Dave's Redistricting App

[Bo]

If the courts draw the lines in Mass, because in 2010 the Pubbies get more than one third of the seats in one of the houses of the legislature, and a GOP governor is elected, causing a deadlock, what would a non partisan map look like that comports with the law, and would that cause any seat to have a GOP lean, or close to it?  Just curious.

So here is one take on a court-drawn map. I assumed that the special master appointed by the court would start by keeping district cores intact, while eliminating the more egregious gerrymander shapes. MA-1 adds Springfield, since adding Worcester creates a strange loop from Worcester across the northern edge to the Berkshires. That puts Worcester in MA-2, and a handful of towns are left over in Worcester County from the two western CDs.

MA-3 AND MA-4 had the least identifiable cores since they were the must strung out. Since MA-3 lost Worcester, I shifted MA-3 south to New Bedford, including almost all of Bristol County. The new MA-4 essentially replaces the current MA-10 in Plymouth, the Cape and Islands.

In the northeast, MA-6 adds Revere and Lawrence to bring up its population and eliminate the long finger that sticks out to Bedford. That moves MA-5 south to Marlborough and the other adjacent "-boroughs." MA-7 pushes south to gain Somerville and Cambridge and west out to Concord. MA-8 maintains its status as a 50-50 white/minority district by linking south to Brockton. That leaves MA-9 to shift west and north, picking up Allston/Brighton, Back Bay, and West Roxbury in Boston, and towns from Newton out to Framingham.



I can now apply the election data from the 2006 Governor's race, using just the R and D numbers, without the other two candidates. Both MA-4 and MA-5 in the map above could be interesting to the GOP. The 2006 two-party split in MA-4 favors the Dems 52.4 - 47.6, and in MA-5 it would favor the Dems 52.8 - 47.2.

Do you think you can calculate how many districts in this map Scott Brown wins?

[muon2]

If the courts draw the lines in Mass, because in 2010 the Pubbies get more than one third of the seats in one of the houses of the legislature, and a GOP governor is elected, causing a deadlock, what would a non partisan map look like that comports with the law, and would that cause any seat to have a GOP lean, or close to it?  Just curious.

So here is one take on a court-drawn map. I assumed that the special master appointed by the court would start by keeping district cores intact, while eliminating the more egregious gerrymander shapes. MA-1 adds Springfield, since adding Worcester creates a strange loop from Worcester across the northern edge to the Berkshires. That puts Worcester in MA-2, and a handful of towns are left over in Worcester County from the two western CDs.

MA-3 AND MA-4 had the least identifiable cores since they were the must strung out. Since MA-3 lost Worcester, I shifted MA-3 south to New Bedford, including almost all of Bristol County. The new MA-4 essentially replaces the current MA-10 in Plymouth, the Cape and Islands.

In the northeast, MA-6 adds Revere and Lawrence to bring up its population and eliminate the long finger that sticks out to Bedford. That moves MA-5 south to Marlborough and the other adjacent "-boroughs." MA-7 pushes south to gain Somerville and Cambridge and west out to Concord. MA-8 maintains its status as a 50-50 white/minority district by linking south to Brockton. That leaves MA-9 to shift west and north, picking up Allston/Brighton, Back Bay, and West Roxbury in Boston, and towns from Newton out to Framingham.



I can now apply the election data from the 2006 Governor's race, using just the R and D numbers, without the other two candidates. Both MA-4 and MA-5 in the map above could be interesting to the GOP. The 2006 two-party split in MA-4 favors the Dems 52.4 - 47.6, and in MA-5 it would favor the Dems 52.8 - 47.2.

Do you think you can calculate how many districts in this map Scott Brown wins?

I haven't run the details, but it looks like Brown wins 2, 3, 4, 5, 6 and Coakley wins 1, 7, 8 and 9. That would be 55-44 and is a good match to the vote split in the race. When the districts split at the same percentage as the statewide election, it's considered to be one sign of a fair map.

[jimrtex]

I can now apply the election data from the 2006 Governor's race, using just the R and D numbers, without the other two candidates. Both MA-4 and MA-5 in the map above could be interesting to the GOP. The 2006 two-party split in MA-4 favors the Dems 52.4 - 47.6, and in MA-5 it would favor the Dems 52.8 - 47.2.

Do you think you can calculate how many districts in this map Scott Brown wins?

Maybe 5.

NYT election results

Coakley would win the 3 inner districts around Boston (6, 7, and 9), and 1.

You might be able to get it up to 6, if you split 7&9 more on north/south basis and really pack the eastern district, and then split CD2 and 5 into 3 districts and crack the remnant of 7&9.

You might be able to bring CD4 north, to get Quincy and Brockton, and shift CD3 eastward.  The northern part of CD3 could then be used to bulk up the new districts.  CD6 could probably be adjusted a bit to take in some suburbs in the eastern part of CD7.

In a more conventional election this might go back9-0 Democrat.  

Some pretty big differential turnouts vs 2008 presidential election:

Metheun 67%, Lawrence 50%

Westfield 71%, Springfield 55%.

Boston 67%, Dedham 84%, Chelsea 52%.

New Bedford 59%, Fall River 58%, Dartmouth 82%.

Norfolk 88%, Wrentham 87%, Walpole 83% (only 3 towns in Norfolk carried by McCain).

[Bo]

I'd like to see if it's possible to create an Obama district in Arkansas.

[Smash255]

I'd like to see if it's possible to create an Obama district in Arkansas.

A district in Pulaski and Jefferson counties combined with some counties in east-central Arkansas would get one.

[True Federalist (진정한 연방 주의자)]

Here's a stab at redistricting SC into 7 districts.



Probably not the real district boundaries as there is a good chance this would put two Representatives in the same district, and I didn't try to gerrymander it into a potential 6 GOP/1 Dem map.

The 6th District remains majority black (53-41) which probably would pass VRA scrutiny, but it could be made blacker and thereby make neighboring seats safer for the GOP by splitting up even more counties than this does.  A 61-34 black majority contiguous district is constructable with no split precincts if you want a really ugly gerrymander.

[JohnnyLongtorso]

Here's an Arkansas map I did a few months ago that has an Obama district (the blue one):

2-2 map of Arkansas:



AR-01 (green, Marion Berry - D) - Eastern part of the state, pretty much. 74% white, 21% black. Formerly 60-40 McCain, now 58-42 McCain and probably won by Kerry in 2004.
AR-02 (blue, Vic Snyder - D) - Expands southwards to take in some of the better for Obama areas of Ross's district. 63% white, 29% black. Formerly 55-45 McCain, now 51-49 Obama.
AR-03 (red, John Boozman - R) - I think this district actually contracted. All the population growth must be in the Walmartistan part of the state. 80% white, 2% black, 11% Hispanic. Formerly 66-34 McCain, now 65-35 McCain.
AR-04 (purple, Mike Ross - D) - Loses some historically-Democratic territory in the south-central part of the state and stretches northeast to take in the more Republican parts of AR-01 and AR-02. 84% white, 8% black. Formerly 60-40 McCain, now 69-31 McCain (ouch!).

[Bo]

Here's a stab at redistricting SC into 7 districts.



Probably not the real district boundaries as there is a good chance this would put two Representatives in the same district, and I didn't try to gerrymander it into a potential 6 GOP/1 Dem map.

The 6th District remains majority black (53-41) which probably would pass VRA scrutiny, but it could be made blacker and thereby make neighboring seats safer for the GOP by splitting up even more counties than this does.  A 61-34 black majority contiguous district is constructable with no split precincts if you want a really ugly gerrymander.

If SC gets seven seats, the Justice Department will likely demand a second black-majority seat. Thus the Republican legislature will likely turn Spratt's district into a black-majority one in addition to keeping Clyburn's black-majority. This will allow the GOP to strengthen its hold on the remaning five seats.

[True Federalist (진정한 연방 주의자)]

If SC gets seven seats, the Justice Department will likely demand a second black-majority seat. Thus the Republican legislature will likely turn Spratt's district into a black-majority one in addition to keeping Clyburn's black-majority. This will allow the GOP to strengthen its hold on the remaning five seats.

I don't believe it's possible to create two majority-minority districts in South Carolina.  If it is, it would be a most hellacious gerrymander.

[Bo]

If SC gets seven seats, the Justice Department will likely demand a second black-majority seat. Thus the Republican legislature will likely turn Spratt's district into a black-majority one in addition to keeping Clyburn's black-majority. This will allow the GOP to strengthen its hold on the remaning five seats.

I don't believe it's possible to create two majority-minority districts in South Carolina.  If it is, it would be a most hellacious gerrymander.

It's possible. Here is a link to a website where someone gerrymandered South Carolina to have two black-majority districts. The SC map on top is really gerrymandered, but the SC map below it is much less gerrymandered. It is the one that makes Spratt's district black-majority in addition to Clyburn's. Thus, it is possible to create two black-majority districts in South Carolina without extreme gerrymandering.

[muon2]

If SC gets seven seats, the Justice Department will likely demand a second black-majority seat. Thus the Republican legislature will likely turn Spratt's district into a black-majority one in addition to keeping Clyburn's black-majority. This will allow the GOP to strengthen its hold on the remaning five seats.

I don't believe it's possible to create two majority-minority districts in South Carolina.  If it is, it would be a most hellacious gerrymander.

It's possible. Here is a link to a website where someone gerrymandered South Carolina to have two black-majority districts. The SC map on top is really gerrymandered, but the SC map below it is much less gerrymandered. It is the one that makes Spratt's district black-majority in addition to Clyburn's. Thus, it is possible to create two black-majority districts in South Carolina without extreme gerrymandering.

There are basically two ways to create 2 black-majority districts in SC using 7 districts, and I posted one on this thread back in Sept. I've posted both below. The major difficulty to making nice districts is the large non-black population along the coast - it's about enough for a district and a half. That means a connection is needed along one side of the state or the other.

The first map maintains the connection that currently exists between Hilton Head and Lexington county. However, it splits Charleston between three districts. The second way is to link Charleston and Hilton Head and avoid a three-way split, but then Myrtle Beach ends up connecting to Rock Hill.

[Verily]

Looking at California, it might be possible to construct an Asian-majority district in the San Gabriel Valley as well as one in the San Jose area. Anyone want to try?

[muon2]

Looking at California, it might be possible to construct an Asian-majority district in the San Gabriel Valley as well as one in the San Jose area. Anyone want to try?

Already done, back on page 18.

Here's a remap of CA with 53 districts using the 2008 data. All districts are within 1000 of the ideal population. There are 17 Hispanic-majority districts, 1 Asian-majority district and 1 Black-majority district. The other districts were drawn to maintain compactness and minimize county splits.



Zooming into LA:



Zooming into the Bay Area:



I'll let the local experts speculate on the partisan balance in these districts.

edit: maps modified to reflect some of the comments, including the addition of an Asian-majority district in and around San Jose, and modifying the northern part of LA county.

[True Federalist (진정한 연방 주의자)]

That's really squeaking by with those SC maps, muon. It won't take much of a difference between estimate and reality for those maps to be impossible.

Also your second map has CD 6 and 7 reversed, since I imagine that the GOP wouldn't be so petty as to make Clyburn change his stationary.

Also if population inflows resume after an economic recovery what you have as CD 6 in both maps would definitely not be majority black by 2020 and might well be majority white by then.

[Bo]

That's really squeaking by with those SC maps, muon. It won't take much of a difference between estimate and reality for those maps to be impossible.

Also your second map has CD 6 and 7 reversed, since I imagine that the GOP wouldn't be so petty as to make Clyburn change his stationary.

Also if population inflows resume after an economic recovery what you have as CD 6 in both maps would definitely not be majority black by 2020 and might well be majority white by then.

The maps on the link I showed were better. BTW, you can always gerrymander if necessary. In addition, if one of those districts stops having a black majority in 2020 or beyond, then policymakers can always dismantle the district at a later date.

[Bo]

Bipartisan incumbent protector:



VA-01 (blue, Rob Wittman - R) - Wittman picks up all of Stafford and part of heavily-Republican Hanover and loses some of the counties that went for Obama, and the parts of Williamsburg and Newport News as well. Also picks up the more Republican part of Hampton from Glenn Nye, as well as the Eastern Shore. This may seem weird, but while the Eastern Shore is not physically connected to VA-01, it is culturally much more like parts of the district than Virginia Beach.
VA-02 (green, Glenn Nye - D) - Dropping the Republican parts of Virginia Beach and Hampton and picks up all of Norfolk, Portsmouth, most of Suffolk and the minority-heavy part of Chesapeake. I've actually gotten it to a coalition district, as it's 49% white. Should be a safe district now, although Nye may not be safe from a primary challenge from the left unless he moves left accordingly.
VA-03 (purple, Bobby Scott - D) - Ladies and gentlemen, may I present a contiguous VA-03. Removing the Southside Hampton Roads parts of the district, I've added all of Newport News, Sussex/Greensville/Emporia, and Petersburg and the black parts of Hopewell. Remains 53% black.
VA-04 (red, Randy Forbes - R) - Losing the black parts of Chesapeake and parts of the district given to Bobby Scott, and adding the Republican parts of Virginia Beach and more of Chesterfield should make Forbes' district more Republican.
VA-05 (yellow, Tom Perriello - D) - District shifts north to Northern Virginia. Dropped all the southern part of the district and pushed it up north to Prince William. The tradeoff for Perriello is having to run in the expensive DC media market.
VA-06 (Bob Goodlatte - R) - Didn't change this district much, it just shifted south a bit.
VA-07 (grey, Eric Cantor - R) - Cantor's district is possibly even more Republican now, taking in much of the heavily-Republican Southside counties and a few Democratic counties out of Wittman's district instead of pushing northwest towards the Shenandoah Valley.
VA-08 (light purple, Jim Moran - D) - Alexandria, Arlington, and eastern Fairfax. Safe Democratic.
VA-09 (light blue, Rick Boucher - D) - Picks up Danville and drops the Ronaoke area. Not much you can do for Rick.
VA-10 (magenta, Frank Wolf - R) - Dropping Prince William and going west into the Shenandoah Valley should make this district more Republican.
VA-11 (light green, Gerry Connolly - D) - Western Fairfax, should be safe Dem.

Johnny, that it a good map. I would remove Greene, Culpepper, and Rapahannock countries, as well as the Republican areas of western Prince William County from VA-05 and put them in VA-10. I would then exapnd VA-05 into Fairfax County and then give the Democratic areas of Loundoun County to either VA-08 or VA-11 in exchange. This should make VA-05 even more Democratic while making VA-10 even more Republican. BTW, what is the white percentage in your VA-10? I'm assuming it's around 80%. Also, good job giving Boucher Henry County and the city of Danville. This makes his district more Democratic, even though it's still Republican at large. What is the white percentage of Boucher's new district? I'm assuming about 85%. I'm hoping you have saved this map.

[muon2]

That's really squeaking by with those SC maps, muon. It won't take much of a difference between estimate and reality for those maps to be impossible.

Also your second map has CD 6 and 7 reversed, since I imagine that the GOP wouldn't be so petty as to make Clyburn change his stationary.

Also if population inflows resume after an economic recovery what you have as CD 6 in both maps would definitely not be majority black by 2020 and might well be majority white by then.

I think both have enough margin, that if I have block-level control of the lines instead of precinct-level, the maps can survive reasonable fluctuations in population. I will admit I didn't think about Clyburn's home, and I left CD-7 towards the northeast in creating the second map. I agree that by 2020 continued spread of the black population in the state as well as other growth make two black-majority districts hard to see in 2020, but that isn't a factor for the VRA.

[MaxQue]

So, I redistricted the only state my computer is wanting to load. Here is New York State. I tried to reduce county splits and lower divisions (that were present in the app) splits. So, the lower division present in this app are townships, except in New York City and Nassau County, where they are Legislative districts. I didn't check representative residence.

NY-1 (Blue, Suffolk): Almost unchanged.
NY-2 (Forest Green, Western Suffolk): Now, all in Suffolk County, less Democrat, more white than before.
NY-3 (Purple, Eastern and Southern Nassau, Southeastern Suffolk): Less white, Obama would have won it.
NY-4 (Red, Western Nassau): No signifiant demographic change.
NY-5 (Gold, Northeastern Queens and Northwestern Nassau): Less hispanic, more Democrat.
NY-6 (Aqua, Southern Queens): Black majority district, go more south and talk whiter areas.
NY-7 (Gray, Eastern Bronx and Northern Manhattan): End of the Bronx-Queens district. Pluriality Hispanic and there is more Blacks than Whites.
NY-8 (Navy Blue, Far Southern Manhattan and Western Brooklyn): More normal shape and less serpent-like shape. Is only pluriality white now. Asians are second, Hispanics third.
NY-9 (Cyan, Western Queens): Much more northern than the previous NY-9. Take much of the former NY-6. Pluriality Hispanic, Whites are second, Asians third.
NY-10 (Pink, Eastern Brooklyn): More to the south than previous NY-10. Still Black majority, but less.
NY-11 (Green, Central Brooklyn): Still Black majority, but less.
NY-12 (Sky Blue, Western Queens and Northern Brooklyn): More normal shape, lost the tail in Manhattan and Western Brooklyn. Much less Asian, still Hispanic pluriality.
NY-13 (Salmon, Staten Island and Southwestern Brooklyn): Won by Obama on new borders, but still very Republican.
NY-14 (Swomp, Southcentral Manhattan): Now all in Manhatten, lost its Queens parts. Less hispanic, more white.
NY-15 (Orange, Northcentral Manhattan): Less hispanic, more white. Hispanics are first, Blacks are second, Whites are third)
NY-16 (Light Green, Western Bronx): Not many changes, still 3% white, Obama still won it with 95%.
NY-17 (Dark Blue, Northwestern Bronx and Southern Westchester): Seems more logic than splitting Yonkers. It is not anymore a coalition district, it is majority white now.
NY-18 (Yellow, Northern Wwestchester and Rockland): More rural, more white, less Hispanic.
NY-19 (Apple Green; Orange, Putham and Southern Dutchess): Slightly less white.
NY-20 (Skin; Washington, Northern Saratoga, Fulton, Herkimer, Southern Oneida, Madison, Chenango, Eastern Cortland and Western Broome): Ugly district created with areas that remained after the rest was drew. Very white district, barely won by Obama (a couple hundreds of votes).
NY-21 (Dark Brown; Albany, Schenectady, Southern Saratoga and Rensselaer): An Albany-based district. No signifiant demographic changes.
NY-22 (Light Brown; Northern Dutchess, Columbia, Greene, Ulster, Sullivan, Delaware, Otsego, Schoharie, Montgomery and Eastern Broome): A rural district between Birmingham, Albany and New York. Whiter and less Democrat, but Obama still won it, by 10 points.
NY-23 (Northen Oneida, Oswego, Jefferson, Lewis, St. Lawrence, Hamilton, Franklin, Clinton, Essex and Warren): North Country. Almost no change.
NY-24 (Dark Purple; Tioga, Tompkins, Chemung, Schuyler, Steuben, Yates, Ontario, Southern Monroe, Livingson, Allegany and Cattaraugus): The former NY-29, a little bit more Democrat, but McCain still won it.
NY-25 (Blood; Onondaga, Wayne, Seneca, Cayuga and Western Cortland): Former NY-25, with rural parts south of Syracuse added and with Rochester parts removed. No significant demographic change.
NY-26 (Dark Gray; Chautauqua, Wyoming, Genesee, Orleans, Niagara and Outer Erie): Former NY-26, but with areas around Buffalo added and areas around Rochester removed. Was the most Republican district in New York State, is a little bit more Republican now.
NY-27 (Light Green; Buffalo and suburbs): Now a very compact district based on Buffalo. More black than former NY-27 and is now the most Democrat upstate district.
NY-28 (Pink; Rochester and suburbs):  Now a very compact district based on Rochester. Less Democrat and black than the former NY-28, who was a Buffalo-Rochester district following the Canadian border.

[True Federalist (진정한 연방 주의자)]

The maps on the link I showed were better. BTW, you can always gerrymander if necessary. In addition, if one of those districts stops having a black majority in 2020 or beyond, then policymakers can always dismantle the district at a later date.

First off, what link?  You mentioned a link, but there was no link.  Second, my point was mainly that a barely majority-minority district crafted based on the 2010 data might well have one or more of the five election cycles of its lifetime when it was not actually majority-minority.

[Bo]

The maps on the link I showed were better. BTW, you can always gerrymander if necessary. In addition, if one of those districts stops having a black majority in 2020 or beyond, then policymakers can always dismantle the district at a later date.

First off, what link?  You mentioned a link, but there was no link.  Second, my point was mainly that a barely majority-minority district crafted based on the 2010 data might well have one or more of the five election cycles of its lifetime when it was not actually majority-minority.

Sorry about that. Here's the link:

http://www.swingstateproject.com/diary/5183/redistricting-south-carolina-2-blackmajority-seats

As for the second part, it is irrelevant since the district only needs to be majority-AA in the 2010 Census.

[muon2]

The maps on the link I showed were better. BTW, you can always gerrymander if necessary. In addition, if one of those districts stops having a black majority in 2020 or beyond, then policymakers can always dismantle the district at a later date.

First off, what link?  You mentioned a link, but there was no link.  Second, my point was mainly that a barely majority-minority district crafted based on the 2010 data might well have one or more of the five election cycles of its lifetime when it was not actually majority-minority.

Sorry about that. Here's the link:

http://www.swingstateproject.com/diary/5183/redistricting-south-carolina-2-blackmajority-seats

As for the second part, it is irrelevant since the district only needs to be majority-AA in the 2010 Census.

The SC map at the link is a mess! I now feel like my 2 different versions of 2 black-majority districts that I offered are things of beauty by comparison.

[Bo]

The maps on the link I showed were better. BTW, you can always gerrymander if necessary. In addition, if one of those districts stops having a black majority in 2020 or beyond, then policymakers can always dismantle the district at a later date.

First off, what link?  You mentioned a link, but there was no link.  Second, my point was mainly that a barely majority-minority district crafted based on the 2010 data might well have one or more of the five election cycles of its lifetime when it was not actually majority-minority.

Sorry about that. Here's the link:

http://www.swingstateproject.com/diary/5183/redistricting-south-carolina-2-blackmajority-seats

As for the second part, it is irrelevant since the district only needs to be majority-AA in the 2010 Census.

The SC map at the link is a mess! I now feel like my 2 different versions of 2 black-majority districts that I offered are things of beauty by comparison.

The second SC map in the link (below the first one--keep on scrolling down) is pretty clean. It is also the one that makes Spratt's district AA-majority in addition to Clyburn's.

[muon2]

The maps on the link I showed were better. BTW, you can always gerrymander if necessary. In addition, if one of those districts stops having a black majority in 2020 or beyond, then policymakers can always dismantle the district at a later date.

First off, what link?  You mentioned a link, but there was no link.  Second, my point was mainly that a barely majority-minority district crafted based on the 2010 data might well have one or more of the five election cycles of its lifetime when it was not actually majority-minority.

Sorry about that. Here's the link:

http://www.swingstateproject.com/diary/5183/redistricting-south-carolina-2-blackmajority-seats

As for the second part, it is irrelevant since the district only needs to be majority-AA in the 2010 Census.

The SC map at the link is a mess! I now feel like my 2 different versions of 2 black-majority districts that I offered are things of beauty by comparison.

The second SC map in the link (below the first one--keep on scrolling down) is pretty clean. It is also the one that makes Spratt's district AA-majority in addition to Clyburn's.

Well it was better than the first, but it still erred by trying to come down both edges of the state to the coast. The best solution is to pick one or the other edge to reach the coast and use border counties on the other side for an AA district. The comments that assume the worst case for a GOP gerrymander, don't understand the more subtle ways the goal can be reached, IMO.

[True Federalist (진정한 연방 주의자)]

I'm not certain those gerrymanders would pass scrutiny.  They not only have to make monstrous black majority districts, they make monstrous white-majority districts.  I'm going to see if I can make a more reasonable gerrymander.

[Verily]

Looking at California, it might be possible to construct an Asian-majority district in the San Gabriel Valley as well as one in the San Jose area. Anyone want to try?

Already done, back on page 18.

That's not the San Gabriel Valley. The San Gabriel Valley is east of LA. It might not be possible. I've been working on it, and I currently have a 47% district that is still about 200,000 people undersized.

Also, there is a plurality Native American block group in Northeast LA, just outside of Chinatown (!!!, but probably a transposition of Asian and NA).

Another edit: Okay, 51% Asian right now with full population. The district connects the main hub of Chinese around Monterey Park to three other areas: Cerritos near the border with OC and a few heavily Asian areas right over the border in OC (mixed Chinese/Korean), the Asian neighborhoods around Rowland Heights, Walnut and Diamond Bar (mixed and wealthy), and LA's Chinatown. The racial split is 51% Asian, 27% Hispanic, 19% white. I'll post a map soon.

Map: