17
   

Get yer polls, bets, numbers & pretty graphs! Elections 2008

 
 
Cycloptichorn
 
  1  
Reply Sun 10 Feb, 2008 10:51 am
sozobe wrote:
CNN's current delegate counts:

Obama (pledged only) = 908

Hillary (pledged only) = 877

Obama (pledged + superdelegates) = 1,039

Hillary (pledged + superdelegates) = 1,100


So odd to me. B/C cnn still isn't counting all the delegates from CO and Al.

Cycloptichorn
0 Replies
 
nimh
 
  1  
Reply Sun 10 Feb, 2008 10:59 am
The Obama site's results center (thanks Butrflynet!) has:

Obama 910
Clinton 882

(With no estimates for superdelegates)

Pretty map, too.
0 Replies
 
realjohnboy
 
  1  
Reply Sun 10 Feb, 2008 12:34 pm
grr. It is a windy day in Charlottesville. Gusts reported up to 60 mph coming over the mountains. Power keeps flickering, wiping out stuff I have written but not sent.
Take 4 on...
VIRGINIA est omnis divisa in partes quattor.

I-64 starts in the Tidewater area of Virginia (Norfolk, Hampton Roads, Virginia Beach). It heads west, passing just to the south of Richmond and Charlottesville. It crosses the mountains into the Shenandoah Valley and on to WVA.
I-95 comes out of NC, cuts through Richmond heading north, getting wider and wider as it passes through suburban Virginia on its way to DC.
Virginia gets divided into four sectors.

Predicting the results of the Repub primary in Virginia is easy: McCain in a total romp over Huckabee. The NE and NW quadrants feel no connection to Huckabee. Folks there are not at all southern. They, unlike me, have no discernible accent. They have no problem with spending an hour driving to work in DC. Their idea of agriculture is listening to their yardboy tell them where to move the sprinkler after he goes home for the day. Solidly McCain.
The SE sector is for the Repubs military, military and more military. McCain rules down there.
That leaves only the SW area. Rural farmers and blue-collar workers looking for work. It's tough down there. The drought-line is moving north, perhaps even crossing the NC border next season. The mills have closed.
The late Lee Atwater summed it up: "It's the economy, stupid."
Huckabee may show up well in the SW, but I have him losing that region also. Folks down there may like Huckabee's religion thing, but it's the economy and the fact that their sons and daughters and grandsons and granddaughters are the ones joining the military for whatever reason, patriotism or economic need.

The Dem primary in Virginia is, in my mind, potentially going to be close.
The Hampton Roads and Richmond area have a lot of black voters. I sense that black voters in the Dem party, particularly older women, are starting to move away from Hillary. SE quadrant goes to Obama.

The NE and NW could go either way. Generalizing, folks up there are well educated and somewhat affluent. I could see it breaking either way.
The SW area has a few pockets of Dems but not enough to be significant.

Virginia is a "Yall Come" primary. I wish I had thought of that phrase. On Tuesday we will walk into our precincts, Dems, Repubs and Indys, and ask for either a Dem or Repub ballot. I think that there will be a certain segment of Indys who will ask for a Repub ballot in order to vote for McCain.

When the dust settles, I have Obama by 5%.

The thing that I will be watching is turnout amongst folks voting for Dems and Repubs. I, a liberal Dem, have been in the wilderness here in VA. The last time we successfully voted for a Dem was in 1964 (LBJ). VA is getting bluer. This year could be the time when, after 40 years in the desert, my vote counts.

Caveat emptor.
0 Replies
 
realjohnboy
 
  1  
Reply Sun 10 Feb, 2008 01:38 pm
After that, I looked at the paper. According to a poll by Mason-Dixon taken over Thursday/Friday. (400 self described Dems and 400 self described Repubs and with a 5% MOE):
McCain over Huckabee by 55-27%. Paul with 5%
Obama over Clinton by 53-37%.
I am skeptical on the Dem side.
0 Replies
 
Butrflynet
 
  1  
Reply Sun 10 Feb, 2008 02:45 pm
Realjohnboy,

Are there any oddities about the ballots that we should inform voters of when making calls? Things such as needing to mark an extra bubble at the top of the page to indicate your party affiliation or needing to be pre-registered to vote in a party, or the phrase an independent or didn't declare party voter needs to mutter to receive the proper ballot?

And your vote will definitely count no matter the outcome of who is the "winner." Your vote might help ensure a delegate from your congressional district is allocated to Obama's campaign.
0 Replies
 
sozobe
 
  1  
Reply Sun 10 Feb, 2008 02:50 pm
That was cool, RJB, thanks for your take.
0 Replies
 
Cycloptichorn
 
  1  
Reply Sun 10 Feb, 2008 02:53 pm
realjohnboy wrote:
After that, I looked at the paper. According to a poll by Mason-Dixon taken over Thursday/Friday. (400 self described Dems and 400 self described Repubs and with a 5% MOE):
McCain over Huckabee by 55-27%. Paul with 5%
Obama over Clinton by 53-37%.
I am skeptical on the Dem side.


Thanks rjb.

Survey USA, who has been dead on this year, has Obama by 20. So...

Cycloptichorn
0 Replies
 
realjohnboy
 
  1  
Reply Sun 10 Feb, 2008 03:07 pm
Not that I am aware of. If there is any kind of chicanery, I will be pissed and will let you know.
I hope that election officials are prepared for a huge turnout and have planned accordingly.
0 Replies
 
blatham
 
  1  
Reply Sun 10 Feb, 2008 03:20 pm
Lovely account, rjb.
0 Replies
 
realjohnboy
 
  1  
Reply Sun 10 Feb, 2008 03:52 pm
Cycloptichorn wrote:


Survey USA, who has been dead on this year, has Obama by 20. So...

Cycloptichorn


Yeah. Northern Virginia perplexes me. I don't know where their heads are at, so I gave half to Clinton and half to Obama. Survey USA seems to detect a strong Obama lead up there.
I still see McCain siphoning off votes.
0 Replies
 
nimh
 
  1  
Reply Sun 10 Feb, 2008 06:21 pm
sozobe wrote:
Interesting analysis of the race issue from a Daily Dish reader:

Quote:
In states where the black population is less than 5%, Obama has a record of 7 wins, 2 losses and 1 undecided (NM). In states where the black population is 20% or higher, he is undefeated at 4-0. However, in states that are between 5-20% black, his record is a fairly dismal 4-10 (with one of those victories being Illinois).

The theory here is that Obama does well where the black population is so low that identity politics isn't an issue. And, he does well where the black share of Democratic primary voters is so high that he needs few white voters to carry the day. He has the hardest time in states that are black enough to have some racialized politics, but without enough black voters to completely tip the scales.

This corresponds to the long held observation that black candidates in general do best in either fairly non-black environments or in heavy black environments but struggle with the in between, where white majority fears that they will be the tool of an aggrieved constituency. (Some, of course, have broken that mold like Doug Wilder, and Carol Mosely-Braun and Obama himself).

For Obama's upcoming contests, this bodes well for Maryland, D.C., Virginia, Nebraska and Vermont and maybe Rhode Island, but not so well for Ohio and Texas.


That was an interesting analysis, and it comes back in that piece about Virginia I quoted above:


Quote:
Obama does disproportionately well in [..] states where African Americans make up more than a quarter of the population (Louisiana, 31.7 percent; Maryland, 29.5 percent), and in lily-white northern-tier states (Nebraska, Maine, Wisconsin, and to a slightly lesser extent Washington). In fact, somewhat astonishingly, among states in which whites make up 89 percent or more of the population, Obama has won seven of eight contests so far, losing only narrowly in New Hampshire.

The problem, like that DD commenter said, is in states in between - but that would include Virginia.

After you posted that comment, I started playing around with the numbers on those. And there does seem to be pattern, although it's a relatively fuzzy one. Wait, I'll put that in a separate post..
0 Replies
 
nimh
 
  1  
Reply Sun 10 Feb, 2008 07:50 pm
OK, so what I did first was (again) list the states that have voted so far in order of how well Obama did among white voters.

Then I looked up the census data from 2000 about the racial demographic make-up of each state. How large was the African-American population?

The intention: to examine the proposed hypothesis that Obama's actually had an easier time winning white votes in "lily-white" states than in states "that are black enough to have some racialized politics".


http://img135.imageshack.us/img135/1668/whitevotesforobamavssimn6.png

Of course not all primaries are equal, which complicates direct comparisons. New York and Arkansas both directly represent home terrain for Hillary, while Obama has the home advantage in Illinois. And in Iowa, New Hampshire, Nevada, South Carolina and Florida, other candidates still took a fair chunk of the white vote. Hence why those states are listed in grey rather than black or orange rather than red (red and orange denote Southern states).

Still, the contrasts remain striking, with Obama getting:

  • just around a quarter of the white vote in most of the Southern states;
  • around a third in the earliest primary states, where he still had to share the votes with Edwards and others too, as well as in Hillary's backyard (NY, NJ);
  • and anywhere between 37-57% in the rest of America, as well as the Southern exceptions Georgia and Missouri.
Now, to the right, there's the data on the size of the state's African-American population. So what can we see? Is there a correlation?

First things first: there is obviously no 1:1 relation. Georgia, Delaware and Illinois have a sizable black population, yet Obama did well among whites there. Oklahoma has a small black population, yet Obama did badly.

Nevertheless, there does seem to be a rough pattern. Four out of five states where Obama did best among whites have small black populations, while the five states he did worst in among whites all have a black population of over 15%.

But is the size of the minority even a determining factor at all here, considering that there are other, possibly more obvious factors? All those five states he did worst in are in the South, for one; well, all seven states he did worst among whites in are. Regional cultural/historical elements are probably at play.

Yet the argument was merely that the more "racialised" local politics is, with long-standing black and white populations, the harder it might be for a black politician to reach out to white voters. And well, Alabama and Tennessee likely have pretty racialised local politics. In "lilywhite states," you might face just as much prejudice, but perhaps not the entrenchment in camps.

On that "lilywhite" note though: that's obviously not the right label. Some of the states that Obama did relatively well among whites in are very multicultural: California, New Mexico. Big Hispanic/Latino populations. So I tried doing a separate comparison, not juxtaposing the size of the state's black population with Obama's score among white voters, like above, but the size of the state's minority population overall. Interestingly enough, that made all semblance of a correlation disappear. While the above juxtaposition at least suggests some kind of relation, there was just randomness if you were looking at the proportion of minority populations overall.
0 Replies
 
nimh
 
  1  
Reply Sun 10 Feb, 2008 07:57 pm
To get a clearer image, I created a graph (of course):


http://img183.imageshack.us/img183/8266/whiteobamavotevsblackpsd0.png


How does this work? Every dot represents a state. States with a status aparte of some sort (primary was held when Edwards and others were still in the race, the state is either Hillary's or Obama's home state) are coloured orange, the others red.

The further to the right the dot is, the more whites voted for Obama in that state. The further up the dot is, the larger the African-American population in the state is.

See a pattern? Well, it's certainly not a very clear one. At first blush, these dots are just all over the place.

I added a trendline though (polynomial, order: 2), and it suggests that there is a correlation of sorts. And if you squint, you do notice that both the bottom left corner and the top right corner are quite empty.

The bottom left corner would have states with a small black population and little Obama support among whites. There isnt much there except, perhaps, Oklahoma.

The top right corner would have states with a large black population where Obama also had a big appeal to white voters. Georgia is there, and nothing much else.

So here's the same graph, but without the gridlines, and instead, with two sectors lined out that most of the states fall into:


http://img442.imageshack.us/img442/5223/whiteobamavotevsblackp2yz1.png


The section at the top left has seven states in it. All states where blacks make up 14% or more of the population; and where Obama got less than a third of the white vote. It's basically got six Southern states, plus, on the very edge, New Jersey.

The section in the bottom right has 10 states in it. All states where African-Americans make up less than 12% of the population, and where Obama got at least a third of the white vote. Or if you ignore the earliest primaries, where the vote was still shared with Edwards etc as well, at least 38%.

This group has five Western states (CA, NV, UT, AZ, NM), three states in New England (NH, MA and CT), plus Iowa and Missouri, both on the edges.

Five states are not part of either section. Oklahoma, as mentioned above, is an overwhelmingly white state without much support for Obama. In Illinois, Georgia and to a lesser extent Delaware, Obama succeeded in rallying both a large(ish) black population and much of the local white vote.
0 Replies
 
nimh
 
  1  
Reply Sun 10 Feb, 2008 08:07 pm
So what does it all mean? Hell if I know. :wink:

The thesis was that Obama actually did very well "in lily-white northern-tier states," while having "the hardest time in states that are black enough to have some racialized politics, but without enough black voters to completely tip the scales." That thesis needs to be extended.

For one, he does no worse among whites in states that are not lily white at all, but where the population is a mix of whites and Hispanics with a relatively small black minority, than in actually "lily-white" states.

Secondly, the analysis focused on potential problems in states with mid-sized black populations. For obvious practical reasons, since the African-American population in states like Alabama and South Carolina is large enough to tip the victory to him anyhow even if he doesnt do so well among whites, those are just not as much a concern.

But it's possibly a concern in general terms. How does he do among white voters in states with a large black population as opposed to states with a mid-sized black population and states with a small black population?

The short answer is: too few datapoints. States with a small black population (<10%) are overwhelmingly in the 30-60% range of white support for Obama, while states with a midsize (10-20%) black population are overwhelmingly in the 10-40% range. Thats a clear contrast. As for the four states with (much) larger black populations, three of them are also clearly on the meagre end (20-30% white support for Obama). But Georgia is the exception; plus, of those three, how many more white voters might Obama have gotten in South Carolina if Edwards had already been out of the race? Remember, Edwards and others got 40% of the white vote there...

So there seems to be a clear contrast between states with a small black minority and those with a significant one, but the problem doesnt, at first blush, become bigger if the black minority is larger still.

But there's also no sign to suggest it becomes better again in such states -- that the problem, so to say, peaks in the "borderland" zone where a still-dominant white majority fears "an aggrieved minority constituency," but subsides again once the two groups are numerically on an equal footing. Whereas there could have been some ground to speculate on that. To go on something of a tangent, in Holland for example, the vote for xenophobic parties has sometimes peaked in the suburbs where immigrants are still a small, but growing group, and actually became somewhat smaller again in the inner cities where immigrants already long make up a big chunk of the residents.

A depressing analysis really, altogether. You'd hope that the more people are familiar with the "Other" -- the more people encounter people of a different race / etc on a day-to-day basis -- the more open they would be for voting for one of them. Whereas this suggests the opposite. Although it's all a very thin sample of course. I dont know.

I dont claim to be an expert on any of this. In the context of Eastern Europe, when I was studying the political mobilisation of minority groups as well as against them, I encountered enough reasons to speculate that such mobilisation occurs especially in times of flux and change and unstable power relations between the groups - one is growing, the other decreasing, societal roles are being redefined, both social advancement and setbacks seem within reach. Whereas if a minority is either mostly absent or perennially marginalised, or constitutes a large group with a stable position, there is less mobilisation on the basis of ethnic identity.

How any of that applies to the US, you tell me. Whether anything in these graphs indicate anything somehow related to it, you tell me. I also have no statistical schooling whatsoever, so what any of the correlations or seeming correlations above mean or whether they mean anything, again, you tell me. But I definitely find it very interesting.
0 Replies
 
nimh
 
  1  
Reply Sun 10 Feb, 2008 09:44 pm
Virginia polls... on the Democrats

...are looking very, very good for Obama.

InsiderAdvantage, 2/7:

Obama 52
Clinton 37

"In the survey Obama led among African-American voters, while Clinton and Obama were evenly splitting the white vote".

Rasmussen, 2/6-7

Obama 55
Clinton 37

"Obama leads among white men while Clinton leads among white women. Overall, the two contenders are even among white voters while Obama leads 72% to 22% among African-Americans."

MSNBC/McClatchy/Mason-Dixon, 2/7-8

Obama 53
Clinton 37

"Blacks support Obama over Clinton, 82 percent to 9 percent.
Obama leads in all regions of the state, except the rural south and west, where Clinton has a negligible advantage. His strongest region, at 62 percent, is Hampton Roads, which has a sizable African-American population."

Survey USA, 2/7-8:

Obama 59
Clinton 39

Obama leads by 31 among men, 11 among women; leads by 37 among youngest voters, trails by 5 among oldest; splits the white vote with Hillary at 49% each, while getting 87% of the black vote and 39% of the small Hispanic vote (6% of the total).

He leads by 19 among Democrats and 29 among Indies; by 18 among moderates and 28 among liberals; and by 35 among those who regularly attend a religious service, while he trails by 2 among those who almost never do.

Obama leads by 46 in the Southeast, by 36 in Central-Virginia, and by 9 in the Northeast, while trailing Hillary by 6 in Shenandoah.

He leads by 39 among those who cite Iraq as most important issue (17% of all respondents); by 11 among those who cite the economy (44% of all); and by 9 among those who name health care (14% of all).

------------------

In short, Virginia is obviously a very different story from states like Tennessee or Missouri.

A couple of posts back, I quoted from an article by Josh Patashnik stipulating that Virginia was hardly necessarily Obama country. He wrote:

Quote:
It's a border state, and only 19.9 percent black. Its closest demographic parallels among states that have voted so far are Tennessee, where Obama lost by 13 points, and Missouri, where he won by only ten thousand votes. [..]

[Obama would] need about 37 percent of the white vote to win the state. That's certainly achievable, given his totals elsewhere. But it's no sure thing. In [..] Southern states his totals have ranged from to 25 percent (Alabama) to 43 percent (Georgia).

Is there reason to believe that Virginia's white Democratic electorate will be more predisposed to vote for Obama than it was in Missouri and (especially) Tennessee? Yes, due to the growing strength of upscale, educated liberals in the Washington suburbs. But while this difference is real, it isn't overwhelming.

Well, according to the polls above, the difference is quite overwhelming. SUSA, Rasmussen and InsiderAdvantage all agree that at the moment, Obama and Hillary are evenly splitting the white vote in Virginia.

Now take a look back at the table I posted some 15 pages ago listing by state how whites divided their support over Obama and Hillary. If whites in Virginia split evenly, that would instantly put the state in the top 6 results for Obama among whites in the whole country! Tennessee this is not.
0 Replies
 
Cycloptichorn
 
  1  
Reply Mon 11 Feb, 2008 02:26 am
Chris Bowers at OpenLeft has been doing a running pledged delegate count.

http://www.openleft.com/showDiary.do?diaryId=3819

Too big to repost here, but here's the wrapup -

Clinton - 934
Obama - 1020

How is this number determined?

Quote:

Notes

1. Delegate counts are derived from the most recent numbers posted by one of CNN, MSNBC, or CBS news. The lone exception is in Colorado, where I have used MetaData's projections.
2. Even though counting continues, I am projecting Hillary Clinton as the winner of New Mexico, and allocating her the one remaining delegate.
3. The 49 remaining delegates from states with completed results are as follows: John Edwards (26), Washington (20), Alabama (1), New York (1), and Tennessee (1).
4. Delegate counts for caucus states might alter following state conventions, which take place variously from March through May.
5. The Michigan Democratic Party has claimed it will send a pledged delegate count of 73 Clinton, 55 uncommitted to the national convention. The DNC has ruled that Michigan has no pledged delegates. This conflict will be worked out by the DNC credentials committee, sometime in June or July. For more on the DNC credentials committee, click here.
6. The Florida Democratic Party has claimed that it will send a pledged delegate count of 95 Clinton, 63 Obama, and 27 Edwards to the national convention. The DNC has ruled that Florida has no pledged delegates. This conflict will be worked out by the DNC credentials committee, sometime in June or July.
7. There are 796 unlpledged, or "super" delegates, to the DNC convention. Historically, they have lined up behind the pledged delegate leader. In the event that there is no pledged delegate leader in June, they will come into play. In that "tiebreaker" event, Clinton currently holds a substantial advantage, 224-126.


Cycloptichorn
0 Replies
 
sozobe
 
  1  
Reply Mon 11 Feb, 2008 09:53 am
This is chock-full of interesting stuff -- I'm still only about halfway through:

http://poblano.dailykos.com/storyonly/2008/2/11/1165/53154/247/454354
0 Replies
 
Cycloptichorn
 
  1  
Reply Mon 11 Feb, 2008 01:08 pm
Now with re-sizing.

http://farm3.static.flickr.com/2228/2258699048_697bb71622.jpg

Still just a matter of time till those two lines intersect. Obama's slope has not diminished, though Hillary has bounced a bit as well.

Cycloptichorn
0 Replies
 
Cycloptichorn
 
  1  
Reply Mon 11 Feb, 2008 01:40 pm
Gallup dailies.

http://media.gallup.com/poll/graphs/021108DailyUpdateGraph1.gif

Cycloptichorn
0 Replies
 
Butrflynet
 
  1  
Reply Mon 11 Feb, 2008 02:13 pm
Wonder why there is such a sharp difference in polling trends between the two graphs....
0 Replies
 
 

Related Topics

 
Copyright © 2024 MadLab, LLC :: Terms of Service :: Privacy Policy :: Page generated in 0.07 seconds on 11/15/2024 at 05:46:13