Georgia 2020 Redistricting Discussion

[Bacon King]

Here's a quick map of county-level population change since the last census that I made in order to visualize how it will affect redistricting. In particular the impact of these population shifts will be greatly amplified on account of the bizarre fact that in spite of being on track to gain a million people by the end of the decade we somehow won't be awarded another seat in the House.

I couldn't find county data for the 2019 estimates so the estimates are a year old but the trends are clear regardless.

Click to embiggen

[Bacon King]

also besides the obvious implications for Congressional redistricting, note that each State House district would have a population of ~58,442 using this estimate

among many other observations to be made, note that the metro counties that voted for Hillary are collectively gaining like eight house seats here

[Bacon King]

TLDR for this thread: The GAGOP can draw 10 Safe R districts. There's no incentive for them not to. The only remaining question is whether it's Bishop that gets cut and McBath packed, or vice versa.

There, are we done?

I would absolutely love to see the GOP attempt a 10-4 gerrymander, because it would absolutely be a dummymander that would completely backfire in their faces.

"GA GOP attempting to draw a 10R-4D map in 2020" would easily overtake "AR Dems attempting to draw a 3D-1R map in 2010" as the worst ever self-inflicted redistricting failure

[Bacon King]

Bishop's seat isn't vra protected, but a 4th aa district should be drawn in atl to preempt any challenges.

1. where is this idea coming from that the 2nd District isn't protected by the VRA? Because it absolutely is.

2. That is not at all how the Voting Rights Act works, for so many different reasons

[Bacon King]

It was created as a D vote sink by the legislature, not court order, and it is not majority black.  But in the case a court interprets the vra to require a 4th aa seat in Georgia, a 4th aa district in Atlanta would satisfy that.  A 10-4 map is doable, as long as all 4 d seats are in Atlanta.  pack all of the bluest precincts into 4 districts, then crack any blue/purple areas left. 

Respectfully, you are incredibly misinformed about the Voting Rights Act. The stated intent behind its creation is irrelevant, whether it was created by the legislature or the judiciary is irrelevant, and whether the district itself is majority black is irrelevant (although, for the record, it is absolutely majority black). What matters is that it passes all three elements of the Gingles test. That alone makes it a protected minority district and therefore its elimination would be an unconstitutional retrogression of minority representation

In addition, there's literally no example anywhere in VRA case law that works the way you're describing. Minority districts aren't suddenly required just because of the minority population of the entire state is above some arbitrary threshold (if it did, for example, Louisiana would absolutely warrant a second black-majority district) and you can't just remove a minority district from one portion of a state and add one to another part of the state if the minority group in the former area still passes the Gingles test. It's not about the racial breakdown of the state as a whole, it's about specific communities of specific minority groups. The most obvious example here is Miller v. Johnson, which directly and explicitly says that for the purposes of the Voting Rights Act, black people in Atlanta aren't members of the same minority community as blacks downstate (in the case itself, Augusta + Savannah). They're different communities of interest, so you can't remove a district from one and give it to the other if the group you're taking the district from still warrants its own district

[Bacon King]

TLDR for this thread: The GAGOP can draw 10 Safe R districts. There's no incentive for them not to. The only remaining question is whether it's Bishop that gets cut and McBath packed, or vice versa.

There, are we done?

I would absolutely love to see the GOP attempt a 10-4 gerrymander, because it would absolutely be a dummymander that would completely backfire in their faces.

"GA GOP attempting to draw a 10R-4D map in 2020" would easily overtake "AR Dems attempting to draw a 3D-1R map in 2010" as the worst ever self-inflicted redistricting failure

https://twitter.com/CARepublicanMap/status/1205669677482930177?s=20

This is incredibly stupid as a take. The Arkansas D districts were literally McCain-Bush-Bush districts, not Trump+28

This map would never be implemented, most obviously because District 1 would cause a long and expensive legal challenge and almost certainly be found unconstitutional. And still, the 3rd district would be at risk almost immediately and the 8th, 9th, 11th, and 12th would all be at risk of falling before the end of the decade.

Though you're right, perhaps I am understating the stupidity of Arkansas Democrats. Instead let's call it the "worst Georgia redistricting plan since the Roy Barnes 2002 monstrosity"

[Bacon King]

While much of this is inarguably true, there remains the fact the GOP could break apart GA-02, at least if Adam Griffin's word is anything to come by. He's told me previously that GA-02 was created mainly to prevent a Democrat from getting elected in Southern Georgia, and packing D voters in the region was a good way of limiting their potential.

again, regardless of the intent behind its creation, it's still protected by the Voting Rights Act regardless.

Though I'd be curious to hear Adam's explanation there? Because I've not heard that before and it doesn't seem right although i could very well be wrong

[Bacon King]

Bishop's seat isn't vra protected, but a 4th aa district should be drawn in atl to preempt any challenges.

1. where is this idea coming from that the 2nd District isn't protected by the VRA? Because it absolutely is.

2. That is not at all how the Voting Rights Act works, for so many different reasons

It was created as a D vote sink by the legislature, not court order, and it is not majority black.  But in the case a court interprets the vra to require a 4th aa seat in Georgia, a 4th aa district in Atlanta would satisfy that.  A 10-4 map is doable, as long as all 4 d seats are in Atlanta.  pack all of the bluest precincts into 4 districts, then crack any blue/purple areas left. 

So, leaving aside whether or not this would be VRA compatible, remember this map is supposed to hold until 2030. Is the GA GOP really reckless enough to assume that this map you're proposing will hold? It'd involve a lot of cracking of the ATL suburbs just so that whatever can't be fed into the D districts are split up, but those suburbs are growing at lightspeed.

this map holds until 2030: https://davesredistricting.org/join/27ac753b-33e0-40f1-8acc-6fb947deb3ee
if they wanna be even more careful, they could do this: https://davesredistricting.org/join/781485f1-efa6-4aad-b355-2976a15a9f3b
but a 10-4 is very possible, if done carefully.  we're talking precinct-by-precinct precision and no regard for anything other than partisanship.  It's ugly, but very doable.

I've already explained why these maps wouldn't be constitutional but I want to compliment you here because these are very skillfully drawn maps! Would you happen to have the 2016 results for these districts?

[Bacon King]

Well then, don't eliminate Bishop's seat, just move a lot of it into Atlanta, keeping a bit in the black belt.

Respectfully, did you even read what you're responding to?

As I literally just said, Miller v. Johnson specifically ruled that urban Atlanta and the rural black belt are not the same community of interest, and therefore a district that stretches to combine the two is an unconstitutional racial gerrymander. Like Brittain33 said, if you do that then it stops representing the will of the district's rural voters even though the nominally remain represented by it.

There is good justification for this, black populations are shrinking in rural areas but growing in Atlanta.

Using the most recent available local level census estimates, it remains trivially easy to draw a compact and normally-shaped majority black district in SW GA -- in fact, it's even easier to draw now than it was during the 2010 redistricting! I drew a 51% black district using 2016 estimate only splitting one county.

Because such a district can be drawn, the conditions of the Gingles test are still met, therefore it's constitutionally mandatory to draw a district that can be reasonably expected to elect that specific minority community's candidate of choice.

In the long term, Republicans should get try to get the court to make a ruling where it's unconstitutional for map makers to even access racial demographic data.  Redistricting should be color blind.  They already struck down part of the VRA.

That would be an insane ruling and contrary even to basic expectations of judicial precedent.

Only a single element of the Voting Rights Act has actually been overturned: the specific formula to determine which jurisdictions are required to do preclearance, and that was only made unconstitutional because the formula was based on data that was over half a century out of date. Literally the only reason the VRA has become less prominent since Shelby County v Holder is because Congress has thus far declined to pass any new/updated formula. Preclearance will almost certainly be back the next time Democrats control both chambers of Congress, and it will be perfectly constitutional when they bring it back.

The evidence for why this should happen is the fact majority white electorates elected blacks like Tim Scott, Antonio Delgado, and Mia Love as well as Hispanics like Ted Cruz and Marco Rubio.  Majority minority districts simply aren't necessary for minorities to be elected anymore. 

EastAnglianLefty is correct here.

Tim Scott is black but he is not the candidate of choice of any black community. Likewise, Steve Cohen is the candidate of choice of the black community of Memphis even though he's white.



edit: this got posted after I started typing

However one can make the argument that the belt no longer has the concentration of AA voters needed to have an opportunity district. I tend to disagree, but it is an argument that can be made if the GOP moves against Bishop. The counties down here are not flipping red because AA voters are now considering the GOP, they.are flipping because the AA pop is shrinking faster than the white pop. We are only a few years away from the Georgian belt being just the urban counties and a straggler or two. A district that goes from Albany to Augusta isn't compact  But the AA pop is concentrating hard into the Atlanta metro. Since the state AA pop is overall growing, a fact magnified by the shrinking belt, that is the ideal place for a new AA seat.

As I said above, it's still very easy to draw a black-majority district covering Albany, Macon, and Columbus.  The African American population is declining in that area, sure, but the white population is shrinking just as quickly.

Also there's definitely not enough population to warrant a fourth black majority seat in the Atlanta metro, attempting to force one would be an unconstitutional racial gerrymander

[Bacon King]

Only thing I have to say to this is that you don't know the demo's of your state . I didn't even try that hard and I got four AA seats in Atlanta between 49% and 52% AA, more than enough to lock in the candidate of choice.

Note that I didn't say it couldn't be done; rather, I expressed my doubt as to its constitutionality. Admittedly these districts are less irregularly-shaped than I had imagined they would be, although it's worth noting that pre-2010 census estimates overestimated the black share of the population growth throughout the metro area, and I remain suspicious the same phenomenon is happening again ten years later. In fact, you could look ten years back on this very board and find posts I made at the time showing how easy it would be to draw four black majority districts in ATL, much as you're doing now

and regardless, even if the estimates are accurate and your map works per the actual 2020 census data, my gut still tells me that the courts could still possibly interpret it as an unconstitutional racial gerrymander - although such a concern is completely moot

[Bacon King]

Racial gerrymandering isn't needed tho.  As a Republican, I don't care if Dem districts are represented by whites or non whites, as long as the number of Dem districts is the lowest it can be without threatening any R seats.  Why would a red legislature use racial data and risk it when partisan data will get you better maps? 

I don't see any point in responding to the rest of your absurd and bad posts itt when other people are doing great work calling you out, but I'm obligated to point out that this isn't how gerrymandering works, in theory or in practice. The act of maximizing the number of Republican seats, would quite obviously also suppress minority representation. In practice it's not about "if the democratic districts are white or black" it's that a gerrymander maximizing the number of GOP districts in the deep south is obviously also going to be a gerrymander that maximizes the number of white districts at the expense of minority representation. And that's still the case even if the people drawing the map had super duper pinky promised for real that "they were only looking at partisan data and definitely weren't looking at racial data at all, it just happened to end up like that i swear"

[Bacon King]

Since some mentioned GA-02 (if kept with a similar composition) being a swing district by the end of the decade despite being a VRA plurality/majority black district, I now wonder if something similar has ever happened.

Has a white Republican (with ~10% black support presumably) ever won a VRA black district? GA-02 seems like it would be extremely inelastic and titanium D despite the low margins

replace "black" with "hispanic" and you're basically describing TX-23. Will Hurd is a non-Hispanic Republican who has narrowly won three times against Hispanic Democrats even though the district is 68% Hispanic

Also with GA-2 specifically, the scenario you're describing actually almost happened in 2010, when Sanford Bishop held on by less than five thousand votes. I doubt it would happen in the upcoming decade, though, because honestly SW GA's white population is shrinking just as quickly as the black population.

[Bacon King]

Well as it currently stands republicans can't draw away the black district in a state like AL, but if that part of the VRA is reinterpreted, there's little that could be done to protect those districts, since it could be argued the motivation us partisan.  Your pompous attitude isn't necessary, if your points held any water, you could stick with those.

I apologize if I seem pompous at all to you, but your posts betray the fact that you are very poorly informed about the topic you're attempting to speak authoritatively about, and when I've attempted to explain it to you, instead of listening you are just doubling down and insisting your incorrect interpretation is the truth. It has honestly become incredibly frustrating.

Let me try to speak clearly here. Attempting to argue intent and motivation is meaningless because US Voting Rights law does not work that way. If redistricting creates an illegal racial gerrymander, then it is an illegal racial gerrymander because all that matters is OUTCOME.regardless of intent. And this isn't something the court can magically reinterpret anyway because it's the explicit word of the law! The Voting Rights Act of 1965 (as amended in 1982), Section 2, "Results Test". Look it up, if you want to keep insisting I'm wrong the burden of explaining yourself is on you.

But it passes the gingles test.  It is large enough and compact enough to make one maybe even 2 SMDs.  It is politically cohesive, OC whites consistently vote Republican.  As to the 3rd part, that's questionable.  Do non whites consistently vote to defeat the white candidate of choice?  That depends, they did in 2018.  If it keeps happening over and over then there would be a good case, it hinges on that. 

No, this is simply not true at all. Even if you assume whites were suddenly treated as a protected racial minority (they aren't, and shouldn't be) the white community of Orange County is absolutely not politically cohesive nor is there any racial bloc voting

[Bacon King]

instead of derailing this thread further what would you guys say to a thread on this board specifically about the Voting Rights Act and its impact on redistricting

[Bacon King]

submitting this map as a proof of concept to demonstrate Sanford Bishop's seat is protected by the VRA

[Bacon King]

Let me offer an explanation for the significance of the above, for anyone unaware

The Supreme Court's unanimous opinion in Gingles v. Thornburg established an important precedent governing the influence of the Voting Rights Act on redistricting.

The "Gingles Test" establishes when a minority district is required. It has three parts:

1. The minority group must be "sufficiently large and geographically compact to constitute a majority in a single-member district"
2. The minority group must vote as a bloc
3. The majority votes sufficiently as a bloc to defeat the minority's preferred candidate

When these three elements are present, then a district must exist that reliably elects the minority's candidate of choice. The court's judgement says that discriminatory intent and causation are both completely irrelevant here, only the test itself matters. If polarized voting exists and a Gingles district can exist for a minority community, then there must be a district that can reliably elect that community's preferred candidate.

Because it's obvious that racially polarized voting occurs in the Deep South, we only need to focus on the first part of the Gingles Test, the compactness test:

Is the specific minority community populous enough and geographically compact enough that its members can constitute a majority of a district's voting age population?

This is called the "compactness test" for a reason. The sprawling 11th District here was ruled unconstitutional because it was not at all compact and the African American population of the district was not part of any coherent community. People living in urban Dekalb County don't have the same interests and priorities as people in the rural black belt or downtown Savannah, even if they're both in the same racial minority. In most circumstances though the Courts have not been very strict about the definitions of "compactness" or "community".

If the General Assembly gets rid of Sanford Bishop's seat, the state's most likely case in the ensuing lawsuit will probably be to push for a new and stricter definition of minority community here. It will probably not be successful because existing precedent covers a much broader definition but it would be the only real argument they could make.

All things considered "African Americans in southwest Georgia" can be reasonably considered a discrete minority community. As you can see, I've drawn a compact district in the area of the current 2nd District with a 50.5% black voting age population. This means a district must exist that can be expected to elect their candidate of choice (who is currently Sanford Bishop). Keep in mind the district itself does not need to be >50% black! It can certainly be less, as long as it will probably still elect the black community's candidate of choice.

Even though it's not strictly necessary, traditionally when drawing VRA districts the mapmakers have typically drawn the districts themselves to be >50% minority, either because they had no pressing reason to do otherwise, or because they deemed it necessary to ensure the election of the minority's candidate of choice

[Bacon King]

After looking at thethem census data for a bit i think the most likely outcome is a 5D, 9R map. Dem districts will be

1) Bishop's seat
2 thru 4) three metro VRA districts in a north-to-south bacon strip configuration. The southern portion of all 3 districts will cover the supermajority black portions of Fulton/Clayton/Dekalb, the northern portion of each district stretching up to the non-black Dem portions of Cobb/North Fulton/Dekalb and possibly a bit into Gwinnett as well. Basically the goal will be for all three districts to be something in the range of 50% to 53% African American VAP (not more, in order to prevent VRA lawsuits over minority vote packing) while being like >80% Dem so they're optimized as a partisan gerrymander.
5) non VRA dem seat probably based in Gwinnett, packing in all the ATL dem voters not in any of the previous three districts

If these explanations don't make sense I'll try to make a map at some point

[Bacon King]

okay, here's my incomplete first draft of what I expect to see from redistricting (please read the notes below the map before you respond!)

click on map to embiggen



Notes

A. Completely disregard all GOP districts except for 6 and 9 (i.e., ignore districts 1, 3, 8, 10, 11, 12, 14) because they do not take incumbency or any other factors into account. They were only drawn to fill out the rest of the map after designing the districts where race and/or partisanship is actually relevant.

B. For the reasons I explained earlier in the thread, I am fairly confident the new 2nd District will be very similar to what I've drawn here, with only minor differences around its border at most. Most notably I could see the General Assembly putting more of northeast Houston County into the 8th. Doing so would put the entirety of Warner Robins Air Force Base into Austin Scott's district, which could improve his standing on the House Armed Services Committee. They did something similar last decade - the current 1st District includes a slice of Lowndes County because they wanted Moody AFB represented by Jack Kingston

C. In my map, the 4th, 5th, and 13th districts remain VRA districts. Their bacon-strip design elegantly allows them to remain optimally packed full of Democrats without excessively packing them with African American voters (which would be an unconstitutional racial gerrymander). As a result the composition of all three districts is something like 51% black, 49% white liberals + hispanics. The current State House map heavily relies on this strategy in Fulton/DeKalb:



D. I'm much less certain of the 7th District's eventual layout, but I believe the legislature will deem a Gwinnett-based Dem sink necessary to ensure every Republican district remains safe all the way to 2030. If they decide on an extremely aggressive map this will be the district they chop apart, but thus far I see no indication that the GA Republicans will abandon the relatively cautious approach that has characterized their gerrymanders since they took over the state government two decades ago.

E. The 6th and 9th districts won't necessarily look exactly how I've drawn it, but the general principle of the design is to take the left-trending northern suburbs and shore them up with the exurbs further northwards that form the new rapidly-growing core of the Georgia Republican party. North Fulton and North Gwinnett are already becoming Dem-friendly, and Forsyth County will join them by the end of the decade. I expect these regions to be paired with portions of Atlanta's exurban belt: Hall and Cherokee Counties, certainly. Quite possibly Bartow County as well

I'll post the statistics for these districts in the morning, might add additional comments as well.

Questions and feedback are welcome!

[Bacon King]

statistics and close-up looks at the seven districts I put effort into:

[Bacon King]

statistics and close-up looks at the seven districts I put effort into:

So the 6th went from R +26.2 to +15.3 from 2016-20, and the 9th from R +35.8 to +22.5.

Yeah, that’s a definite dummymander, those two would go blue by the middle of the decade making a 7-7 map. And if the 1st and 12th flip it becomes 9-5.

this is still a first draft - i definitely believe both districts will be safer than what I've drawn above. In my second draft I expect the 6th will include Bartow, the 9th will include Jackson and Barrow, and Forsyth will probably be included in the 14th (though this depends on how exactly the numbers work)

In general though, be wary before you take four years of trends and assume they'll extrapolate linearly for another decade. The current ongoing trends will certainly affect Forsyth County but I'd be very hesitant to assume it will continue spreading further into suburbia (or reach the exurbs at all)

[Bacon King]

I gave my initial thoughts on the Senate Draft to Adam Griffin and a few others on discord yesterday, but now I finally have time to sit down and take a good look at it so I'll offer my thoughts here as well.

Frankly I have no idea what the State Senate is thinking. Honestly it's as if an intern accidentally attached the wrong image to the press release and what we're all looking at is actually just a discarded sketch from early in the design process

They literally didn't even take incumbent residency into account! When I first saw that Bordeaux lives in Clyde's district and McBath lives in Loudermilk's district I assumed the committee was just being petty, or perhaps just didn't care about Democratic incumbents. But no, it happened to Republicans too: Clyde now lives in the district of Jody Hice while Loudermilk is represented by MTG!

This map honestly looks closer to something I'd expect to see from an independent commission, not a creation of Republican officeholders. Honestly this map would only need a few relatively minor changes to become something I'd genuinely consider fair (mostly just keeping the suburbs and exurbs in different districts)


Honestly the only part of this map that actually appears to be an effective GOP gerrymander is, surprisingly, the 2nd District. It's designed so it's juuuust safe enough to reliably re-elect Sanford Bishop, thereby adhering to VRA case law, but when Bishop retires it becomes a tossup and definitely winnable by a Republican. From a strictly technical point of view it's honestly kind of clever: they've managed to create a mandatory black VRA district that will probably elect a Republican in at least 1 of the next 5 elections with little-to-no risk from a lawsuit against it

[Bacon King]

They didn't baconstrip Appalachia.

Indeed. As I said earlier in this thread, the GA GOP has never done that sort of thing and they've not really given any indication that they're going to start now

(not that they've ever done... whatever the hell this new congressional map is, either)

Wait... Lt. Gov Duncan and Senator WHAT?

The fact that we have our very own SENATOR JOHN F. KENNEDY never ceases to amuse me

I remember when all the Conservative ET kids said they would dismantle GA-02.

While that was obviously never going to happen, this subtle gerrymander they're doing with the 2nd district is even worse because it could actually hold up in court

So why did they give Bordeaux a safe seat but not McBath? I guess the Republican logic is that they don't like having a black woman represent a white traditionally Republican district like GA-06. Whereas at least Bordeaux is white, so from their perspective she is more tolerable. But if voters in GA-06 get too used to voting for McBath, next thing you know they'll be voting for Stacey Abrams as well, and can't have all those black women running around with so much power.

IMO attributing it to race is very much overthinking it - the "Republican logic" here is nothing more than geographic necessity.

I explain it in more detail in my map prediction post earlier in this thread, but the tl;dr is that without bacon-stripping Appalachia, basically any GOP-friendly congressional map will require a non-VRA Dem vote sink in Gwinnett County. They didn't decide to favor Boudreaux over McBath because it wasn't even a choice in the first place - Boudreaux's safe seat was pretty much inevitable.

Basically the only options on the table are as follows:

a) 3 Dem seats in the metro. Neither Bourdeaux nor McBath get a district, GOP wins back both seats by drawing a bunch of snakes that extend from the suburbs to the mountains

b) 4 Dem seats in the metro. A fourth Dem vote sink joins the three black-majority VRA seats. Gwinnett is too big and racially diverse to contain a VRA district so it gets the new seat. Bourdeaux gets it since her current district is Gwinnett-based

c) 5 Dem seats in the metro. Both sophomore Democrats get safe seats. Republicans permanently surrender both seats but keep every incumbent safe for the decade.

Giving a district to McBath but not Bourdeaux would require them to draw some weird districts. If they were willing to do that the question would be moot, because they would then have no reservations drawing districts that got rid of both of them

[Bacon King]

This map honestly looks closer to something I'd expect to see from an independent commission, not a creation of Republican officeholders. Honestly this map would only need a few relatively minor changes to become something I'd genuinely consider fair (mostly just keeping the suburbs and exurbs in different districts)

behold! this bad failomander has evolved into... A FAIR MAP?

(click either image to embiggen and view in DRA)







edit: partisan lean map

[Bacon King]

Are you serious? Their idea of drawing out one of McBath or Bourdeaux involves creating a lean Republican district that Biden might win anyway in 2024? That’s an own goal.

But wait, there's more! They also left Loudermilk with a more vulnerable district lmao

[Bacon King]

Can't wait for 2031 after GOP judges completely dismantle Sections 2 & 5 of the VRA and we get a 9D-5R (or 10D-5R) map!

01: Biden +10
02: Biden +12
03: Trump +33
04: Trump +29
05: Trump +35
06: Trump +57
07: Trump +52
08: Trump +4
09: Trump +20
10: Biden +25
11: Biden +36
12: Biden +45
13: Biden +43
14: Biden +41

If they're throwing out the VRA we could imo definitely net a couple more seats by unpacking ATL