Georgia 2020 Redistricting Discussion

[AustralianSwingVoter]

https://davesredistricting.org/join/f320c87d-0c0f-4ee8-ae06-bd4eb4b95aed
A Georgian Experiment. The Second helps pack Atlanta while still being a SW VRA seat.
1 - R+8.62, Trump+15.1
2 - D+15.93, Clinton+35.2, (51.3% Black)
3 - R+16.36, Trump+27.5
4 - D+33.32, Clinton+70.6, (54.0% Black)
5 - D+31.47, Clinton+67.4, (54.3% Black)
6 - R+20.88, Tump+29.2
7 - R+21.92, Trump+41.8
8 - R+9.38, Trump+18.1
9 - R+20.95, Trump+33.5
10 - R+18.16, Trump+33.7
11 - R+19.72, Trump+29.3
12 - R+9.03, Trump+18.4
13 - D+24.20, Clinton+52.9, (54.2% Black)
14 - R+18.34, Trump+30.5

[AustralianSwingVoter]

https://davesredistricting.org/join/f320c87d-0c0f-4ee8-ae06-bd4eb4b95aed
A Georgian Experiment. The Second helps pack Atlanta while still being a SW VRA seat.
1 - R+8.62, Trump+15.1
2 - D+15.93, Clinton+35.2, (51.3% Black)
3 - R+16.36, Trump+27.5
4 - D+33.32, Clinton+70.6, (54.0% Black)
5 - D+31.47, Clinton+67.4, (54.3% Black)
6 - R+20.88, Tump+29.2
7 - R+21.92, Trump+41.8
8 - R+9.38, Trump+18.1
9 - R+20.95, Trump+33.5
10 - R+18.16, Trump+33.7
11 - R+19.72, Trump+29.3
12 - R+9.03, Trump+18.4
13 - D+24.20, Clinton+52.9, (54.2% Black)
14 - R+18.34, Trump+30.5

That GA-02 is definitely going to fail the 2012 NC-01/VA-03 test, and while the VA-03 case reached SCOTTUS during the 4-4 deadlock in 2016, Roberts concurred in striking down the 2012 version of NC-01 as a racial gerrymander.  The better bet for getting a 10-4 map upheld should be just explicitly moving the 4th VRA seat to Atlanta than trying to connect the current GA-02 to it. 

VA-3 was about 57% Black and qualified as a Black pack.
Meanwhile my GA-02 is actually (slightly) less black than the current GA-02, so there's absolutely no reason to think the courts would strike it down. The arm to Atlanta is a partisan gerrymander, taking in white suburban Dems in Gwinett, not a racial gerrymander packing black voters.

[AustralianSwingVoter]

https://davesredistricting.org/join/d6cf2d99-f0ed-477f-be41-016398a086eb
1 - R+9.20, 56.7% Trump
2 - D+14.52, 63.5% Clinton, (56.9% Black)
3 - R+17.01, 63.3% Trump
4 - D+30.84, 81.4% Clinton, (56.5% Black)
5 - D+33.93, 84.6% Clinton, (53.0% Black)
6 - R+18.18, 63.5% Tump
7 - R+20.25, 62.2% Trump
8 - R+13.9, 62.7% Trump
9 - R+19.89, 62.7% Trump
10 - R+18.57, 63.3 Trump
11 - R+19.98, 63.1% Trump
12 - R+8.99, 57.7% Trump
13 - D+28.20, 79.0% Clinton, (57.8% Black)
14 - R+20.42, 68.0% Trump

If the Georgia GOP decides to stick with the current arrangement of D packs rather than moving the 2nd to Atlanta and hoping the Courts don't complain, one would imagine something like this would suffice as a rock solid gerrymander. And in case anyone complains about VRA, all 4 seats are less black than the current 3 Atlanta seats, all of which are around 60% black.

[AustralianSwingVoter]

Given how no one has drawn a white Dem district (is one even possible?); the number of Dem seats equals the number of black seats

So now the question is:

1: Can GA-02 be dismantled? Can it be brought to Atlanta or other black areas of the state?
2: How many black majority districts are needed in the Atlanta area?

Once you answer those 2 questions you can start drawing. The VRA essentially means Georgia will have a court designed map in practice as 11-3 is impossible.

You can very easily draw 2 white dem seats in Atlanta, it's just that there's literally no reason the GOP would possibly draw them.

[AustralianSwingVoter]

Given how no one has drawn a white Dem district (is one even possible?); the number of Dem seats equals the number of black seats

So now the question is:

1: Can GA-02 be dismantled? Can it be brought to Atlanta or other black areas of the state?
2: How many black majority districts are needed in the Atlanta area?

Once you answer those 2 questions you can start drawing. The VRA essentially means Georgia will have a court designed map in practice as 11-3 is impossible.

You can very easily draw 2 white dem seats in Atlanta, it's just that there's literally no reason the GOP would possibly draw them.

Wouldn't such districts be illegal as they would require packing the black voters?

Like if you have a white Dem district in Atlanta that means the remaining black seats will have to be 60%+ black; which would be struck down by courts?

So I guess while easily possible, not only has the GOP no reason to draw them but also they would be struck down anyways

All 3 current Atlanta districts are 58-60% black, for reference. And the courts haven't complained.

[AustralianSwingVoter]

https://davesredistricting.org/join/2d5fdc7d-5ed8-4c5c-9d27-c438f9ca43b0

I've been tinkering with this for a while to see what the GOP could do if they really don't want to concede a 4th pack in Atlanta. It's just as VRA compliant as the current map (and the very Trumpy 5th circuit shouldn't find any problems no matter what). All 10 republican seats should be totally rock-solid for the entire decade, and they would have decent odds of picking up the 2nd. Kemp won all the Atlanta suburban seats by at least 25% so I really doubt they'll be marginal in 10 years time.


GA-01: Buddy Carter's seat is basically unchanged, still stretching along the Georgia Coast. Trends still mildly favour the Republicans in this seat going forward.

GA-02: Sandford Bishop's seat becomes 0.1% more Black than its current iteration while the PVI halves. Abrams' margin of victory goes from 12% to 5%. Bishop still has some crossover appeal and outran Biden by 6%, but when he retires this seat is a ripe pickup opportunity. And it's just as VRA compliant as the current district.

GA-03: Drew Ferguson's seat loses it's eastern arm out to Henry and instead stretches further north up to Cobb. However the character of the district barely changes with the small slice of blueing Atlanta suburbs easily outweighed by rural West Central Georgia.

GA-04: Hank Johnson's seat remains based in DeKalb though it gains a significant amount of deep blue territory in Gwinnett.

GA-05: Still based in Atlanta, John Lewis's old seat loses most of Clayton for more of DeKalb. This seat is Nikema Williams's for as long as she wants it.

GA-06: Marjorie Taylor Greene is the big winner under this map, with her seat now stretching from Dalton, her base in the primary and the runoff, to take in her old home town of Alpharetta in suburban Fulton County. And with Kemp winning this seat by 27% she should have no problems in the general. Northern Fulton and Forsyth are both zooming leftwards however Gainesville and Appalachia completely overpower them.

GA-07: Carolyn Bordeaux is the one big loser in this map with absolutely nowhere to go. The radically redrawn 7th is now R+21 PVI and voted for Kemp by 27%. A wide-open seat it should see a divisive primary between a Gainesville-based Republican and a Gwinnett-based Republican but whoever wins shouldn't have to sweat about winning election at any point in the next decade.

GA-08: Austin Scott's district snakes through Central Georgia up to Athens and drops 6 PVI points in the process. It wouldn't be the snoozefest he's accustomed to but Scott shouldn't have any problem holding this seat.

GA-09: Freshman Andrew Clyde's Northeastern seat is rather radically altered and he now has to deal with a significant democratic constituency with Kemp's margin of victory falls from 59% to 26%. Gwinnett may be zooming leftwards but this is still the seat where Deliverance was filmed.

GA-10: Jody Hice's district loses the rural eastern counties and instead wraps around west to Henry and Fayette and south to Macon. Not quite as safe as the Northern seats however Kemp still won it by 24%.

GA-11: Barry Loudermilk's seat keeps its basic shape but loses about half of Cobb in exchange for taking two more deep red Northern counties. It moves 4 PVI points right and Kemp won it by 27%, so Loudermilk should have no problem holding this seat.

GA-12: Still anchored by Augusta Rick Allen's seat has mild changes that don't affect the partisan balance. Trends should only strengthen the Republican hold on this seat.

GA-13: Sets up a very nasty primary between Lucy McBath and David Scott in this Rorschach blot (which is still less ugly than MD-03). David Scott is a noted DINO who was almost forced into a runoff against an unfunded nobody this cycle, meanwhile McBath is hindered by currently representing exactly 0.7% of this district, including herself. A dark horse in the vein of Keisha Waites could even come through the middle and defeat both of them.

GA-14: Though it's the successor to the current 14th MTG should move to the 6th making this one of two Open seats. It keeps its basic shape taking in Northwestern Georgia however it loses MTG's base of Dalton for a large slice of rapidly diversifying Cobb. Whoever wins the Republican primary should have no problem holding this seat, given Kemp won it by 26%.

[AustralianSwingVoter]

Welcome to the redistrictingverse!

FYI your districts are a little too wide in terms of population deviation--ideally you try to get that number as close to 0 as possible, though it can be tough, especially in Georgia where there are dumb huge precincts. Personally I shoot for more like -/+1000.

DRA also has 2018 data, so you can use more up to date info.

That's good to know, thank you!

You are allowed deviation up to 1% of the overall quota. The gap between largest and smallest district cannot exceed that.

That's simply incorrect.
The Supreme Court decided in Karcher v. Daggett to overturn NJ's redistricting plan with a variance of only 0.7% for failing to adhere to one man, one vote and then adopted a plan with a variance of only 0.01%. Meanwhile in Tennant v. Jefferson County Commission the Supreme Court upheld a variance of 0.8% only because no other plan would satisfy the state's requirement of no county splits.
The clear precedent of the supreme court is that the state must adopt a plan with the smallest population variation possible unless there is an overriding interest such as requiring county integrity enshrined in state law, which is why the only state to have a congressional variance higher than 0.1% is West Virginia due to its constitutional requirement for whole counties.