had to work with.

 

As it turns out, the national polls were spot on.