The 1,000 Districts Series

[Miles]

The Southeast NC is too big to pack into one district and D-leaning. With two districts, it almost inevitably wastes Republican votes, although exactly how you draw the map affects things. Additionally, a lot of the urban areas occur in something of a belt, and most of them are too big for one district, so they swallow right-wing suburbs, wasting votes.

Yep, thats why I had to make two Democratic sinks in southeastern NC, one in Fayetteville and another one from Lumberton to Wilmington.

My (very quick) attempt at what Republicans might draw in NC:



This would be 22-9.

Democratic sinks:
- 2 majority-black seats where Butterfield's district is currently located.
- 3 in the Triangle
- 2 in near Fayetteville; the Lumberton-Wilmington CD is less than 60% Obama, but is very heavily D downballot.
- 1 in Charlotte, which is 85% Obama. Black VAP majority.
- 1 in the Triad, which is also (barely) over 80% Obama. Black VAP majority.

I've actually been meaning to clean this map up for a while.

[minionofmidas]

Montana might be slightly beneficial to the Democrats, at worst even. It's kind of tricky to judge and depends a lot on the election.

I got one seat for either party with the third being slightly Republican:

MT:



There wasn't pre-loaded election data in DRA, so I crunched the numbers myself.

CD1:
P- 60.5/36.7 Romney
S- 48.2/45.0 Rehberg
G- 51.8/44.6 Hill

CD2
P- 48.8/48.1 Romney
S- 53.9/40.0 Tester
G- 53.7/42.5 Bullock

CD3
P- 57.5/39.7 Romney
S- 46.9/46.3 Rehberg
G- 48.3/47.9 Hill

Basically, CD1 would be Safe R, CD2 would be Likely/Safe D and CD3 would be swingy, but tilting more R than the state.

I'm sure the green district could be 'unpacked' so that the the purple one would move to the left.

The purple district is also quite unfortunate. Marias Pass is the only real connection between its eastern and western halves.

[traininthedistance]

The purple district is also quite unfortunate. Marias Pass is the only real connection between its eastern and western halves.

Some quick fiddling in Montana has convinced me that any map, especially one that keeps counties whole, is going to have to cross some mountains somewhere, and probably in an even worse spot.

I'm 99.9% sure that Miles' map cannot be improved upon.  (And, thankfully, 2R-1D is an accurate reflection of the state's political lean.)

[minionofmidas]

The purple district is also quite unfortunate. Marias Pass is the only real connection between its eastern and western halves.

Some quick fiddling in Montana has convinced me that any map, especially one that keeps counties whole, is going to have to cross some mountains somewhere, and probably in an even worse spot.

I'm 99.9% sure that Miles' map cannot be improved upon.  (And, thankfully, 2R-1D is an accurate reflection of the state's political lean.)

I haven't done the attempt, but from what I remember from the 100k thingy, I suspected as much.

[traininthedistance]

Florida.  In 61.

So, there are two really salient features when it comes to Florida map-making.  One is that it is way way easier than you would expect to follow county lines.  Lots of easy whole-county groupings here, and a few of them quite unexpectedly felicitous.  For example, Pinellas is exactly three districts (well, the three most underpopulated districts on the map) AND Hillsborough is exactly four.  Lots of happy stuff like that.

The flip side is that this map kind of really sucks by VRA standards.  The state is only 61 percent white by VAP (even less by total population), and "should" yield about 8 black districts and between 12-13 Hispanic districts.  But the minority population is pretty well spread out, with many concentrations that can't quite support a full district.  Except for one area, Miami-Dade, which has a gigantic super-packed Hispanic population (well, the Cuban parts are super-packed at least) tucked away in a corner of the state such that you can't really unpack it very well.  So Hispanics really do get some short shrift here- only eight Hispanic-majority districts are realistically possible, seven of them are in Miami-Dade, and most of those seven are really packed pretty heavy.  Blacks don't do any better- only four actual black-majority districts are possible.  

I did try to put together a number of minority-majority districts in addition to the strict VRA ones, and was even willing to get a little ugly on a few of them.  While I'm pretty satisfied with the map as a whole, I really don't like the choices that were available in two areas of the map, namely around Orlando/Polk and Palm Beach.  Those areas took the vast majority of my time here, and not coincidentally, are the areas where I'd consider revisiting things.  In particular, at some point I'll put up alternate versions which focus more on clean lines and less on min-maj opportunities in these two regions.    I fixed the Palm Beach area and now have lines there I'm happy with.  The old version will remain up as a curiosity, the final section will be a revised PB.  I may fix Orlando/Lakeland/Winter Haven etc., but it is not the highest priority ATM.

Onto the map!  It'll take three four posts.  



The Panhandle.  



DISTRICT 1: PENSACOLA.  Pop 307,969.  O 39.8%.  70W/20B.  All of Escambia and a sliver of Santa Rosa.  Not a whole lot of choices here in the Panhandle.  No, I did not seriously see if you could make a Pensacola-to-Tallahassee black district.  It would ruin the whole-county grouping of the first three districts, anyway.  Safe R.

DISTRICT 2: SANTA ROSA-OKALOOSA.  Pop 309,156.  O 26.8%.  Workin' across the Panhandle, this district is the vast majority of the two named counties; Fort Walton Beach, the Eglin Air Force Base, and Pensacola burbs appear to be the main sources of population here.  Safest R in the state, narrowly beating out its neighbor 3.

DISTRICT 3: PANAMA CITY- WEST FLORIDA.  Pop 308,818.  O 27.9%.  82W/10B.   The rest of Okaloosa, Bay County (Panama City), and five surrounding rural counties in the Panhandle.  Crossing the Apalachicola River makes it harder to craft perfect names for these districts, but that's how you get it in whole counties.  Safe R.

DISTRICT 4: TALLAHASSEE-GADSDEN-JACKSON.  Pop 308,199.  O 63.0%.  57W/34B.  The two named counties, and Tallahassee in Leon.  You can't avoid splitting at least one county in the Tallahassee area, and you can't get a min-maj district either.  But this version keeps the splits minimal (only Leon, and no cities are split) while also keeping most of the region's black voters together.  Safe D.

DISTRICT 5: APALACHICOLA FOREST-NORTH FLORIDA.  Pop 306,495.  O 38.4%.  74W/19B.  The rest of Leon, and ten more counties stretching across North Florida from the rest of the Panhandle east to the Okefenokee Swamp.  The most thoroughly rural district in the state.  Safe R.

Jacksonville.  Districts six through nine are all in Duval, and take in four more whole counties.



DISTRICT 6: JACKSONVILLE AREA BEACHES, NORTH, AND WEST.  Pop 306,618.   O 32.9%.  78W/12B.  Baker, Nassau, and two sections of Duval- the western edge and a larger portion in the north and east, including the Beach towns and most of the Northside and Arlington.  Drawn around 7 and 8, obviously.  Safe R.

DISTRICT 7: JACKSONVILLE CENTER AND NEAR WEST.  Pop 307,951.  O 68.6%.  38W/52B.  First of two all-Duval districts, and our first, black-majority, VRA district.  Corrine Brown's current district is kind of a monstrosity, but you can definitely get two sensible smaller districts out of its two poles, and this is the first one.  (The Orlando portion is merely black-plurality, but what can you do.)  Safe D.

DISTRICT 8: JACKSONVILLE SOUTHEAST.  Pop 307,750.  O 39.3%.  Entirely within southeast Duval east of the St. John's.  Nice squarish shape.  Safe R.

DISTRICT 9:  CLAY-MARION-JACKSONVILLE RIVERSIDE.  Pop 307,594.  O 33.6%.  79W/11B.  Starting with the Riverside/Avondale neighborhoods, this district snakes down the St. Johns (helpfully excising white areas from 7) and then takes in two counties- suburban/exurban Clay (which has most of the population here), and Marion (home of the Palatka micropolitan area).  Safe R.

North-central Florida.



DISTRICT 10: GAINESVILLE AREA.  Pop 308,330.  O 55.4%.  69W/18B.  Alachula County and three small surrounding counties (Union, Bradford, and Gilchrist, which is technically in the metro area).  The only district here that is actually whole counties.  Lean D.

DISTRICT 11:  OCALA AREA.  Pop 306,642.  O 44.4%.  77W/11B.  Marion County is larger, so it gets bites taken out instead- a little each to 12 and 13, which together with 11 make another whole-county grouping.  Safe R, but not quite as monolithic as parts north.

DISTRICT 12: HOMOSASSA SPRINGS-THE VILLAGES.  Pop 306,704.  O 38.8%.  Two rural counties along the least-touristy part of the Gulf Coast (Dixie and Levy), a small bite of Marion, but mostly Citrus and Sumter counties, in all their micropolitan, age-restricted sprawldivision glory.  The Villages is actually split with Marion and Lake, but the bulk is here.  Safe R.

DISTRICT 13: LAKE.  Pop 306,883. O 43.2%.  79W/10H.  Except for another small bite of Marion, this is all Lake County.  Safe R.  

DISTRICT 14: ST. AUGUSTINE-PALM COAST.  Pop 308,216.  O 39.5%.  St. John's, Flagler, and a small, rural part of northern Volusia; basically the Atlantic coast just south of Jacksonville.  It gets harder to do whole-counties at this point; the twelve-county, fifteen district grouping that starts here and is centered on Orlando is the largest such grouping in the state.  Safe R.

DISTRICT 15:  DAYTONA BEACH-CENTRAL VOLUSIA.  Pop 308,368.  O 53.5%.  81W/11B.  The all-Volusia district, it takes in almost all of the beach portions and Deland (with some buffer districts because the lines are wacky)- really, most of the county's populated areas except for Deltona.  It would be better to split things entirely beach/inland, but the beach section is not quite large enough; this configuration preserves road connections and town lines at well as possible, though.  Tossup; I can't tell who won this district in 2012, but I do know it was razor-thin.

DISTRICT 16: DELTONA-WEST SEMINOLE.  Pp 309,446.  O 52.9%.  69W/11B/17H.  Workin' down the chain of counties to Orlando, southern Volusia and about a third of Seminole.  I thought about splitting southern Volusia with Brevard, but that particular county line corresponds to a lull in development, and Deltona is much more a natural partner to Sanford.  I also tried to see if Sanford could fit in the Orlando-area black-pluraility district, but it in fact does not.  Especially if you insist on preserving road contiguity.  Romney probably won this in 2012, but since I'm not 100 percent sure, I'll say Tossup despite a possible R tilt.

The rest of Central Florida, coming up in Part II!

[traininthedistance]

Closer-up on Orlando, with Kissimmee-St. Cloud and the Space Coast too.



DISTRICT 17: EAST SEMINOLE-WINTER PARK.  Pop 309,381.  O 46.1%.  74W/15H.  Almost entirely a Seminole district, but dips down into Orange County to take in most of Maitland and Winter Park.  Not all, though, because it doesn't need all of them, whereas District 18 does want a couple of their precincts.  Safe R.

DISTRICT 18: WEST ORLANDO-APOPKA.  Pop 308,804.  O 71.5%.  33W/43B/19H.  You can't get a black-majority district in Orlando (too many mixed districts, not enough blacks), but you can get a pretty strong plurality this way.  As a bonus, it's one of three all- Orange districts.  Safe D.

DISTRICT 19: CENTRAL ORLANDO-EASTERN ORANGE.  Pop 308,004.  O 55.9%.  63W/22H.  It doesn't really hew to town lines perfectly (given the jagged lines and minority-opportunity districts around it), but most of Orlando is here, and then it hooks around the most heavily Hispanic precincts in 20 to newer subdivisions (and subdivisions to be) in the east of the county.  Lean D.

DISTRICT 20: KISSIMMEE- SOUTH AND EAST ORLANDO.  Pop. 310,724.  O 65.5%.  33W/52H.  You can make one Hispanic-majority district in the Orlando area, but it can't be majority by much due to the large number of mixed precincts, and it also forces extra county splitting since it has to enter both Orange and Osceola.  You can goose the percentage a point or two if you get uglier in Orange and/or go deeper into Osceola, but I liked the idea of getting another minority-majority district out of the area (24) instead.  Safe D.

DISTRICT 21: DISNEY WORLD.  Pop 308,696.  O 52.6%.  55W/12B/23H.  Our third all-Orange district includes towns like Ocoee and Williamsburg and even the southwest corner of Orlando itself, but obviously it's most famous for that eponymous theme park.  Tossup.

DISTRICT 22: CAPE CANAVERAL-ST.CLOUD.  Pop 310,320.  O 43.9%.  79W/10H.  Very much this part of the map's leftovers district.  I wanted to have an all-Brevard district that didn't split towns, and St. Cloud has to be separated from Kissimmee (and really, the whole Orlando urbanized area) if you're making a Hispanic district there, so linking St. Cloud with the (smaller) northern half of Brevard through a thinly-populated corner of Orange was the least bad option.  Titusville is actually the largest city here. Better than throwing St. Cloud in with a district to its south, that's for sure.  Safe R.

DISTRICT 23: PALM BAY-MELBOURNE.  Pop 311,048.  O 45.6%.  The southern half of Brevard, mostly those two cities.  Safe R I guess?

Tampa Bay and environs.  



DISTRICT 24: POINCIANA-WINTER HAVEN-CENTER LAKELAND.  Pop 309,094.  O 58.3%.  54W/18B/24H.  The rest of Osceola (save one district in the undeveloped southern portion necessary to connect 35) and much of Polk.  This district is minority-majority by total population, and it's my attempt to do that which creates this admittedly weird shape.  Polk is almost perfect for two districts, but it's not quite right, and surrounding county groupings force it to take three districts within its  boundaries anyway: since Pinellas and Hillsboro are whole-county, one of Lake's districts has to take the Pasco-Hernando leftovers, leaving too much left.  This is roughly the more Orlando-oriented Polk district, I guess.  Lean D; presumably it would be a tossup if I threw minority voting power to the wind and just made clean lines.  Hewing to town boundaries is out of the question here, though: they're just too squiggly and don't line up with precincts.

DISTRICT 25: LAKELAND AREA.  Pop 309,626.  O 41.5%.  78W/11H.  Mostly northern and western Polk fitting around 24, also takes in Dade City in Pasco, more Tampa-fringe than the last one.  24 and 25 together take most of Polk, the less-developed southern bit is given to 35 instead.  Safe R.

DISTRICT 26: HERNANDO-OUTER PASCO.  Pop 308,103.  O 47.4%.  83W/10H.  All of Hernando, some of Pasco, the northern reaches of the Tampa Bay area.  An obvious district, but how exactly to draw the line between this and 27 doesn't seem to make a whole lot of difference to my outsider eyes.  Lean R, bordering on safe I guess.

DISTRICT 27: NEW PORT RICHEY-INNER PASCO.  Pop 308,805.  O 49.1%.  Entirely in Pasco.  Mostly along the Gulf Coast, but not exclusively.  Right in the middle of the Lean R range.

DISTRICT 28: DUNEDIN- PINELLAS NORTH.  Pop 305,585.  O 49.4%.  Pinellas perfectly works out to three districts, with the caveat that all of them are pretty much maximally underpopulated by the deviations I'm allowing myself.  North/Central/South is pretty much the only sensible way to divide it far as I can tell.  Most of Clearwater, the county's second largest city, is here, but since it is by necessity split, it doesn't get top billing anywhere.  Lean R.

DISTRICT 29: LARGO- PINELLAS CENTRAL.  Pop 305,204.  O 52.0%.  More of the  same.  At 3K underpopulated, the highest deviation here, or pretty much anywhere.  Tossup.

DISTRICT 30:  ST. PETERSBURG-PINELLAS SOUTH.  Pop 305,753.  O 61.2%.  73W/17B.  Just takes a little of Pinellas Park from 29 for population.  Safe D.

DISTRICT 31: TAMPA WEST-TOWN N' COUNTRY.  Pop 305,865.  O 58.1%.  47W/10B/38H.  The Hispanic-influence Tampa district, and given the geography of Tampa Bay (and the heavily white areas to its south) it would require throwing road connectivity to the wind to do much better on that front.  Still white-plurality by total population, but closer (44 to 40).  Yadda yadda usual disclaimer about trying to hew to town lines but not being perfect.  Lean D.

DISTRICT 32: TAMPA EAST-BRANDON.  Pop 307,618.  O 69.8%.  44W/32B/20H.  And the black-influence Tampa district.  Brandon is actually split with 34 to keep the numbers up, same with a couple other localities.  Safe D.

DISTRICT 33: HILLSBOROUGH NORTH.  Pop 307,593.  O 44.0%.  68W/18H.  The only one of our four Hillsborough districts to not actually enter Tampa at all.  Safe R.

DISTRICT 34:  TAMPA BAY-HILLSBOROUGH SOUTH.  Pop 308,150.  O 45.6%.  72W/16H.  And the rest of Hillsborough.  Hillsborough is even more accommodating than Pinellas, as 32, 33, and 34 are all pretty close to the ideal population.  Still Safe R for now, but perhaps not for that long.

DISTRICT 35:  INDIAN RIVER-SOUTH POLK-HIGHLANDS.  Pop 308,726.  O 41.7%.  77W/12H.  The biggest "leftovers" district on this map, it starts with the remainder of Polk (part of Bartow, but not very developed otherwise) and goes in search of capping off Central FL in whole counties.  To that end, it cuts through empty southern Osceola to the shore county of Indian River (host of its own small metro area, Sebastian/Vero Beach) and goes south to Highlands, which is mostly Sebring and farms.  I was initially hoping to have Highlands anchor some sort of expansive, rural "interior southern Florida" district, but that region kept being necessary for whole-county groupings otherwise.  Oh well.  This district and 38 hopefully capture some of that spirit.  Safe R.

South from Tampa, it's the Suncoast!  (Okechobee County is slightly inaccurate here.)



DISTRICT 36:  BRADENTON AREA.  Pop 307,522.  O 47.1%.  78W/12H.  Southwest Florida is almost as felicitous as Pinellas/Hillsborough for whole-county groupings, and we have another one with 36/37/38.  Manatee is just a little too big for a district, so we have an all-Manatee district along the coast (Bradenton and assorted other retiree havens) and the rural inland is given to 38.  A pretty strong Lean R.

DISTRICT 37: SARASOTA-VENICE BEACH.  Pop 307,673.  O 50.3%.  Sarasota County is larger, so this district sticks closer to the coast, and abandons one of its larger cities (North Port) to 38.  Retiree heaven just like everything else in the state. Lean R despite the Obama win.

DISTRICT 38: CHARLOTTE-SOUTHWEST.  Pop 309,657.  O 45.3%.  82W/10H.  The rest of Manatee and Sarasota, the Port Charlotte/Punta Gorda area (Charlotte County; also North Port feels like to belongs here too), and two more rural interior counties (DeSoto and Hardee).  Still mostly Suncoast subdivisions, but the most rural district in the area all the same.  Safe R.

DISTRICT 39: CAPE CORAL-FORT MEYERS-LEE WEST.  Pop 310,802.  O 46.4%.  75W/15B.  Lee perfectly subdivides into two districts; it should say something (something very damning) about the nature of development in Florida that this part has both major cities, not much else, and is still basically half of the county's land area.  Safe(-ish) R.

DISTRICT 40: BONITA SPRINGS-LIBERTY VILLAGE-LEE EAST.  Pop 308,499.  O 43.2%.  77W/15H.  More of the same.  The division here is actually more northwest/southeast, as 40 has Sanibel Island sticking out into the ocean south of Cape Coral.  Safe R.

DISTRICT 41:  NAPLES AREA.  Pop 310,379.  O 38.6%.  72W/21H.  And the southern edge of the Suncoast, Naples and environs, and then taking in part of the Glades.  Collier is just a little too big for a district, and it has to lose a jagged bite from the north because a) Miami/Broward/Monroe is a whole-county grouping (which really was a shock when I realized), and b) Immolakee is all one huge precinct.  So my hand was forced. Anyway, Safe R.

South Florida in the last part.

[traininthedistance]

Part three!

The Port St. Lucie/Fort Pierce area, Palm Beach County, and Lake Okechobee:



DISTRICT 42: PORT ST. LUCIE-FORT PIERCE-OKECHOBEE NORTH.  Pop 307,923.  O 55.5%.  66W/16B/15H.  All of Okechobee and almost all of St. Lucie; something has to be split here.  This is a late revision, as I was trying for quite a while to keep St. Lucie and take the rest from either parts of Okechobee or Martin.  However, I also wanted to try and accommodate minority-opportunity districts in Palm Beach, especially since that area features much of Alcee Hastings' district, and with more than twice the number of districts, it feels like retrogression to replace one 700K VRA district with only one 300K VRA district, no matter how horribly erose.  And Hastings' 2003-2013 district is about as horribly erose as they come, with cutouts of Fort Pierce and West Palm Beach and Fort Lauderdale (which is actually easy to make one smaller VRA district out of), connected by rural areas but not by road.  Anyway, losing the Fort Pierce bit was easy, it's gone in the current district and can't help the black percentage anyway because it's so isolated.  So I wanted to to a district with all of St. Lucie, but eventually found that taking all of Okechobee instead and microchopping St. Lucie was necessary to get at least something with a non-Anglo plurality in the area.  Urf.  Still, it's not that bad as far as clean lines go here.  Lean D.

DISTRICT 43:  MARTIN-JUPITER-PALM BEACH GARDENS.  Pop 306,434.  O 45.4%. That tiny bit of St. Lucie, the vast majority of Martin, and some area in the north of Palm Beach, mostly the two named towns.  For the longest time this was all of Martin, but eventually I put Indiantown in what became the Hispanic-plurality 45.  The county split is unfortunate, but it's at least a clean, coherent CoI of relatively conservative coastal communities on the northern fringe of South Florida.  Safe R.

DISTRICT 44: COASTAL PALM BEACH.  Pop 309,277.  O 72.1%.  46W/34B/16B.  So, 44 and 45 are gerry-rigged pretty badly here, to be honest.  This is, despite appearances, the northern successor to Hastings' district, collecting the county's black areas (except for inland Belle Glade) as non-horribly as possible, which means going down the coast between northern West Palm Beach and Boynton/Delray.  I tried and tried and tried to get a black-plurality district and it was impossible, this is just about the best you can do here.  I felt I had to do something VRA-ish here, but next order of business will be to throw up an alternate map in this area that goes for cleaner lines and doesn't bother to salvage/maximize VRA representation.  Safe D.

DISTRICT 45: INLAND PALM BEACH-OKECHOBEE SOUTH.  Pop 309,227.  O 60.9%.  38W/18B/41H.  A lot of inland southern FL- the rest of Collier and Martin, all of Henry and Glades, the agricultural and protected western half of Palm Beach.  And then a snake around 44 and 46, carefully making sure to maintain road contiguity (unlike Hastings' current district, which cuts across Loxahatchee) to mixed and Hispanic areas just south of West Palm Beach.  I eventually realized that, despite black-majority Belle Glade, the inland areas are actually more Hispanic, and trying to get a Hispanic-plurality district here is pretty much the only way to make anything without an Anglo plurality.  So that's what I did.  Safe D.

DISTRICT 46: WELLINGTON-BOYNTON WEST-MID PALM BEACH.  Pop 305,982.  O 61.2%.  73W/13H.  Most of the population here is actually in the unincorporated urbanized areas between Wellington and Boynton Beach.  Basically it cuts out the whitest areas from 45, and is thankfully not nearly so ugly itself.  Safe D.

DISTRICT 47:  BOCA RATON- PALM BEACH COUNTY SOUTH.  Pop 308,012.  O 59.6%.  82W/10H.  Except for the cutout of the black community in Delray Beach for 44, this is the simplest Palm Beach County district: just Boca, Delray Beach, and adjacent built-up areas in the south of the county.  Safe D.

Disregard these districts, part four will have better versions of them.

Broward and Miami-Dade.  These two counties, plus Monroe (which is mostly the Keys) combine perfectly to fit fourteen districts.



DISTRICT 48: CORAL SPRINGS-TAMARAC-NORTHWEST BROWARD.  Pop 310,214.  O 67.2%.  51W/21B/21H.  Couldn't find a way to avoid the split of Sunrise with 49, but everything else is along town lines.  (That's why there's that weird little spit into 51.)  While the physical northwest corner of Broward is in 49, that's all empty Glades area- the northwest corner of urbanized Broward is here.  This district is also minority-majority by total population- 48 percent white.  Safe D.

DISTRICT 49: DAVIE-PLANTATION-WEST BROWARD.   Pop 310,129.  O 60.5%.  55W/10B/29H.  Mostly Davie, Plantation, Weston, and half of Sunrise, also the empty Glades part of the county.  Safe D.

DISTRICT 50: LAUDERDALE AND POMPANO EAST-DEERFIELD BEACH.  Pop 309,029.  O 57.7%.  71W/11B/15H.  Northeast Broward, splits Fort Lauderdale and Pompano Beach with 51 for VRA purposes.  Lean D.

DISTRICT 51: CENTRAL BROWARD-LAUDERDALE AND POMPANO WEST.  Pop 305,629.  O 85.4%.  28W/55B/14H.  Only Fort Lauderdale and Pompano Beach are split in this compact black-majority district, the southern and far more sensible half of Alcee Hasting's monstrosity.  Safe D.

DISTRICT 52:  HOLLYWOOD-PEMBROKE PINES EAST.  Pop 308,512.  O 67.4%.  47W/16B/32H.  Southeast Broward- mostly just Hollywood and a slight majority of Pembroke Pines, which is the only town it really splits.  There are a lot of mixed precincts here, so a mixed min-maj district is easy and appropriate.  Safe D.

DISTRICT 53: NORTHWEST DADE-SOUTHWEST BROWARD.  Pop 305,607.  O 55.5%.  14W/15B/66H.  First of two districts to split Broward and Miami-Dade, first of seven Miami-Dade Hispanic districts.  Splits several towns with 52 and 54, and takes in some Glades area because that's what works best for road contiguity.  Lean D; the Miami-Dade part has a lot of Cubans, but the Broward part doesn't (while still having quite a few Hispanics that should go in a VRA district of some sort).

DISTRICT 54: MIAMI GARDENS AND MIRAMAR EAST-AVENTURA.  Pop 309,024.  O 85.8%.  18W/51B/27H.  The black community in Miami-Dade (and just north of it) can squeeze out one compact district in the current map, and can comfortably get two here.  This is the northern half of that community, and it also goes a little further east to take the mostly-Anglo Aventura (declining to maxpack blacks) so that only two districts have to straddle Miami-Dade and Broward.  Miramar and Miami Gardens (which looks like "Carol City" on this map) are the only towns split, obviously for VRA purposes.  Safe D.

DISTRICT 55: HIALEAH-DORAL.  Pop 308,600.  O 38.7%.  93H.  Pretty much exactly just Hialeah, Doral, and Hialeah Gardens.  Enough of a super-pack of the Cuban community that I feel dirty- that 93 percent Hispanic is the highest percentage of any non-white ethnicity we've seen so far.  It will likely remain the highest in the country with one probable exception- El Paso, where such a pack is well and truly unavoidable.  (Possibly also the Rio Grande Valley.)  But you can't really unpack this district much, seeing as it is entirely surrounded by other VRA districts, and the geography of South Florida is a de facto corner.  So, why not hew to town lines.  Anyway, Safe R- the Cuban areas of Miami-Dade are a demographic anomaly matched only by the Orthodox areas of Southern Brooklyn.

DISTRICT 56:  NORTH MIAMI-MIAMI NORTH.  Pop 307,267.  O 86.9%.  14W/53B/31H.  Back to blue, the colors have made a full cycle around now.  The black neighborhoods in the north of Miami, and a number of (mostly) black-majority towns, well, north of Miami.  Safe D.

DISTRICT 57:  MIAMI EAST-MIAMI BEACH.  Pop 308,766.  O 60.6%.  26W/65H.  A couple other tiny towns, but mostly Miami Beach and a big chunk of Miami, including downtown.  It looks like pretty much all of Miami's non-Cuban Hispanics are here- it may be the one Miami-Dade district where the Hispanic vote actually leans Dem within the county.  (61 is a possibility, too, but this one I'm pretty sure about.)  Safe D.

DISTRICT 58: FOUNTAINBLEAU-TAMIAMI-WEST OF MIAMI.  Pop 309,382.  O 36.0%.  91H.  Like 55, overwhelmingly a) Cuban, b) Republican, c) whole towns (almost exclusively CDPs in this case), and d) surrounded by more minority districts.  Marginally less Hispanic, and more Safe R. Without a doubt the most Republican urban district in the nation outside of Boro Park.

DISTRICT 59:  MIAMI WEST-CORAL GABLES-KENDALL.  Pop 309,950.  O 47.9%.  22W/72H.  Most of the Cuban parts of Miami proper, and several cities and CDPs to the city's southwest.  (Kendall and Coral Gables being easily the two largest.)  Lean R; the Anglo vote is probably quite Democratic here, but it's overshadowed by Cubans, who are still Republican.  But for how long?

DISTRICT 60: CUTLER BAY-KENDALE LAKES- NEAR SOUTHWEST DADE.  O 58.8%.  19W/15B/62H.  More Miami suburbs, going southwest from 59.  The lines here, as is the case for most of the county, are meant to ensure whole towns and a strong Hispanic majority (this district, though, is certainly not Cuban majority).  Lean D.

DISTRICT 61: HOMESTEAD-GLADES-KEYS.  O 56.6%.  32W/11B/55H.  The rest of Miami-Dade, down to Homestead and the Everglades, and all of Monroe, which despite having some land area in the Glades, is by population entirely the Florida Keys.  This district does throw a finger up to The Hammocks to ensure another Hispanic majority, but it's clearly the most tenuous majority, as Homestead is mixed and the Keys are mostly white.  No getting around that, though.  Lean D.

A fixed Port St. Lucie/Palm Beach etc. in Part 4.

[traininthedistance]

A much better Palm Beach, Port St. Lucie, etc.:



DISTRICT 42: PORT ST. LUCIE-FORT PIERCE.  Pop 308,381.  O 55.0%.  67W/16B/15H.  Just St. Lucie and most of Okechobee (which keeps Martin whole).  Lean D.

DISTRICT 43:  MARTIN-JUPITER-PALM BEACH GARDENS.  Pop 307,387.  O 45.9%. All of Martin and the northern fringe of Palm Beach.  Almost exactly the same as the earlier version, just with fewer county splits.  Safe R.

DISTRICT 44: WEST PALM BEACH SOUTH-GREENACRES.  Pop 307,306.  O 66.0 %.  43W/17B/36H. Cleaning up the lines doesn't mean I need to lose a minority-majority district, both this and 45 remain less than half Anglo.  I do give up on connecting the inland Hispanic areas to this district, which is the area to the south and west of West Palm Beach (and including it somewhat).  Compared with the old map this area thus take some neighboring precincts, cleans up town lines (WPB is still split, as is Lake Worth), and expands a bit.  Somehow, the result is still plurality-Hispanic by total population, just barely (38.8 to 38.1)!  Safe D.

DISTRICT 45: WEST PALM BEACH NORTH-OKECHOBEE.  Pop 309,909.  O 65.2%.  45W/31B/21H.  The rural bits (this time small sections of Collier and Okechobee, all of Hendry and Glades, western Palm Beach) get attached to the black areas of WPB again (as in the actual map).  And the lines are further cleaned up by taking Anglo areas north of Wellington and retreating from Boynton/Delray.  But the black percentage doesn't go down that much, because it retreats from the very white coast as well.  Safe D.

DISTRICT 46: WELLINGTON-BOYNTON BEACH-COAST.  Pop 307.336.  O 62.3%.  75W/11B/11H.  Instead of an ugly cutout of 45 north of Wellington, it takes all of Boynton then goes up the beachfront of Palm Beach, which has much the same effect.  Lake Worth and Lantana are split, but that's it.  Safe D.

DISTRICT 47:  BOCA-DELRAY-PALM BEACH COUNTY SOUTH.  Pop 306,536.  O 60.5%.  78W/10H. All of Boca Raton, all of Delray Beach, and unincorporated subdivisions to their west.  The Delray cutout is eliminated.  Safe D.

…


So.  

Four black-majority districts (7, 51, 54, 56).  Eight Hispanic-majority districts (20, 53, 55, 57, 58, 59, 60, 61).  One more black-plurality district (18) and one more Hispanic-plurality district by total population only (44); four more minority-majority by VAP (31, 32, 45, 52) and two by total population only (24, 48).


Safe D 17
Lean D 9
Tossup 4
Lean R 6
Safe R 25

That's… actually a pretty reasonable snapshot of the state, which has what I'd call a fairly faint neutral-year R tilt.

[I spent the winter writing songs about getting better]

How is that 42 Lean R? It was no doubt won by Obama in 2012 as well, and maybe even Kerry.

[traininthedistance]

How is that 42 Lean R? It was no doubt won by Obama in 2012 as well, and maybe even Kerry.

Because it's a typo, that's why.  I meant Lean D, and had it as Lean D in the final tally.

Thanks, fixing it right now.

[traininthedistance]

Alabammy.

The black population in AL is just over a quarter by total population, and just under a quarter by VAP.  So four black districts would be ideal, and four is certainly possible.   This map only actually has three black-majority districts, though, since it puts a lot of emphasis on CoI and avoiding county splits.  (Only six counties are split, and three of them are too large for a district.)  There is a fourth Dem-leaning district that is barely white-majority, which at least gets us close to ideal while still keeping lines easy.

The whole state:



The Mobile area:



DISTRICT 1: MOBILE.  Pop 297,525.  O 53.7%, D 51.9%.  54W/40B.  This is the one white-majority district with a Dem average in the state, and it's not very white-majority (by total population, it's 51/43).  Almost all of the City of Mobile, and some surrounding districts mostly to the south (which gives 2 better road contiguity).  The fact that Mobile/Baldwin together fit perfectly into two districts, and are a distinct community, is the main reason I declined to attach Mobile to an actual black-majority district.  Should still be Lean D, though- those aren't strong numbers, but the Deep South is so polarized and inflexible already that it's probably pretty secure.

DISTRICT 2: BALDWIN-NORTH OF MOBILE.  Pop 297,732.  O 23.3%, D 26.6%.  84W/10B.  All of Baldwin, the rest of Mobile County, Safe R.

DISTRICT 3: BLACK BELT WEST-TUSCALOOSA SOUTH.  Pop 297,391.  O 57.4%, D 63.9%.  46W/52B.  Twelve whole rural counties, either within or directly adjoining the western Black Belt, and part of Tuscaloosa.  The largest city entirely in this district is the 20K large Selma, which should give you an idea of how rural this area is.  I'm not going to bother with closeups for districts 3 through 7, but the Birmingham zoom will show the Tuscaloosa split at least.  Safe D.

DISTRICT 4: DOTHAN-ENTERPRISE-WIREGRASS.  Pop 298,805.  O 28.8.%, D 35.7%.  73W/22B.  Eight whole counties in the state's southeast, quite rural aside from Dothan and Enterprise.  By strict CoI concerns Dale County (Ozark) belongs here, and parts of Conecuh could go to 3, but I'm not going to turn down whole counties.  Safe R.

DISTRICT 5- AUBURN-OPELIKA-OZARK.  Pop 301,208.  O 40.8%, D 43.8%.  66W/27B.  North of 4 and wrapping around the black-majority 6, this district has all of Pike, Dale, Barbour, and Lee, and splits Russell (just east of Columbus, GA) with 6.  Not exactly urban, but more large towns than 3 or 4.  One of the state's more Democratic districts, and yet still Safe R.

DISTRICT 6- MONTGOMERY-BLACK BELT EAST.  Pop 296,568.  O 61.9%, D 58.5%.  40W/55B.  All of five counties: Montgomery and the more rural Butler, Lowndes, Macon, and Bullock; and the split of Russell.  Our second black-majority county; Safe D.

DISTRICT 7: NORTH OF MONTGOMERY-ALEXANDER CITY.  Pop 301,732.  O 28.7%, D 34.3%.  Eight whole counties; the northern suburbs and exurbs of Montgomery, the micropolitan area of Alexander City (which is still within Mongomery's orbit), and a few completely rural counties.  Safe R.

Closeup on Birmingham and environs:



DISTRICT 8: SHELBY-ST. CLAIR-BIBB.  Pop 301,593.  O 21.8%, D 25.4%.  83W/11B.  Those three counties (Shelby dominates, though), all in the Birmingham metro and all exurban or worse.  The state's worst Democratic average, and I think the most anti-Obama district we've seen so far (though I don't know that the numbers have been crunched for Miles' LA map) but incredibly not the most anti-Obama district in the state.  Safe R, duh.

DISTRICT 9: GADSDEN-ANNISTON-TALLADEGA.  Pop 297,964.  O 34.5%, D 42.5%.  75W/20B.  Four whole counties in the east-central part of the state: Talladega, Calhoun (Anniston), the rural Cleburne and Cherokee.  Splits Etowah with 10, but has just about all of Gadsden.  Somehow the small towns of Gadsden and Anniston are considered "metropolitan areas" by the Census Bureau.  Safe D.

DISTRICT 10: CUMBERLAND PLATEAU.  Pop 300,624. O 18.3%, D 31.7%.  Safest R in the state by Obama-in-08 numbers, this district includes four whole counties north of the Birmingham region and south of the Tennessee Valley: DeKalb (Fort Payne), Blount (technically in the fringe of the Birmingham metro), Cullman, and Winston; also splits Etowah and dips into Jefferson a bit.  3/9/10/11/12/13 is the one part of the map I couldn't get into small county groups.  Safe R of course.

DISTRICT 11: HOOVER-BESSEMER-SOUTH JEFFERSON.  Pop 297,619.  O 39.1%, D 38.5%.  68W/25B.  Suburban Jefferson County south of Birmingham (which is where most of the suburbs are), attempting to follow city lines as much as possible.  Most of it is lily-white and deeply Republican, but Bessemer modulates the final numbers a bit.  Safe R.

DISTRICT 12: BIRMINGHAM CITY.  Pop 296,698.  O 75.2%, D 72.9%.  33W/61B.  Mostly just the city of Birmingham, and a couple precincts to its immediate north.  Easily the state's most urban, most black, and most Safely D district.

DISTRICT 13: JEFFERSON AND TUSCALOOSA NORTH- WEST CENTRAL.  Pop 297,798.  O 24.1%, D 38.2%.  87W/10B.  The rest of those two metro counties, and four more mostly rural counties (Walker, Marion, Fayette, Lamar).  Seals off Central Alabama, letting North AL be three clean districts. Safe D.

And Northern Alabama:



DISTRICT 14: DECATUR-SCOTTSBORO-ALBERTVILLE.  Pop 300,045.  O 27.3%, D 41.8%.  Four whole counties: Lawrence, Morgan, Marshall, and Jackson.  This whole region has a Dem Average far higher than the Obama average, since the TVA was active here and it held on to its Dixiecrat tendencies longer than the rest of the state.  But those tendencies, and their last beneficiary (Parker Griffith) appear to be gone with the wind.  Safe R.

DISTRICT 15: FLORENCE-MUSCLE SHOALS-NORTHWEST.  Pop 299,048.  O 33.7%, D 49.2%. 82W/12B.  Franklin, Colbert, Lauderdale, Limestone, and a little bit of Madison, which is too large for a district.  I am going to operate under the assumption that the Dem Average here is now well and truly irrelevant, and say Safe R.

DISTRICT 16: HUNTSVILLE.  Pop 297,386.  O 42.8%, D 48.2%.  68W/23B.  Most of Madison, all of Huntsville.  This is almost certainly the district in the state with the most liberal white electorate, where "most liberal" probably still implies something less than 30 percent Obama.  With Huntsville being an actual city in addition to the Dixiecrat factor, it is the closest Republican district in the state and the only one that I'd consider even remotely competitive outside of dead girl/live boy situations.  How about an optimistic Lean R.

3 Safe D (all black-majority), 1 Lean D, 1 Lean R, 11 Safe R.  Incredibly, this is better for Dems than what they have now.

[Benj]

Surely Birmingham City has the most liberal white electorate in the state. There are a bunch of 70+% white precincts in the SE of Birmingham proper that voted for Obama, some with more than 60% of the vote.

[traininthedistance]

Surely Birmingham City has the most liberal white electorate in the state. There are a bunch of 70+% white precincts in the SE of Birmingham proper that voted for Obama, some with more than 60% of the vote.

Yeah, good catch.

[traininthedistance]

Oh, hey, forgot I had this lying around for the last month.  It's an alternative Mississippi that prioritizes keeping counties together over strict population equality (though it still hews to my rule of keeping districts no more than 1 percent away from the ideal).  With a lot of rural area, a lot of small counties, and a lot of areas where county government is the smallest subdivision, doing something like this makes sense here.  Regions are still mostly kept together, and a fair ratio of black-majority districts is also maintained.  The state is almost 35 percent black by VAP, and there are three black-majority districts plus a fourth where they break 40 percent and could conceivably elect a Democrat as well.

Only two counties are split- something must be split in the Gulfport/Biloxi/Passacougla area (EDIT: if you want to keep the cities there together, as I do), and I also split Madison, which has obvious CoI reasons and may be necessary as well (is not strictly necessary but enables one district to remain entirely within the Jackson metro, and improves the black percentage of another).

The state:



DISTRICT 1: SOUTHHAVEN-OXFORD-NORTHWEST.  Pop 296,040.  O 38.6%.  69W/25B.  The heart of this district are the Memphis suburbs/exurbs in DeSoto, which take up over half the population here.  Also included are two more rural counties technically in the Memphis metro (Tate and Marshall); Lafayette (Oxford), Benton and Yalobusha.  Safe R.

DISTRICT 2: TUPELO-NORTHEAST.  Pop 295,205.  O 29.3%.  80W/17B.  Ten counties, Tupelo is easily the main population center here but it's mostly rural.  This is the least black district in the state, though 9 is a smudge more white.  Safe R.

DISTRICT 3: DELTA NORTH.  Pop 297,676.  O 59.6%.  41W/56B.  Fifteen whole counties, entirely rural (the largest town is Clarksdale, 17K as of the last Census), not entirely within the Delta but close to it.  Safe D.

DISTRICT 4: GOLDEN TRIANGLE-MERIDIAN-EAST CENTRAL.  Pop 295,080.  O 47.7%.  57W/40B.  The largest city here is Meridian, but the Starkville/Columbus/West Point area is larger in aggregate.  Nine whole counties consisting basically of those four towns and surrounding rural areas.  This is the one district outside of the Delta or Jackson to include a sizable amount of the Black Belt, and as such is far more closely divided than the rest of the state.  In addition, there might even be a couple white liberals in Starkville.  A fairly mild Lean R sounds right, if the state flips, this district will lead the way.

Might as well show things a little closer in southern MS, to get the two split counties at least:



DISTRICT 5: DELTA SOUTH-NATCHEZ TRACE.  Pop 296,873.  O 58.9%.  41W/55B.  The second black-majority district, it takes in thirteen whole counties and rural north Madison.  The rest of the Delta is here, as well as heavily-black areas in the southwestern corner of the state.  Quite rural, but a few more recognizable towns than District 3, with Vicksburg, Greenville, and Natchez.  Safe D.

DISTRICT 6: JACKSON CITY AND WEST.  Pop 296,047.  O 64.8%.  35W/62B.  All of Hinds, and then it takes in Copiah to the south and a few precincts of Madison for population.  One of two districts in the state with any even vaguely urban flavor at all (10 is the other), and also the most heavily black and Safest D.

DISTRICT 7:RANKIN-EAST OF JACKSON.  Pop 298,780.  O 28.2%.  72W/22B.  Mostly Rankin County and the suburban, white parts of Madson; also takes four more counties in the center of the state to fill out population (Simpson, Scott, Leake, Neshoba).  I expected this to be the safest R district, but it's actually not.  It is still obviously quite Safe R, though.

DISTRICT 8: LAUREL-BROOKHAVEN-SOUTH CENTRAL.  Pop 294,104.  O 35.6%.  66W/31B.  Thirteen counties, very rural, no famous identity like the Delta. The state's highest deviation, at -2,626.  Safe R.

DISTRICT 9: HATTIESBURG-PINE BELT.  Pop 298,822.  O 26.2%.  78W/17B.  Six whole counties and a split of Harrison with 10; reaches down to a little of the Gulf Coast but most of that area is in 10.  The safest R in the state.

DISTRICT 10:  GULF COAST.  Pop 298,670.  O 35.8%.  70W/21B.  All of Jackson and most of Harrison, the Gulfport/Biloxi/Passacougla area.  With an Obama percentage over 14 points higher than the black percentage, this appears to be the least polarized district in Mississippi, which is, erm, not exactly a high bar.  Assuming that Gene Taylor stays retired, it's Safe R.

...

3 Safe D (all black-majority), 1 Lean R, 6 Safe R.  A pretty obvious arrangement, about the only gerrymandering you can do here is give the Dems a fourth minority-majority district (which would be based on 4), or break up 4 for the Pubs' benefit.

[dpmapper]

DISTRICT 3: BLACK BELT EAST-TUSCALOOSA SOUTH.  Pop 297,391.  O 57.4%, D 63.9%. 

DISTRICT 5- AUBURN-OPELIKA-OZARK.  Pop 301,208.  O 40.8%, D 43.8%.  66W/27B.  North of 4 and wrapping around the black-majority 6, this district has all of Pike, Dale, Barbour, and Lee, and splits Russell (just east of Columbus, GA) with 6.  Not exactly urban, but more large towns than 3 or 4.  One of the state's more Democratic districts, and yet still Safe R.

DISTRICT 6- MONTGOMERY-BLACK BELT WEST.  Pop 296,568.  O 61.9%, D 58.5%.  40W/55B. 

Love the series, and I don't mean to be nitpicky, but I've noticed here (and in many previous posts) a tendency for you to get east and west mixed up.  Just an FYI. 

[traininthedistance]

DISTRICT 3: BLACK BELT EAST-TUSCALOOSA SOUTH.  Pop 297,391.  O 57.4%, D 63.9%.  

DISTRICT 5- AUBURN-OPELIKA-OZARK.  Pop 301,208.  O 40.8%, D 43.8%.  66W/27B.  North of 4 and wrapping around the black-majority 6, this district has all of Pike, Dale, Barbour, and Lee, and splits Russell (just east of Columbus, GA) with 6.  Not exactly urban, but more large towns than 3 or 4.  One of the state's more Democratic districts, and yet still Safe R.

DISTRICT 6- MONTGOMERY-BLACK BELT WEST.  Pop 296,568.  O 61.9%, D 58.5%.  40W/55B.  

Love the series, and I don't mean to be nitpicky, but I've noticed here (and in many previous posts) a tendency for you to get east and west mixed up.  Just an FYI.  

Oh, jeez, that's embarrassing.  Fixed.  I've caught myself doing it a couple times, actually, but thought I had edited it out.

Please, everyone, feel free to pick any other nits you find.  I want to get the details right, and will happily go back to edit to that end.

[muon2]

Oh, hey, forgot I had this lying around for the last month.  It's an alternative Mississippi that prioritizes keeping counties together over strict population equality (though it still hews to my rule of keeping districts no more than 1 percent away from the ideal).  With a lot of rural area, a lot of small counties, and a lot of areas where county government is the smallest subdivision, doing something like this makes sense here.  Regions are still mostly kept together, and a fair ratio of black-majority districts is also maintained.  The state is almost 35 percent black by VAP, and there are three black-majority districts plus a fourth where they break 40 percent and could conceivably elect a Democrat as well.

Only two counties are split- something must be split in the Gulfport/Biloxi/Passacougla area, and I also split Madison, which has obvious CoI reasons and may be necessary as well.

I couldn't resist checking. It turns out you can keep all counties intact, stay within 1% and keep three BVAP-majority districts.

[traininthedistance]

Oh, hey, forgot I had this lying around for the last month.  It's an alternative Mississippi that prioritizes keeping counties together over strict population equality (though it still hews to my rule of keeping districts no more than 1 percent away from the ideal).  With a lot of rural area, a lot of small counties, and a lot of areas where county government is the smallest subdivision, doing something like this makes sense here.  Regions are still mostly kept together, and a fair ratio of black-majority districts is also maintained.  The state is almost 35 percent black by VAP, and there are three black-majority districts plus a fourth where they break 40 percent and could conceivably elect a Democrat as well.

Only two counties are split- something must be split in the Gulfport/Biloxi/Passacougla area, and I also split Madison, which has obvious CoI reasons and may be necessary as well.

I couldn't resist checking. It turns out you can keep all counties intact, stay within 1% and keep three BVAP-majority districts.

Very nice.  FWIW, I do think the two splits on my map are worth it, mainly for the purpose of metro area contiguity.

[Benj]

Yeah, that map is a CoI mess, as over-adherence to county lines often creates.

[muon2]

Yeah, that map is a CoI mess, as over-adherence to county lines often creates.

Actually I think that's really only true on the coast. One of the three BVAP-majority districts is generally not going to look very good, and my map would have been much neater if I was willing to take 47% for the third one. Compare the green district on my map to train's yellow district. Also the VRA would cause some split of the Jackson metro in any case.

I will agree that my three coastal districts aren't very pretty. Hancock, Harrison, and Pearl river are about 9K too small and adding Stone puts it over by 8K. One could take my light and darker blue districts down there, rearrange, and split Pearl River between them.

[traininthedistance]

I made a map for Texas, but don't feel like writing up eighty districts just yet.

So, Missourah in the meantime.  19 districts.  The state is about 11 percent black, which should mean two VRA districts; we actually have one each black-majority, black-plurality (by total population only), and minority-majority.  And, no, that is not *quite* it for the Dems, but the overall lay of the land is pretty bad for them here.  The Dem average is a little better (I think it's mostly based on Jay Nixon's win?), but Obama only won seven of these in 2008, which was basically a 50/50 race.  The downballot opportunities are probably somewhat better, and I'm fairly liberal with giving safe Romney districts a mere "lean R" rating because there could always be another Todd Akin… but still.  Going by Akin's results makes about as much sense as going by Martha Coakley, that is to say it's an extreme outer bound.

Whole-county districts are legion- I was able to get everything outside of the two big metros in whole counties, and as such there aren't a whole lot of close-ups here.  It does also mean deviations are fairly high.

The whole state:



And the St. Louis metro:



DISTRICT 1: ST. LOUIS CITY.  Pop 317,891.  O 83.7%, D 84.8%.  47W/45B.  The city of St. Louis is just a tiny bit too large for one district, so it gives a precinct to 3 and otherwise takes up all of 1.  Pity about the packing.  This district is black-plurality by total population (49 to 42).  Safe D.

DISTRICT 2: NORTH COUNTY (ST. LOUIS).  Pop 316,246.  O 77.1%, D 81.0%.  44W/51B.  The northern part of St. Louis suburbs, mostly black-majority suburbs.  In a break from what I do on the Midwest maps, I went with town and village lines in St. Louis County (and by Kansas City) despite the nominal presence of townships here; it seems like townships really don't matter and even get their boundaries changed from time to time, while towns basically cover all the boundaries we need over here.  Florissant is the largest city here.  Safe D.

DISTRICT 3: MID COUNTY (ST. LOUIS).  Pop 315,476.  O 62.9%, D 64.3%.  77W/15B.  Just west of St. Louis proper (and including one precinct for population), this is largely inner-ring suburbs, and probably counts as the state's one honest-to-goodness white liberal district.  University City is the largest town, but there are a lot of towns here and none of them are particularly large in area or population.  (There are other potential white Democratic districts, but they're all quite closer.)  Safe D.

DISTRICT 4: ST. CHARLES.  Pop 317,127.  O 45.2%, D 46.7%.  Exurban St. Charles County is a little too large for a whole district, so 4 takes most of it (including the largest cities of St. Charles and O'Fallon), and leaves a chunk in the south of the county for 5.  The lines here are crazy, so I couldn't really keep to town boundaries like I did in St. Louis County.  I'll go with Lean R rather than safe, even if it's still for now a very strong lean that would require an Akin-type to flip.  (It did flip for McCaskill in 2012, which is the main reason I refrained from a "safe" rating.)

DISTRICT 5: WEST COUNTY (ST. LOUIS)-SOUTH ST. CHARLES.  Pop 314,885.  O 41.2%, D 42.2%.  The rest of St. Charles, but mostly western St. Louis County (and some of South County, I guess).  Chesterfield and Wildwood are the largest towns in this suburban/exurban district, which is without a doubt the richest in the state.  Safe R.

DISTRICT 6: JEFFERSON-SOUTH COUNTY (ST. LOUIS).  Pop 315,841.  O 49.9% (McC 48.8%), D 55.7%.  Jefferson County and St. Louis County along the river, south of the city. The St. Louis metro worked out well county-wise, with the city and the three closest suburban counties fitting exactly six districts.  This is the sort of white working-class area that's historically been very Democratic but Romney won in 2012; some areas are gone for good but this one is sufficiently non-rural that I'd be cautious about making that pronouncement here just yet.  Tossup.

DISTRICT 7: LEAD HILLS-FRANKLIN-FARMINGTON.  Pop 317,280.  O 43.6%, D 48.1%.  Twelve counties in southeast Missouri, including the Farmington micropolitan area, two counties on the St. Louis outskirts (Washington and Franklin, the latter of which has one-third of the district's population), and a lot of rural areas.   There's some historic Dem strength from the area's mining heritage, but this has to be a secure-in-most-cases Lean R anyway.

DISTRICT 8: CAPE GIRARDEAU-POPLAR BLUFF-BOOTHEEL.  Pop 316,624.  O 34.3%, D 37.7%.  Eleven counties in the far southeast of the state; largely rural outside of Cape Girardeau.  Even Akin won this.  Safe R.

DISTRICT 9: SOUTH CENTRAL-OZARKS SALEM PLATEAU.  Pop 317,384.  O 33.6%, D 38.2%.  Ten counties, kind of an ugly shape because I'm trying to keep to whole counties.  Mostly rural, with several micropolitan centers (West Plains in the south; Lebanon, Ft. Leonard Wood, and Rolla in the north) as well as some southern Springfield suburbs.  Safe R.

DISTRICT 10:  SPRINGFIELD.  Pop 313,808.  O 40.2%, D 41.6%.  Greene and Franklin counties, over half the population is in Springfield proper.  It would have been nice to pair Greene with another county technically in the metro, but rural Franklin was the only one with the right population.  Claire McCaskill actually won this by about 1500 votes but who are we kidding.  Safe R.

DISTRICT 11: JOPLIN-BRANSON-SOUTHWEST OZARKS.  Pop 318,075.  O 31.3%, D 31.7%.  Six counties in the southwest corner, the other southern Missouri district which is mostly urbanized.  The Joplin metro takes up most of the population here, but you've also got Branson (which is actually much smaller than its fame) and some Fayetteville, Arkansas exurbs too.  Safe R.

Might as well show Kansas City and the I-70 Corridor here.



DISTRICT 12: OSAGE LAKES-NORTH OZARKS.  Pop 313,080.  O 35.6%,  43.8%.  Thirteen whole counties, and most of Cass except for the towns of Belton and Raymore- so it is actually part of the Kansas City county grouping, which had to include a rural district unlike St. Louis.  Mostly rural, with the geographic heart of the district around the several manmade reservoirs in the Osage Plains, but it also includes northern Springfield outskirts, southern KC outskirts, and Western Missouri towns such as Sedalia.  While McCaskill may have won it (I can't tell from a cursory glance), the Ike Skelton days are over and I feel pretty safe calling it Safe R.

DISTRICT 13: JEFFERSON CITY-CENTRAL.  Pop 312,504.  O 36.1%, D 42.7%.  Twelve counties; all of the Jefferson City metro and everything else is basically rural (though the Census does count Mexico as micropolitan).  Safe R.

DISTRICT 14:  COLUMBIA-MISSOURI RIVER.  Pop 312,235.  O 49.9% (McC 48.4%), D 54.2%.  Seven counties almost all along the river, most of the population lives in Boone (Columbia).  Warrensburg and Marshall are the secondary population centers.  Barely won by Obama in '08.  The numbers are almost equivalent to 6, but for whatever reason I feel like this district falls just on the Lean R side of things, rather than the tossup I rated the other one.

DISTRICT 15: MARK TWAIN-NORTH AND NORTHEAST.  Pop 317,343.  O 40.5%, D 44.2%.  Twenty-two counties.  Despite technically including two outlying counties in the St. Louis metro, this is easily the most rural district in the state; its largest population centers appear to be the sub-20K Kirksville and Hannibal.  Safe R… probably.

DISTRICT 16: NORTHWEST-ST. JOSEPH-PLATTE.  Pop 314,538.  O 44.5%, D 46.7%.  Thirteen counties in the northwest.  Some rural areas, the St. Joseph area, and a significant amount of Kansas City countryside.  Platte County, in the southwest of the district, actually contains part of KC proper, but most of that is the airport.  Another Lean R that's probably secure for most non-Akin candidates in most years.

DISTRICT 17: CLAY-NORTH KANSAS CITY.  Pop 313,555.  O 56.0%, D 57.7%.  You can't keep Kansas City together as neatly as you can for St. Louis, because it spans four counties, with a couple enclaves, so I didn't sweat the splitting of KC that much- no other towns are split around here at least.  This district is all of Clay County (which includes much of the city) and the northern bit of Jackson, with a strip running down the Missouri-Kansas border in Kansas City to help 18 be minority-majority.  Most of KC is actually in 18, but technically downtown is here.  Lean D.

DISTRICT 18: KANSAS CITY CENTRAL AND SOUTH.  Pop 312,478.  O 72.7%, D 74.8%.  47W/40B.  Almost black-plurality by total population, but not quite.  It's almost entirely KC, with a few adjoining towns- Grandview and Raytown in Jackson, Belton and Raymore in Cass. Safe D.

DISTRICT 19: INDEPENDENCE-EAST JACKSON.  Pop 312,557.  O 46.7%, D 50.6%.  Entirely within Jackson, dominated by Independence but including several other suburbs and a small part of KC as well.  Lean R.

One black-majority (2), one black-plurality but only by total population (1), one min-maj (18).

4 Safe D
1 Lean D
2 Tossup
4 Lean R
8 Safe R

[traininthedistance]

Just want to say that I'll get back to this eventually.  Burnout plus being busier in real life means this project has gone on the back burner, but I still intend to finish it.  Um... before the year is out, I hope.

[○∙◄☻¥tπ[╪AV┼cVê└]

I'm fairly liberal with giving safe Romney districts a mere "lean R" rating because there could always be another Todd Akin…

Akin won 6 House elections by large margins because people don't realize how crazy their representative is. They pay more attention to Senate elections.