US House Redistricting: South Carolina (user search)
       |           

Welcome, Guest. Please login or register.
Did you miss your activation email?
April 28, 2024, 06:31:59 AM
News: Election Simulator 2.0 Released. Senate/Gubernatorial maps, proportional electoral votes, and more - Read more

  Talk Elections
  General Politics
  Political Geography & Demographics (Moderators: muon2, 100% pro-life no matter what)
  US House Redistricting: South Carolina (search mode)
Pages: [1]
Author Topic: US House Redistricting: South Carolina  (Read 19553 times)
minionofmidas
Lewis Trondheim
Atlas Institution
*****
Posts: 58,206
India


« on: January 10, 2011, 09:09:13 AM »



Northern end cropped.

Grey district is new. Lopping all of Horry off CD-1 was my starting point... after that I had to rummage around for white or at least >60% Black areas on the Black side of the current CD1/CD6 divide, but I found enough. All incumbents remain in their seat (if we count Duncan's phony registration address; he actually lives in Greenville.) This kinda forced a three-way split of Columbia:



I *think* all white-majority districts should "usually" be safe for Republicans. Race is CD1 69-24, CD2 65-27, CD3 77-18, CD4 70-20, CD5 71-24, CD6 38-58, CD7 66-27. If you know the state better than me (not that hard) and beg to disagree and have an idea how to rectify things, just say so.
Might be possible to draw an alternative map based on a Greenville-Spartanburg split, as the current district is getting quite urban in this iteration. Scott is from Charleston, so removing that end from CD1 is probably not an option. Wilson lives in the western suburbs of Columbia, so, again, yeah.

Logged
minionofmidas
Lewis Trondheim
Atlas Institution
*****
Posts: 58,206
India


« Reply #1 on: January 10, 2011, 10:25:06 AM »

Wilson will not like splitting up Lexington County.  He'd be vulnerable to a primary challenge if he loses half of his base county.  People here think the GOP will be able to get away with providing for only a single minority-majority district, and they really have nothing to lose by trying.  If they get overruled and have a court-drawn plan with two minority-majority districts, it certainly will be a 5 GOP 2 Dem delegation no matter how it is drawn.
The conventional wisdom on this board has been that South Carolina would have to draw two majority-minority districts that would elect Democrats, but there was an article in The State today that indicates that both Clyburn and the GOP are thinking that the General Assembly will be able to draw only one such district and carve things up so that the other six would elect Republicans.

(link)
Yeah, I remember that. It was actually part of the reason why I drew South Carolina. And after this, I think I'm fairly confident they'll get through with it. One might try looking what a 2 Black seats map would look like... I have a feeling it would have to be hideous.

As to the Wilson objection (though after his 2010 GE showing, the Assembly just might feel that getting rid of him would be a good thing)... well he didn't lose that much of Lexington. It would be possible to draw the other seats into Aiken instead for an even uglier map. He lost the white areas on the east side of Columbia though... if these were an indispensable part of his primary base then he has a problem. One that's insoluble in *this* setup - might require drawing the grey district all along the state line from Horry to Mulvaney's home in Lancaster, give Wilson a district that stretches north from Columbia, and have the southwestern seat be the open seat?
Logged
minionofmidas
Lewis Trondheim
Atlas Institution
*****
Posts: 58,206
India


« Reply #2 on: January 11, 2011, 11:20:59 AM »



Gives Joe Wilson a seat that will elect a Columbia Republican. As a special bonus, Jeff Duncan gets to choose where he really wants to live as both seats are free of other incumbents.
Logged
minionofmidas
Lewis Trondheim
Atlas Institution
*****
Posts: 58,206
India


« Reply #3 on: April 02, 2011, 04:07:00 AM »

I think you set up Wilson to lose a primary to Duncan.
Logged
minionofmidas
Lewis Trondheim
Atlas Institution
*****
Posts: 58,206
India


« Reply #4 on: May 31, 2011, 12:50:36 PM »

The VRA wasn't about redistricting at all (except by implication). It's all VRA-related case law.
Logged
minionofmidas
Lewis Trondheim
Atlas Institution
*****
Posts: 58,206
India


« Reply #5 on: June 13, 2011, 02:11:04 PM »

Here is a horrible gerrymander I've worked on, that does provide for two majority-minority districts.  While it can likely be improved on, it does show that it is doable.


CDPopDev%W All%B All%W 18+%B 18+%Rep%Dem
1661,272+50675.514.878.314.665.434.6
2661,383+61775.916.878.115.966.333.7
3660,894+12875.515.878.014.866.034.0
4661,498+73276.614.178.813.668.731.3
5660,016-75069.223.271.821.962.737.3
6660,204-56239.952.842.750.938.161.9
7660,097-66935.856.238.854.133.266.8

What is needed is ugly, but not anywhere near that ugly.



Columbia seat 52.4-40.3 Black total, 50.3-43.2 VAP, 62.5 Obama
Charleston seat 52.3-41.0 Black total, 50.3-43.9 VAP, 59.8 Obama - and thus actually the most marginal district in this map, as the southwestern seat is 59.9 McCain. The Blackest of the white seats is the eastern one, at 20.1.
Logged
minionofmidas
Lewis Trondheim
Atlas Institution
*****
Posts: 58,206
India


« Reply #6 on: June 13, 2011, 02:28:30 PM »

I was just idly wondering... it obviously wouldn't fly in real life... but since our argument here is "SC is over 2/7th Black and thus must draw two Black-opportunity districts if it can also be shown to have the areas necessary to drawing two majority-Black VAP districts to at least arguably represent reasonable communities of interest"... and any two Black districts map will rely on one district of Charleston Blacks plus rural/smalltown Blacks in the state's south-central portion and one of Columbia Blacks plus rural/smalltown Blacks in the state's central portion... wouldn't it actually make far more sense, CoI wise, to draw one wholly rural/smalltown Black seat and one urban Black seat that's in two noncontiguous parts, half in Columbia and half in Charleston? Grin
Logged
minionofmidas
Lewis Trondheim
Atlas Institution
*****
Posts: 58,206
India


« Reply #7 on: June 14, 2011, 12:06:50 PM »

Not really. The divide between the low country and the rest of the state is far more significant than urban/rural as far as CoI.
Is that why you drew the areas by the Savannah River into the Columbia seat? It makes the map very ugly. Where exactly would you say the Low Country ends?
Logged
minionofmidas
Lewis Trondheim
Atlas Institution
*****
Posts: 58,206
India


« Reply #8 on: June 15, 2011, 11:43:47 AM »

That's right, Democrats took about 3/7th of the SC vote, so they should by right get about 3/7th of seats. Right? Tongue
Logged
minionofmidas
Lewis Trondheim
Atlas Institution
*****
Posts: 58,206
India


« Reply #9 on: June 15, 2011, 11:45:08 AM »

According to the newly added partisan figures in DRA, my 1st is 58.6 McCain.  The app's numbers are (on average) 2.7% redder than the real numbers due to the lack of absentee ballots, so figure about 56% for the true total.  It should be safe for Tim Scott.  
Ah, right, that explains my partisan figures per district. They did seem a bit off...
Logged
minionofmidas
Lewis Trondheim
Atlas Institution
*****
Posts: 58,206
India


« Reply #10 on: June 15, 2011, 12:21:13 PM »

Just for hilarity's sake, I'll attempt to draw one (correcting for the figures issue, so I'll be content with McCain by less than 5.)
Logged
minionofmidas
Lewis Trondheim
Atlas Institution
*****
Posts: 58,206
India


« Reply #11 on: June 15, 2011, 12:57:37 PM »



Didn't draw the R parts.
Green district is 50.5-48.4 McCain on app figures, which translates to approximately 51.0-47.9 Obama in real life. Yellow district is 49.7-49.1 Obama even in the app. Both are 56% White (total, 59% in VAP) - you'll notice yellow has the white dem bits of Charleston. As an added bonus, the grey district is over 60% Obama (59.8% in the app) and has a barest of Black pluralities, though not in VAP.
Obviously it's possible - probably not particularly hard - to draw four rl Obama districts. Though they wouldn't look as pretty, I reckon.

The point here is: the uglyness of the above two black democratic seats maps is due only to their having to reach an unnaturally high threshold of Black population, thanks to the way Republicans on the SC have interpreted the VRA. Two Black Democratic (defining a white man whose primary and general electorate is Black-dominated as a Black Democrat) representatives from SC would occur naturally on any not gerrymandered map, unless all the White Liberals decide to vote for the Republican over the Black.
(The three seats wouldn't - fptp even with fair districts doesn't provide for fair representation of political minorities unless they're highly concentrated, and my split of Charleston here is quite unnatural.)

Logged
minionofmidas
Lewis Trondheim
Atlas Institution
*****
Posts: 58,206
India


« Reply #12 on: June 15, 2011, 02:28:22 PM »

That's right, Democrats took about 3/7th of the SC vote, so they should by right get about 3/7th of seats. Right? Tongue

A.  Dems didn't get 3/7th of the vote.
Sure did. More than that. (Okay, so I'm specifically talking of Barack Obama in 2008. Not going to look up the congressional total for each and every year. Especially seeing as the state had 1.5 non-joke contests in both 2008 and 2010.)
Quote
You must be logged in to read this quote.
Of course not. But you know what it was in reference to (though I suppose you weren't being serious in using that as a defence of your own NY gerrymander either.)

1) I see the VRA is now being reintrepreted to protect White liberals. In case you have forgotten, White liberals have never been subjected to the types of acts that the VRA was suppose to remedy.

2)  This presumes a closed primary system in which White voters whom typical vote Republican don't nominate the White Democrat in the primary.
Lolwut?

For your reading comprehension:

Obviously this map does not conform to the VRA in any way - if anything, it might be conceivable if the VRA were overturned and Dems had control of redistricting (though they'd make the southern seats more erose to shore them up a bit). And how likely is that, exactly?
The part about
Quote
You must be logged in to read this quote.
refers to a map not posted - a cleaner-looking version of the two black seats map. These would still be Black enough to elect what I called Black Democrats (though not Bobby Rush or Cynthia McKinney, obviously. Oh, and on rereading, my wording was a bit unclear: I meant to include Steve Cohen or even John Barrow with the Black Democrats because they have much the same electoral coalition as Black Democrats. Basically, I was saying the districts would elect a Sanford Bishop / GK Butterfield / Artur Davis type or possibly a John Barrow type.). They wouldn't be over 50% Black VAP is all. The 50% thingy in case law creeps in due to the need to demonstrate to a court that SC (arguably; according to current case law) must draw the two seats. If South Carolina had a commission for these things that was forced to interpret the VRA in good faith, it would draw two such seats.

The gist of the lawsuit, he said, would be to encourage the U.S. Justice Department to approve a S.C. plan that moves African-American voters out of majority-black districts to other districts, giving them more sway over who is elected.
LOL, what a bunch of hacks! They say the opposite in so many other states, and for that matter, in the same state for the Congressional map.
So they're saying Blacks are overpacked - seems to me like that's exactly what they're saying about the Congressional map in the same state. (Guilty as charged on being a bunch of hacks and saying the opposite in some other states, of course!)



And now, for comic relief again. An attempt at a "colorblind" but otherwise "fair" map of South Carolina - something a British boundary commission might have come up with (except with very little leeway for deviation, as common in US congressional districts. Maximum deviation here is 177). And like a British boundary commission's map, this is a first draft - details would be likely to change after a hearing with people on the ground.



Charleston (blue) Charleston, Berkeley and Dorchester Counties - the three counties to be partly in the Charleston built-up area - are barely too large for one constituency. Using a uniform correction, McCain barely edged Obama here. However, as it's almost exactly made from three counties, we can just use the actual figures, and the lead was 1.odd% in real life. Guess the gap between absentee and in-person votes was smaller in Charleston than elsewhere in the state.

South Carolina South West (green) takes in the remainder of the low country and some adjoining country to the north. Uniform correction makes this McCain by 2.3 or so.

Pee Dee (purple) Based on Horry, with its (blacker and less Republican) hinterland. McCain by 5 or so.

Columbia (red) Richland and Lexington are too small for one district, but not by much. Obama by 3 or so.

Rock Hill & South Carolina North East (yellow) Bit unfortunate design, but not much choice (see also next para.) McCain by 4 or so.

Spartanburg & Greenville South (cyan) Doesn't actually take in any part of Greenville proper (if precinct names can be trusted). Still, had to split the Greenville metro. The alternatives like drawing a donut around Greenville County or a donut around just the urban cores of Greenville and Spartanburg or including Anderson in the green district and anything at all removed from the Savannah in other districts are all worse. Very safe Republican (almost 60% McCain).

Greenville North & Anderson (grey) Even safer - 64.odd% McCain.

Now... the issue with this map (which I drew without looking at the race or election figures) is... every seat is at least 56% White. That's because the semi-solid belt of rural Black settlement is split between four constituencies. Which obviously doesn't fly if race is any consideration at all - nvm the only specially legally protected consideration as in the US right now.
Oh, and instead of a 5-2 map, I came out with more of a 3-0-4 map. Lol.
Logged
minionofmidas
Lewis Trondheim
Atlas Institution
*****
Posts: 58,206
India


« Reply #13 on: June 15, 2011, 02:33:06 PM »

SC as a whole, bear in mind that the underlying assumption is that the app has a structural GOP bias of about 2,5% , as someone said in this very thread. It's entirely possible that I'm mistaken.
You can easily test that claim with the app (just look at the blank map, click on unassigned and let it show you the election stats). I did. Smiley
Logged
minionofmidas
Lewis Trondheim
Atlas Institution
*****
Posts: 58,206
India


« Reply #14 on: June 18, 2011, 05:48:28 AM »
« Edited: June 18, 2011, 05:50:46 AM by Jakob Bronsky »

Here's a version of the last map where race was taken into account as one of many factors. Max deviation 366. Purple district is majority Black, plurality Black VAP. Getting it to majority Black VAP requires really ugly stuff on the eastern edge and/or entering the Charleston or Columbia built-up areas.



Politically speaking it's 4-1-2, with the Charleston (& points west) marginally Republican and the Columbia seat marginally Democratic.
Logged
minionofmidas
Lewis Trondheim
Atlas Institution
*****
Posts: 58,206
India


« Reply #15 on: June 18, 2011, 06:01:49 AM »

And because I was unhappy with Kershaw County in that one...



Aiken town has been racially split. Still doesn't take purple over 50% Black VAP. Maximum Deviation has been gotten down to 97, though.
Logged
minionofmidas
Lewis Trondheim
Atlas Institution
*****
Posts: 58,206
India


« Reply #16 on: June 28, 2011, 05:29:25 AM »

Ooh, pretty good. That's a map you could take the state to court with.
Logged
minionofmidas
Lewis Trondheim
Atlas Institution
*****
Posts: 58,206
India


« Reply #17 on: July 27, 2011, 03:33:59 AM »

You have a map?
Logged
Pages: [1]  
Jump to:  


Login with username, password and session length

Terms of Service - DMCA Agent and Policy - Privacy Policy and Cookies

Powered by SMF 1.1.21 | SMF © 2015, Simple Machines

Page created in 0.1 seconds with 12 queries.