US Elections - furniture shopping with disaster

[Swordfish]

19 minutes ago, IheartIheartTesla said:

As I mentioned before, there is no way we have reached 200 million registered voters. For starters, the number of eligible voters in 2012 was 218 million. A good rule of thumb is that the US population grows about ~10 m a decade, so the number of eligible voters wont be much higher in 2016 compared to 2012. Getting 90% of the entire eligible voting population registered? Not happening or has not happened yet.

Think of the logistics involved in getting 55 million new people entered in. Not to mention that human nature being what it is, I'd say a change from 67% participation to 75% participation would be a big deal and require something transformational (like say a  2008 election). A 67% to 90% jump is unthinkable.

 

Could this be affected by 'motor voter' auto registration though?

I know this is relatively new in Oregon, but I'm not sure it's widespread enough to impact the numbers that much.

[OldGimletEye]

http://www.msnbc.com/rachel-maddow-show/trump-says-hell-accept-election-results-if-i-win

Wow. Even LePage thinks Trump is acting crazy.

[Kalbear]

Just now, Manhole Eunuchsbane said:

Good article. I have to admit I'm probably not giving her as much credit as she deserves. The point they made regarding the Alicia Machado set-up was especially adept. That said, it works as well as it does because of who Trump is. 

Well, yes. All of this works because of who Trump is. 

She wouldn't have done it against Cruz - she would have made him look like a lizard man instead. 

She wouldn't have done it against Rubio - she would have made him look like a robot instead.

That she had to egg him on in a fairly subtle way while not looking too 'mean' or unpresidential or unwomanly or even make him think that she was attacking him all that much is the real amazing thing. It's such a difficult performance to do, and she nailed it not once, not twice, but all three major debates. The fact that she's not getting that much credit for it should tell you what a good job she did getting under Trump's skin. Think about how many times other Republicans tried to do that and failed - and not just failed, but failed catastrophically. 

Now, I don't want to give Clinton all the credit here, because her campaign in the POTUS part has been pretty well masterful. There have been some missteps - largely with Clinton the candidate, honestly - but their being able to coordinate web videos and ad campaigns and talking points and her surrogates and fresh and superfresh news stories has been amazingly impressive in the social media age. 

[Dr. Pepper]

2 hours ago, Maithanet said:

Every day I check the early voting stats, and somehow it just makes me happy.  Elections are exciting, and thinking of all those people casting votes makes me smile.  Florida has already had 800,000 votes cast!  That's great!  Like every year, I'm hoping for a good turnout. 

Democracy isn't perfect, but it certainly has it's moments. 

Another day of super long lines in Georgia.  Drove by this morning, then around lunch, and just a few minutes ago.  This seems crazy!

[Kalbear]

Also, in case you had any considerations for Stein...she's giving interviews to InfoWars. Yes...InfoWars. 

[Mr. Chatywin et al.]

7 minutes ago, Swordfish said:

I was surprised to find out it was a real word, but not surprised to find out he was using it wrong.

Well technically it was"bigly/big league." My buddy Max drew the slip of death: "Trump interrupts with "Wrong."" 

[lokisnow]

57 minutes ago, Kalbear said:

So people asked me before about why I was saying more voter turnout might mean a Trump win. This'll be long.

The voter turnout in 2012 was 126 million people. While there were more registered than that (about 140m) it was well short of the 200 million mentioned registered voters in this election. If 90% of that voted, we would still have by far the largest turnout in US history, and it wouldn't be remotely close to anything that the US has had either as a percentage (it would be close to 70% - the next highest was 62% with Obama) and as a total amount of voting.

Now, here's where it gets wonky. The way polling works is that they have a sample of voters and break them up into demographic groups, and then weigh them based on those demographics. So, for instance, if you have 10 white male college educated voters, their selections are tallied and then weighed on some value. That value is essentially arbitrary and differs from poll to poll, but is usually correlated with the amount of voters in that group that have voted in the past. So, for instance, if you have 10 white male college educated voters you'd weigh their choices based on how many turned out in the past - which, as you can see from the 538 block, is about 78%.

This is also why one AA man in Illinois who is voting for Trump skews the LATimes/USC poll a bit.

The thing is, however, that all polls are based on this representative weighing, period. And that weight is based on previous elections and turnouts. If you change that significantly - by, say, 30% or more, which would be the biggest change in history - you can have a lot of different results. 

So going back to that tool 538 published, there are some interesting and scary things there. Namely, one of the groups that turns out the least is white men (61%) and specifically non-college educated white people (55%). If you move up the non-college educated white men to 100%, you get a very, very narrow Clinton win. (272 EV). If you move up all white men, you get a slightly better win for Clinton (285). I don't know what happens if you move up white non-college men, but you get the point here - which is that when you add 30% more, there is absolutely no guarantee that those 30% more are going to be in the same proportions as they were before.

And furthermore, the non-registered voters will come more from groups of white men, hispanics, and Asian-Americans than they will African Americans and White women - simply because AA and women tend to have the highest voter turnout. And while Hispanic voting matters it isn't in the right 'places' to affect electoral college voting to any major degree save in Florida. 

So yeah, that's why I think that the polling might not represent what happens. There are a lot of good signs - particularly that in battleground  states the new voting registrations appear to favor Democrats - but it makes the polls potentially based on a lot of assumptions that really aren't fair.

I appreciate and laud your pessimism!

as you said, Hispanic and Asian voters (like Ross Perot voters) aren't distributed evenly amongst all states, but they have the most room for growth as 48% of eligible hispanics vote and 45% of eligible asians vote. Often unnoticed is that asians immigrants have outnumbered Latino immigrants for several years now and there a couple million more eligible Asian voters than 08. Increasing the turnout of these two demographics may not tilt swing states but might make a difference diwnballot and long term. Whites and blacks are the strongest turnout group with 65% and 67% respectively, note however that two thirds of whites lack a college degree and only 55% of said vote, so that 10% gap of 55%to 65% is entirely made up by the one third of whites with degrees voting in very high numbers, I think it's like 89%. Usually these white college voters break republican, and comprise much of the overall margin in a Romney Obama scenario. So that means that if whites with college degrees are not providing that margin, whites without college degrees need to make up that loss and also offset gains with minority groups. And whites without college degrees who don't usually vote are not monolithic, trump is not going to get all their votes, so they need really high margins to also offset the additional Clinton voters amongst their numbers.

but the scary thing is that such white voters are distributed in such a way as to maximally hurt Clinton in the rust belt and swing states, where they can play the most offense she has the least defense. Ultimately, such typically nonvoters are very well distributed to help trump if they do turnout as they simply have enough people in the right states to maximize their candidates chances. So in an atypical turnout election may basically come down to college whites providing just enough of a firewall to keep the noncollege whites from flipping every swing state. North Carolina might wind up being more safe than Wisconsin because the more diverse electorate gives Clinton more defense against a noncollege white turnout surge.

[Maithanet]

7 minutes ago, Dr. Pepper said:

Another day of super long lines in Georgia.  Drove by this morning, then around lunch, and just a few minutes ago.  This seems crazy!

Glad to hear it!  Or at least, I'm glad plenty of people are motivated to vote. I think that lines for voting should be a lot less common than they are in this country, but that's a separate issue. 

[Kalbear]

1 minute ago, Maithanet said:

Glad to hear it!  Or at least, I'm glad plenty of people are motivated to vote. I think that lines for voting should be a lot less common than they are in this country, but that's a separate issue. 

Again - Washington and Oregon have 100% vote by mail ballots, and they absolutely love it. Encourage your states to use it. It's cheaper, it results in fairly high turnout, and it's faster to do. It greatly encourages voter turnout amongst those who are busy. 

[Manhole Eunuchsbane]

8 minutes ago, Kalbear said:

Well, yes. All of this works because of who Trump is. 

She wouldn't have done it against Cruz - she would have made him look like a lizard man instead. 

She wouldn't have done it against Rubio - she would have made him look like a robot instead.

 And I'm sure she would've worked those angles well, but let's face it, Donald is especially vulnerable to her. His personality, lack of polish and inherent disrespect for women that he doesn't want to have sex with make him a perfect target for her. 

[lokisnow]

1 minute ago, Kalbear said:

Again - Washington and Oregon have 100% vote by mail ballots, and they absolutely love it. Encourage your states to use it. It's cheaper, it results in fairly high turnout, and it's faster to do. It greatly encourages voter turnout amongst those who are busy. 

I'm pretty amazed that high area low population  states like Wyoming, Montana Alaska Idaho don't do universal mail voting, it's a no brainer in terms of savings

[Kalbear]

Just now, Manhole Eunuchsbane said:

 And I'm sure she would've worked those angles well, but let's face it, Donald is especially vulnerable to her. His personality, lack of polish and inherent disrespect for women that he doesn't want to have sex with make him a perfect target for her. 

Yep. I am amused to think a few months ago that I was worried that Clinton wasn't nearly as good a foil as Sanders was for Trump, but I don't see how Sanders could have pulled off the types of egging on that Clinton has. Sanders getting mad at Trump would have been two old white men shouting over each other again and again; it's hard to imagine Sanders remaining calm in the face of Trump's lies over and over again. 

[lokisnow]

23 minutes ago, Kalbear said:

Well, it's closer to a jump of 67% to 80% - the eligible voter base is around 240 million right now. 

Apparently a very big part of that registration is heavy efforts to register voters, especially hispanic voters - and especially especially hispanic young voters. 27.9 million Hispanic Americans are now eligible to vote, which is 10 million more than it was in 2008. That alone is a pretty big jump.

I agree that it does seem like a really huge thing - but the ability to register in 2016 is significantly improved compared to 2008, as well as the ability to check your registration status. I'm also wary of dismissing actual data because it seems like it wouldn't be right. Both parties appear to be taking it seriously, and all indications - early voting, early polling, level of engagement in debates - seem to indicate that the number of registered voters is a lot higher. 

Before progressives and confederates started restricting the expanding franchise with voter registration in the late 1800s you had percentages in the 80s & 90s voting (of eligible voters).

as we are now lessening these restrictions I wouldn't be surprised to see percentages of voters rise.

[The guy from the Vale]

6 minutes ago, Manhole Eunuchsbane said:

 And I'm sure she would've worked those angles well, but let's face it, Donald is especially vulnerable to her. His personality, lack of polish and inherent disrespect for women that he doesn't want to have sex with make him a perfect target for her. 

You think Donald Trump treats women he does want to have sex with with respect either? Where are those sexual assault accusations coming from, then?

[Kalbear]

Sigh.

 

[Kalbear]

2 minutes ago, lokisnow said:

Before progressives and confederates started restricting the expanding franchise with voter registration in the late 1800s you had percentages in the 80s & 90s voting (of eligible voters).

as we are now lessening these restrictions I wouldn't be surprised to see percentages of voters rise.

That's interesting. Though another part of that was suffrage, of course. The graph of voting by population in the US was eye opening - basically like 5% of the people decided back in the day. 

[lokisnow]

24 minutes ago, Swordfish said:

Could this be affected by 'motor voter' auto registration though?

I know this is relatively new in Oregon, but I'm not sure it's widespread enough to impact the numbers that much.

No, only Oregon has it in place, CA and WV haven't yet activated their new motor voter laws.

fundamentally this is the ascendency of the echo boom now all being old enough to vote,  millenial voters now outnumber boomer voters, additionally there's like 800,000 new immigrant citizens every year also growing the pool.

democrats have also historically declined to focus on registering latinos and asians since their votes are mostly distributed to states that don't help democrats win new outcomes much like CA, but that seems to have changed in this election. 

Texas is a unique case, their new John crow laws restricting voter registration in the wake of 2013 gutting section 4 of the vra means democrats can do very little registration there, so this surge of voter registrations seems to be in spite of the second largest state massively restricting the possibility of such a surge. So these numbers are also even more impressive since Texas is contributing very few new voters in spite of having one of the largest available pools of new votes to tap.

[Manhole Eunuchsbane]

Just now, theguyfromtheVale said:

You think Donald Trump treats women he does want to have sex with with respect either? Where are those sexual assault accusations coming from, then?

No, but it's a different dynamic at work there. I suspect he might be more likely to be slightly more diplomatic in an exchange with a woman he finds to be attractive. He has some use for them, whereas I imagine that Hillary is just useless to him.

[Fez]

Kalbear,

While what you say is plausible, there are some problems with your analysis.

1) There hasn't been a significant surge in voter registrations among white voters, white working class voters, or new Republicans. This data is available and there are no states (or at least, no swing states, I don't know if anyone's bothered to look at safe states) where the demographics of registered voters is significantly different than where it was expected to be.

2) Polls don't just use demographic screens, they also ask questions about self-reported enthusiasm, likelihood to vote, attention paid to the election, etc. that are all pretty strongly correlated to whether people actually vote. And they don't show any indication that there are huge numbers of Republicans out there and ready to vote who don't already regularly vote.

3) In all the swing states with early voting and/or absentee voting that publish data, Trump is vastly underperforming Romney, except in Iowa. These voters may still show up on election day, but if there was a large contingent of uncounted Trump voters; at least some of them would be expected to have voted already. Instead, its looking entirely possible that Trump is going to significantly underperform the polls.

[The Anti-Targ]

2 hours ago, sologdin said:

a colossal listing, yes.

Wonder Woman endorses Hillary, Superman endorses Trump.