Texas, it's bigger than France

[jimrtex]

This is an updated version that includes more details in urban areas.

In the thread about a possible Georgia redistricting, I was challenged to produce a Texas map without the bacon strips districts.  This is a map of a plan that I outlined in that thread.



Basically, I divided the state into 6 regions (Houston, DFW, Central (I-35), Border, Eastern, and Western) and apportioned the 32 representatives among them.  I then shifted a few counties so that the population of each region was close to a whole number of districts of an ideal population size (651,619).  In the process of delineating the districts in each region, I identified a few other inter-regional shifts of counties that would provide better inter-district balance.

The plan as presented would have a maximum deviation of 0.5% (or around 3,000 people).  9 counties are split, the 6 that have more than one district's worth of population: Harris, Dallas, Tarrant, Bexar, Travis, and El Paso, and 3 others: Williamson, Nueces, and Cameron.  9 districts do not contain any split counties, including all 8 in the more rural Eastern and Western regions.   The Border region, while having vast tracts of rural areas, is actually highly concentrated in the cities of El Paso, Laredo, Lower Valley, and Corpus Christi.

East (5 districts).
Liberty, Bastrop, Guadeloupe, and Live Oak shifted in.  Falls and Chambers shifted out.  Population equivalent to 5.005 districts.  The districts pretty much drew themselves based on 3 of the districts starting in the corners of the region, and the other two having to form an east-west split in the center.

District 1: North East Texas, 1.005

District 2: Deep East Texas, 0.996

The deviation betwen 1 and 2 (the 2 greatest for the state) could be reduced by swapping Rains and Wood for Panola and Shelby.  This would make two a much odder shape, but would mantain the Tyler-Longview-Nacogdoches-Lufkin core.

District 3: South East Texas, 1.000

Combines the industrial Golden Triangle, and the timber/recreation areas to the north, and Huntsville into a single district, but avoids sticking the two parts in Houston dominated districts.

District 4: East Central Texas.  1.001

District 5: Mid Gulf.  1.003

West (3 districts).
Cooke, Montague, Palo Pinto, Somervell; Bosque, Erath, Hamilton, Mills, Lampasas, Loving, Wink, Ward, Crockett, Sutton, Edwards shifted out.  None shifted in.  The west region is entitled to about 3.3 districts.  Rather than attempting to draw 4 districts, some of which would soon find them being drawn into major metropolitan areas, the excess population is shifted to a pair of districts that are somewhat less densely populated.  Over the next decade, the western districts will likely need these areas returned.

District 6: Panhandle-Red River 1.000

District 7: South Plains-Abilene 0.999

District 8: Permian Basin-San Angelo-Hill Country 0.999

Border (4 districts).
Loving, Wink, Ward, Crockett, Sutton, Edwards shifted in,
Live Oak shifted out.  This has very close to the correct population for 4 districts.  The population of this region is concentrated in the far opposite ends.  Even District 13, most of the population is in the south eastern section, with about 1/3 in Laredo.

District 12: El Paso 0.997

Roughly 0.046 or 30,000 persons in south east El Paso County are moved out.

District 13: Mid Valley-Trans Pecos-Southwest Texas 0.998

Includes 0.046 from El Paso County, and 0.033 (22,000) from southwest Nueces County.

District 14: Lower Valley.  1.000

Includes 0.044 (29,000) from Cameron.

District 15: Coastal South Texas.  1.000

Excludes 0.033 (22,000) from Nueces County (including Robstown and Bishop), and 0.044 (29,000) from Cameron County (La Feria, Santa Maria, and areas south of Harlingen and San Benito.  This district links Corpus Christi and Brownsville with little population in between.  Removing a little from both ends, avoids picking a favorite.

Central (5 districts).
Bosque, Erath, Hamilton, Mills, Lampasas, Falls shifted in.  Bastrop and Guadeloupe shifted out.  Most of the area shifted in is added to the less populated Waco-Temple-Killeen area on the north end.  The rest of this area is sliced in pieces along I-35.

District 9: Central Texas 0.995

Includes 0.034 (22,000) from eastern Williamson County, including Taylor.

District 23: North Austin 1.005.  Roughly 1/3 in Williamson County and 2/3 in Travis County.   The boundary between districts 23 and 24 is the Colorado River, except for an area including downtown Austin and UT which is in District 24.

District 24: South Austin-San Marcos-New Braunfels 0.998

Travis County is split between 23:24 0.655:0.592 426,000:385,000.  0.136 (89,000) of Bexar is included in in District 24 (small cities in northeast Bexar County, including Selma, Univeral City, Converse, Live Oak, and Windcrest).

District 25: North San Antonio 1.001

District 26: South San Antonio 0.999

The boundary between 25 and 26 is a little bit north of downtown San Antonio.

DFW (8 districts).
Cooke, Montague, Palo Pinto, Somervell shifted in, none shifted out.  In my original outline, I underestimated the population of Dallas County which is entitled to 3.4 districts, and overestimated the population of the southern suburban counties.  So I have created a Tarrant-Dallas Mid Cities district, and combined the southwestern and southeastern counties with districts that are largely based in Tarrant and Dallas County.

District 10: Denton and Northwest DFW Metro. 1.002

7/8 of the district is in the counties adjacent to Tarrant County (2/3 in Denton alone).

District 16: South Tarrant and Southwest DFW Metro 1.003

0.735 (479,000) is from Tarrant County (Arlinton, Mansfield, Kennedale, Forest Hill, Rendon, Everman, Crowley, and an area of Fort Worth south of I-20.

District 17: East Dallas County and Southeast DFW Metro 0.999

0.653 (426,000) is from Dallas County, including Garland, Mesquite, Sachse, Rowlett, Sunnyvale, Balch Springs, Seagoville, and Combine.  Rockwall County was shifted from District 18 in my previous map.

District 18: Collin and North Dallas County 0.997

0.242 (158,000) is from Dallas County, including Richardson, Carrollton, Farmers Branch, and Addison.

District 19: Fort Worth and Northwest Tarrant County 0.995

Fort Worth and areas to the north and , including Haltom City, Watauga, Keller, Saginaw, Eagle Mountain Lake, Azle, White Settlement, River Oaks, and Lake Worth.

District 20: Tarrant-Dallas 0.999

Tarrant:Dallas contribute 0.489:0.510 318,000 :332,000 along the northern part of their boundary, including, Coppell and Irving in Dallas Counties, Grand Prairie in both counties, and Grapevine, Hurst, Euless, Bedford, Richland Hills, North Richland Hills, Colleyville, Southlake, and Westlake in Tarrant County.

District 21: North Dallas 0.1001

Dallas from Downtown northward and including Park Cities.

District 22: South Dallas 1.000

South Dallas and cities in southern Dallas County, inlucing Cedar Hill, Duncanville, De Soto, Lancaster, Wilmer, and Hutchins.

Houston (7 districts).
Libery shifted out, Chambers shifted in.  My original proposal called for 5 districts in Harris County, and one linking Montgomery and Fort Bend.   The addition of Chambers made creation a southeatern district easier.  Montgomery and Fort Bend are instead paired with adjacent areas in Harris County.

District 11: Galveston Bay-Brazoria 1.002

Includes 0.209 (136,000) from shoreline of Galveston Bay in Harris County, including Baytown, La Porte, Seabrook, El Lago, Taylor Lake Village, and Webster.

District 27: Montgomery-Northeast Harris 1.001

Includes 0.551 (359,000) from Northeast Harrist County, including Spring, Humble, Kingwood, Atascosita, Lake Houston, Crosby, Barrett, Highlands, Channelview, and Sheldon.

District 28: Fort Bend-Southwest Harris 1.001

Includes 0.457 (298,000) from Southwest Houston roughly along the Southwest Freeway.

District 29: Northwest Harris County 1.001

Includes outer edges of Houston, Tomball, Jersey Village, Katy, Cypress, and 1960 area.

District 30: West Houston 1.002

Includes enclaves of West U, Bellaire, and Memorial villages.

District 31: North Houston 1.002

Also includes Aldine, Jacinto City, and Galena Park.

District 32: Southeast Houston and Southeast Harris County.

Includes Pasadena, Deer Park, and South Houston.

[jimrtex]

Looks nice.
I haven't checked, but from memory, many of the rural districts (outside the Southern bacon strips of course) look a lot like what existed pre-regerrymander.

I assume you mean the map used for the 2002 election (descendant of the
1990s Frostrocity).  Not really.

The 2002 had a lot of quirky little boundaries (see 4, 5, 8, and 23).  It was also
intent on preserving incumbents, and had many districts running from rural areas
into the major metropolitan cities.

1 on the 2002 plan is more NS, and wraps around Tyler-Longview.  The 2004
plan and my plan are more EW.  The main difference between the 2004 and my
plan was that I wasn't trying to draw a district for an incumbent.

2 on the 2002 plan is more southerly (the 2002 plan splits up the cities in East
Texas).

3 on the 2002 plan was more than 1/2 in the Houston area.  My plan goes
north rather than west.

There is nothing like 4 on my plan.  Districts like 5, 6, and 31 in the 2002 are
very much urban districts. that gather up pieces of the area.  The 4 on the
2002 plan was intended to pack Republicans.

5 on my plan is quite similar to 14 on the 2002 plan.  The biggest difference
is that my plan cuts out the incumbent.  The original 1990s plan was drawn
for a Democrat incumbent.   The claw into Brazoria County was to pull Republicans
out of 14.  The Democrat switched parties, then lost the GOP primary to
Ron Paul, who happens to live in the area of Brazoria County beyond the
claw.

13 on the 2002 plan is similar to 6 on my plan.

7 and 8 on my plan are similar in configuration to the 2004 plan.  The
big difference, was that I wasn't trying to pull certain counties away
from the incumbent.

23 on the 2002 plan and 13 on my plan are somewhat similar.  But
this is not entirely true since 22 has a sizeable portion in Bexar county,
including the incumbent.

9 in my plan is similar to 11 in the 2002 plan.

So I'd say that 5, 6, and 9 are similar to the 2002 plan.  1 and 13
are to some extent, but also fairly different.

Four of my rural seats, 2, 5, 8, and 13 would have been open seats.
1, 3, and 4 had a sitting incumbent, but with a quite different
district.  7 would have paired 2 incumbents.

Based on the 2004 election, 1, 3, 4, and 5 would be open seats.

[jimrtex]

Great job on the districts.

I have some idle questions about the regional groupings. I spent quite some time in TX during the early 1990's particularly in the area from Dallas to Waco where the SSC was planned. Whereas College Station seemed East Tx to me, just about everything west seemed more like central TX. This feeling would apply to areas like Hill, Navarro, and Limestone Cos in that area, and Grayson Co. up north. Based on this visitors impressions I would have said that your CD 4 splits between East and Central TX. Similarly your CD 5 seemed much more like San Antonio and Corpus than Houston, so it could be Border or Central, but I didn't visit there as much, so I'll freely admit any impressions may be off.

Certainly not piney woods East Texas.

My primary objective was to separate out the larger metropolitan areas.  Because of
the position of the I-35 corridor and the DFW area, there is a split between the more rural/
smaller cities to the east and west.  My definition of East and West Texas is on that basis.
If Navarro County is not in the I-35 corridor, and not in DFW, it must be in East Texas.

Under the current redistrictricting plan, many of the districts extend outward from
the larger metropolitan areas, leaving the less populous areas without representation.
Using my regional definitions, the current representation is DFW (9, 1 extra), Houston
(8, 1 extra), Central/I-35 (7, 2 extra), East(1, 4 deficit), West (3, no imbalance), Border
(4, no imbalance).

While Grayson would be a small fish if placed with Collin, it is the most
populous county in my District 1.

This map shows my regional boundaries (heavy black line).  Counties in yellow were
shifted for population balance.



The West region had a population equivalent to about 3.3 representative.  The DFW and
East regions had a deficit.  The transfer from the West region to the East region was via
the Central/I-35 region.

A) The border region had a very small deficit which was satisfied by these transfers.

B) A shift from from the West region to the DFW region.  Originally Jack, rather
than Somervell had been shifted.  Removing Jack permitted District 10 to be drawn
without splitting counties.

C) A shift from the West region to the I-35 region.  The counties in A and B are the
easternmost in the West region.  Removing them somewhat reduces that extent.

D) A shift from the Central/I-35 to the East region.  Falls was countershifted
for population balance.  As you note, this moves suburban areas of San Antonio
and Austin into the East region.

E) A shift from Houston to East Texas to remove Housto's slight excess.  The
countershift of Chambers is for population balance.

F) A final pair of shifts made after the basic districts had been set out, and
District 5 was found to have a slight deficit that couldn't be corrected by swaps with
District 4, and West Texas still had a slight excess.  In a sense, this could be considered
as an extension of the C and D shifts which moved population from west to
east Texas.

[jimrtex]

Although not as many as the current one...

But not serving any purpose other than preserving
the disticts drawn in 1991.

[jimrtex]

While it would no doubt mess up your regionality criteria, the districts could probably be made considerably more compact of Dostricts 6 and 7 excjanged territory so that D7 touched the Red River instead of New Mexico and if D8 and D13 were seperated by boundary roughly perpendicylar to the one you gave.

The population of 6/7 is highly concentrated in 4 cities: Lubbock, Amarillo, Wichita
Falls, and Abilene (The 5 counties - Amarillo straddles the Potter/Randall line
have about 55% of the total population).   Extending 7 north to the Red River
would include Wichita Falls.   This would mean that 6 would have to include Lubbock.
But Lubbock is bigger than Wichita Falls, so 7 would have to take in a large swath
of the lightly populated eastern Panhandle.  A compact Amarillo-Lubbock district
can be drawn (two counties from the first 4 tiers in the Panhandle, and 3 counties
from the next 4 tiers in the South Plains).  But this leaves a really huge eastern
district.

The Trans Pecos of 13 is very lightly populated, and about 1/2 of that is on
the extreme east and west ends (the remnant of El Paso County and Val
Verde County).  If you take the 18+ counties west of the current District 8
and gave it to District 8, this could be swapped for 6 counties north of
San Antonio.  This would greatly extend  the extent of 8.  It might appear
more compact than the current 13 (because it would somewhat rectangular),
but part of the oddness of the shape of District 13 is due to the Big Bend.

If South Texas continues to grow, then it might be all of a sudden that it
won't need the Trans Pecos area.  District 13 would be the a more
compact area where most of the population is.   This would make the
Trans Pecos region available for West Texas districts.

[jimrtex]

I added some details to the original map.

[jimrtex]

Nice map, jimtex

and the results would be ?

East Texas (1-5) 4 R; 3 (SE Texas) could be competitive because of Jefferson County, but would need the right candidate to hold Democrat votes in the rest of the district.

West Texas (6-10) 5 R; though Chet Edwards might be able to hold 9.

Gulf Coast (11) 1 R, though could be competitive.

Border (12-15) 4 D.  15 might be slightly competitive.

DFW (16-22) 6 R and 1 D (22).

Central (23-26) 2-3 R and 1-2 D.  23 and 25 would be R, with 26 D.  24 would be competitive with mix of Travis county with Comal and NE Bexar county.

Houston (27-32) 4 R and 2 D (31 and 32).

Overall, R 21 to 24, D 11 to 8.

[jimrtex]

Sorry, for my ignorance, but is that more Republican seats than now, or less?

Same basically, but Jim is being very optimistic for his party in East Texas, possibly elsewhere.

The current split is 21:11 (22:10 if Edwards were defeated).  Bush carried 25 CDs (4 with a Democrat congressmen - Cuellar, Hinojosa, Ortiz, and Edwards).

A lot would depend if the seats were open, or there was an incumbent Democrat.  Remember that there is one less State Senate seat in Texas than CDs, and there are no Democrat Senators from East Texas.  One GOP senator is running for Ag Commissioner, and no Democrats are seriously challenging for open seat.

1 (Northeast) Sandlin might have held on here, but he had been challenged even under the 90 Democrat gerrymander.
2 (East Texas).  No chance for Democrat in district that includes Tyler, Longview, Lufkin, and Nacogdoches.  Turner represented a district that carefully avoided more populated areas.
3 (South East)  Lampson might have held on, but remember that Jack Brooks was beaten when the district included Galveston and Baytown.  The areas to the north are going to be reluctant to support a candidate from Beaumont.
4. (East Central) The last Democrat to carry this area was Phil Gramm.  Brazos and Washington counties are too strongly Republican for the Democrat areas further north to overcome.
5. (Upper Gulf) Current Democrat Senator from this area is not running for re-election, and GOP pickup is almost certain.  Includes suburbs of San Antonio, Austin, and exurbs of Houston.
6-8. (West Texas)  These are Nebraska-like areas.
9. (Central Texas)  Only Chet Edwards can win this seat for Democrats.  He almost lost it in 2002.
10. (Denton and Parker and nearby areas).  Democrats wouldn't even run.
11. (Galveston Bay).  Houston suburbs balance Galveston and southern mainland of Galveston County, but Brazoria is very Republican, and will outvote Baytown.  SE Harris suburbs will vote GOP.
12-15. (Border) 13 would be a pickup for the Democrats with Bexar Country excluded.
16-22 (DFW).  There is currently one Democrat from the area, and there is no reason that the districts that I have drawn will perform differently.
23-26 (Central).
23. Williamson County and northern Travis County will vote GOP.
24. Includes UT and central and south Austin, but southern parts of district towards San Antonio are extremely Republican areas.
25 and 26.  The two Bexar County seats will split (remember that Bexar is the large county that most closely tracks the national result).
27-32 (Houston).
27-30 will vote Republican, with Democrats only running in 28.
31 and 32 will be reliably Democrat, though primaries will be very interesting since they mix Hispanic and Black areas.

So that's 8 seats the Democrats would win (12-15, 22, 26, 31-32)
4 that they could win (1. 3. 9. 24) and
5 that they would bother to run a candidate (5, 11, 16, 19, 28).

[jimrtex]

11 is similar to Ron Paul's old district.  I would be more optimistic than jimrtex is on Democrat chances, but the exurban growth of Brazoria County tempers my optimism a bit.  It certainly could be competitive.

Galveston County is about evenly split between suburban Houston (area just south of NASA) and the island and Texas City areas, and has recently elected Republicans to countywide office.  Brazoria County has about the same population as Galveston County, and has high suburban growth in the northern portion of the county due to the opening of the South Freeway.  This is the area of undeveloped land that is closest to downtown Houston (5 to 10 miles closer than areas to the north and west).  DeLay's district became less favorable to him because he gave up parts of the Brazosport area (the claw that extended into Brazoria County was part of the Democrat's GOP packing scheme).   About 20% of the district is in Harris County.  While Baytown will vote slightly Democrat, the areas closer to Clear Lake will vote strongly GOP.

You must be logged in to read this quote.

24 includes the UT and capital areas north of the Colorado River (elsewhere the river is the boundary).  This will make 24 the more competitive seat.   Even if the Democrats managed a small majority in the Travis County portion of the district, the Williamson County portion would tip the district.

24 is split 59% Travis; 27% Hays and Comal; and 14% Bexar.  This would be a highly polarized election, with the respective parties possibly piling up 70% of their votes in their half of the district.   I would make this a probable Democrat district, and shift 23 further to the top of your list of probable Republican districts.

You must be logged in to read this quote.

Over the long haul, 11, 27, and 28 will shift more into the suburban counties.  Currently they are: 11, 79% in Brazoria, Galveston, and Chambers; 27, 45% in Montgomery; and 28, 55% in Fort Bend.  A new seat would then be created in Harris County, which would likely be competitive Democratic.

[jimrtex]

16-22 (DFW).  There is currently one Democrat from the area, and there is no reason that the districts that I have drawn will perform differently.

Oh yes there is. Dems actually topped the congressional poll across Dallas County (and Harris County too) in 2004. Obviously the 3 outer districts are utterly safe Rep, I'd have to find me some more details before making predictions on the other three, but there is certainly not material for two safe Rep districts in them. At best (for you), one almost-as-safe-as-now (where all Republicans have been surgically removed) Dem seats and two lean Rep ones, but I need convincing.

In Harris County, this was because 2 candidates did not have GOP opponents, while one GOP candidate did not have a Democrat opponent.  If you look at the RRC race, the GOP had a +86K plurality in Harris County; along with +65K in Montgomery, +18K in Fort Bend, +23K in Brazoria, and +6K in Galveston.

The 4 seats in Harris County will split 2:2, all with at least a 60% majority.  The suburban counties will ensure the other 3 districts are GOP held.

In Dallas County, it was a similar situation, with two uncontested seats, one Dem held and entirely in Dallas County, and one GOP held but mostly in Collin County.   The GOP won the RRC race by +4K.  My districts for Dallas County are remarkably similar to the districts created by the Texas Legislature in 2003.  Pete Sessions defeated Martin Frost by 20K in the North Dallas district.

The two seats wholly in Dallas County will split 1:1.  The suburban (and Tarrant) counties will ensure the other seats are GOP held.

[jimrtex]

1 (Northeast) Sandlin might have held on here, but he had been challenged even under the 90 Democrat gerrymander.
2 (East Texas).  No chance for Democrat in district that includes Tyler, Longview, Lufkin, and Nacogdoches.  Turner represented a district that carefully avoided more populated areas.

Basically, if D2 can't elect a Democrat, D1 is highly competitive. If D1 is safe-ish for Reps, then D2 is only safe-ish either. I'd have to have a closer look at population distribution, but looks like the first of these is correct. Of course it would have to be the right type of Democrat.

57.7% in RRC race, with big margins in the more western counties especially Grayson (Sherman-Dennison) and Hunt (Greenville) which are the 1st and 3rd largest counties.  The GOP also carried Bowie (Texarkana), Lamar (Paris), and Harrison (Marshall).  The Democrat candidate carried 4 counties, mostly very small.

You must be logged in to read this quote.

The areas to the North are not more Republican than the areas to the East of Jefferson County. Probably less so. (Not thinking of presidential elections here.) This isn't safe, but it sure favors Dems.[/quote]
The areas to the east of Jefferson County don't vote in Texas elections.   The RRC race was 48.2% GOP to 51.8% Dem.   The rest of the district nibbled a 17.4K margin from Jefferson County down to 7.6K

You must be logged in to read this quote.

No contest (although Charlie Stenholm did carry quite a few counties last time around.)[/quote]
In 2002, he almost lost.  He failed to carry his home county (Taylor) as well as large counties of Erath, Brown, Tom Green.  The largest county that he carried was Wise.   Wise is not in my district, but Lubbock is.

You must be logged in to read this quote.

When it was cut more explicitly with the purpose of getting him out.[/quote]
He almost lost in 2002 when the districts imposed by the federal judges tried to preserve the partisan 1990s gerrymander.  Like the 2002 district, the core of my district is McLennan, Bell, and part of Williamson, plus smaller counties to the northwest.

[jimrtex]

Yes, Austin can complain about this map too, that's true.

Travis County is entitled to about 1.25 representatives.  It makes more sense to split the county than to create one whole district, and having another district that includes areas to the north and south that are not linked by any highways.

You must be logged in to read this quote.

GOP representatives had a +10K majority (all 4 races were contested by both parties).  The areas in Bonilla's seat (+45K), and Smith's seat (+49K) would be in the northern district.  The areas in Gonzalez's district (-58k) and Cuellar's district (-26K) form the southern district.  The area along I-35 that would be in 24 is a fairly small portion of the county (about 6%).  Comal County (New Braunfels) voted 80% Republican, and is the heartland of German Texas (even San Antonio is historically a German rather than a Hispanic city).  Hays County voted 60% GOP.

[jimrtex]

(The other seat I'd been wondering about was the between-Dallas-and-Fort Worth one, o/c)

This would be Grand Prairie, Irving, Carrollton, and the Tarrant suburbs west of DFW (airport).

[jimrtex]

He did carry Taylor in 2004., when Wise was not in his district, but Lubbock was. (He carried 5 counties - Castro, Fisher, Kent, Taylor, and the portion of Nolan to be included in his district. Most of these are tiny.) Mind you, that train's departed, this is a safe Rep district now.

He carried Taylor County (Abilene) when he was facing a candidate from Lubbock County.  He did not carry Taylor County when facing a candidate from Tom Green County (San Angelo).