Georgia 2020 Redistricting Discussion

[lfromnj]

Anyone else working on State Senate maps?
(Reminder you can split precincts now)

The ability to splint precincts is so cool, but I noticed when I was doing some maps for fun that the President data isn’t accurate yet (atleast for South Carolina), and that it just assumed that every block voted the same as the precinct

I mean I don't see how they can get data by block. The main reason for it is just to reduce population deviation and perform extreme gerrymanders.

[President Punxsutawney Phil]

Anyone else working on State Senate maps?
(Reminder you can split precincts now)

The ability to splint precincts is so cool, but I noticed when I was doing some maps for fun that the President data isn’t accurate yet (atleast for South Carolina), and that it just assumed that every block voted the same as the precinct

Even if the data might be 99.97% or so accurate as opposed to 100% accurate, the ability to split precincts in places like Cobb is a game-changer.

[Born to Slay. Forced to Work.]

Anyone else working on State Senate maps?
(Reminder you can split precincts now)

The ability to splint precincts is so cool, but I noticed when I was doing some maps for fun that the President data isn’t accurate yet (atleast for South Carolina), and that it just assumed that every block voted the same as the precinct

Even if the data might be 99.97% or so accurate as opposed to 100% accurate, the ability to split precincts in places like Cobb is a game-changer.

Oh no yea it definitely is, the annoying city/precinct overlap is so annoying to work with, but my point of it was doing smaller districts (state house/senate) it can make a difference. I noticed it when I was splitting some precincts that had stark racial divides and the precinct may have been mostly republican/democrat, but it was showing the results the exact percentage that the precinct had which would be unrealistic if they are accurate. It doesn’t matter much for congressional and I am just so thrilled and happy and appreciative they added it, and the discrepancy won’t change much for congressional maps, just something to be conscious of as you make small districts.

And also it’s theoretically possible for them to get the block data. I know when I’ve doorknocked the campaign has data on how households voted going back a few cycles, but I know how easy that would be to implement as especially because I’m not a programmer. I don’t fault them for adding it at all.

[President Punxsutawney Phil]

Anyone else working on State Senate maps?
(Reminder you can split precincts now)

The ability to splint precincts is so cool, but I noticed when I was doing some maps for fun that the President data isn’t accurate yet (atleast for South Carolina), and that it just assumed that every block voted the same as the precinct

Even if the data might be 99.97% or so accurate as opposed to 100% accurate, the ability to split precincts in places like Cobb is a game-changer.

Oh no yea it definitely is, the annoying city/precinct overlap is so annoying to work with, but my point of it was doing smaller districts (state house/senate) it can make a difference. I noticed it when I was splitting some precincts that had stark racial divides and the precinct may have been mostly republican/democrat, but it was showing the results the exact percentage that the precinct had which would be unrealistic if they are accurate. It doesn’t matter much for congressional and I am just so thrilled and happy and appreciative they added it, and the discrepancy won’t change much for congressional maps, just something to be conscious of as you make small districts.

And also it’s theoretically possible for them to get the block data. I know when I’ve doorknocked the campaign has data on how households voted going back a few cycles, but I know how easy that would be to implement as especially because I’m not a programmer. I don’t fault them for adding it at all.

Oh, no disagreement there overall. However, even then, the room for error isn't necessarily that large for, say, state senate districts or even state house districts, depending on circumstances. The room for error comes when a precinct is split in 2 (or more) with substantial variation between those two halves in regards to how those different slices vote for. That probably won't be very important for even a state house district, since the error that is generally possible is not so large it renders the numbers worthless or even un-usable for general purposes (unless you are relying on a bunch of split precincts with varying politics, i.e. a string of precincts in northern Fulton designed to collect Latino precincts whenever possible).

The smaller size of the district, the more precinct splitting in the abstract can cause error. State Senate districts are around 190k, while House districts are around 60k. That's quite larger than individual precincts. It's quite unlike the North Dakota map I made with 125 seats (district size: just over 6k), where I would advise not putting too much stock into how one district is held to have voted.

[beesley]

https://davesredistricting.org/join/4e7384a5-0390-4bd8-b1c6-88bd78118c32

Another 'my definition of fair map' using the 2020 Presidential race figures. I tried to balance out the black % in Districts 4, 5, and 13, which is why there isn't a district made up of only DeKalb. Districts 2 and 7 are also maj-min. Loudermilk would be drawn out, but District 9 has no incumbent.

[Torie]

Here is my latest iteration of an aggressive but VRA compliant Georgia Pubmander using the 2019 population estimates, and the Trump 2020 numbers that are now in the DRA data base.



https://davesredistricting.org/join/b78dd1ca-18c4-4e7c-aa2e-97bd2272a06a

[lfromnj]

No real reason to post this, just thought it looked nice







https://davesredistricting.org/join/664342a1-f3ad-4993-9fe2-3828ddb87e3a

Seems almost criminal not to give Cobb and Dekalb their own districts, they're pretty much perfect for them.

Dekalb and Fulton share the urban area of the city of Atlanta and its inner ring suburbs. Pretty reasonable argument to split it although keeping them whole makes just as much sense. Cobb has the argument to be split between South Cobb and the highway cities compared to North Cobb. This would allow one to form a 4th black majority district as Sol did earlier in this thread.

I am not saying your map doesn't work fine in the Atlanta metro(GA02 is streching it pretty far though)

[Torie]

Somebody on RRH decided it was time to cut the Dem CD quota in Georgia down to 4 CD’s.

https://davesredistricting.org/maps#viewmap::c76a08aa-466a-44e8-a548-ccc2be568e93

I replied by linking the map below that has five 50%+ BCVAP districts all of which are compact enough to trigger Gingles in my view. So well, you guessed it, illegal! 

https://davesredistricting.org/join/2ded45e3-4c87-49b6-b0ce-9ecad601db89

[President Punxsutawney Phil]

behold, a map where every district in GA voted for Joe Biden!
https://davesredistricting.org/join/03b115b8-9974-4d69-b67f-c2d276fb8288

[Torie]

behold, a map where every district in GA voted for Joe Biden!
https://davesredistricting.org/join/03b115b8-9974-4d69-b67f-c2d276fb8288

How long did it take to thread that tiny needle?   

[President Punxsutawney Phil]

behold, a map where every district in GA voted for Joe Biden!
https://davesredistricting.org/join/03b115b8-9974-4d69-b67f-c2d276fb8288

How long did it take to thread that tiny needle?   

The entire map took about 30-45 minutes probably.

[Torie]

Very impressive that you can effect that so efficiently. I stand in awe of your talents.

[Skill and Chance]

Somebody on RRH decided it was time to cut the Dem CD quota in Georgia down to 4 CD’s.

https://davesredistricting.org/maps#viewmap::c76a08aa-466a-44e8-a548-ccc2be568e93

I replied by linking the map below that has five 50%+ BCVAP districts all of which are compact enough to trigger Gingles in my view. So well, you guessed it, illegal! 

https://davesredistricting.org/join/2ded45e3-4c87-49b6-b0ce-9ecad601db89

It is far from certain that a 5th VRA district would be mandated in SW GA, and I think there is a good chance it will continue to be uncertain even after census figures are released. Looking at your map, I see a number of potential issues.

The 2nd District on your map will be underpopulated by 15,000 in 2019, and likely will be underpopulated by 20,000 or more with 2020 census numbers. Even if you introduce a sixth county split, cherry-picking the most black heavy neighboring precincts to remedy this population imbalance would still push the BCVAP down to just about 50% BCVAP on the mark. If you don't introduce further splits, you are very likely going to end up underneath the 50% mark.

Another thing worth mentioning is that given longstanding turnout differences in that part of Georgia, Whites were either the majority or plurality of voters in your district for the 2014, 2016, and 2020 elections, even before you add more precincts to it (which you will have to).

I also wonder if the 11th Circuit or SCOTUS (with their recent Trump appointees) would view a district that needs to split 5 counties (three of which are quite small) by cherry-picking the whitest precincts to leave out as "compact". Maybe they would, maybe they wouldn't, but it seems very odd to assume that the GOP wouldn't at least try to find out given that the worst remedy that would happen is reversion back to 9-5.

Actually think SCOTUS is going to be very reluctant to further loosen the VRA redistricting provisions or chuck them outright.  Roberts and Kavanaugh watch the news closely enough to know the biggest threat to SCOTUS as an institution would be taking a hatchet to the VRA while Dems control the senate.

[Sol]

The VRA is also fairly beneficial to Republicans in many other cases, and of course it was the first Bush admin which pushed for stricter rules in preclearance.

[Alcibiades]

The VRA is also fairly beneficial to Republicans in many other cases, and of course it was the first Bush admin which pushed for stricter rules in preclearance.

Back when VRA districts first became a thing and Southern legislatures were for the most part still controlled by Conservadems, Republicans and black Democrats would ally against the white Democrats to push for VRA districts, as they would obviously help the Republicans by effectively amounting to Democratic packs.

[Torie]

Somebody on RRH decided it was time to cut the Dem CD quota in Georgia down to 4 CD’s.

https://davesredistricting.org/maps#viewmap::c76a08aa-466a-44e8-a548-ccc2be568e93

I replied by linking the map below that has five 50%+ BCVAP districts all of which are compact enough to trigger Gingles in my view. So well, you guessed it, illegal!  

https://davesredistricting.org/join/2ded45e3-4c87-49b6-b0ce-9ecad601db89

It is far from certain that a 5th VRA district would be mandated in SW GA, and I think there is a good chance it will continue to be uncertain even after census figures are released. Looking at your map, I see a number of potential issues.

The 2nd District on your map will be underpopulated by 15,000 in 2019, and likely will be underpopulated by 20,000 or more with 2020 census numbers. Even if you introduce a sixth county split, cherry-picking the most black heavy neighboring precincts to remedy this population imbalance would still push the BCVAP down to just about 50% BCVAP on the mark. If you don't introduce further splits, you are very likely going to end up underneath the 50% mark.

Another thing worth mentioning is that given longstanding turnout differences in that part of Georgia, Whites were either the majority or plurality of voters in your district for the 2014, 2016, and 2020 elections, even before you add more precincts to it (which you will have to).

I also wonder if the 11th Circuit or SCOTUS (with their recent Trump appointees) would view a district that needs to split 5 counties (three of which are quite small) by cherry-picking the whitest precincts to leave out as "compact". Maybe they would, maybe they wouldn't, but it seems very odd to assume that the GOP wouldn't at least try to find out given that the worst remedy that would happen is reversion back to 9-5.

My only comments are that I used the 2019 numbers that the DRA now has in its data base, and the county chops are small, and the VRA when it comes to the definition of compactness is not concerned with county or municipal lines, and the courts have had a pretty broad definition of what is "compact" beyond that for purposes of triggering Gingles. The downside for the Pubs is if they lose a case before SCOTUS on this issue after they lose the governorship in 2022 (if that happens). In such event the Courts will draw the lines, and that means a 7-7 map. So by reaching for a 9-4 map with VRA risk, they may end up with a 7-7 map rather than a 9-5 map. The risk-reward ratio of that for me simply does not pencil. YMMV.

Another problem is that the BCVAP numbers will be estimates, and it is uncertain what data base the courts will use. If the latest estimate before the map is passed, even if at the last minute, make clear that there is a very low risk any court will find that five 50%+ BCVAP seats can be drawn that it deems "compact," then the Georgia Pubs could dismantle the SW corner CD and snatch away a 5th black seat and take the political hit if they want to, without undue legal risk. I think that would be extremely short term thinking on their part;  unless they find a way to attract more black voters, and more  of the higher SES white voters in the Atlanta metro that have been abandoning them, their days as a competitive party in Georgia will be at an end, and sooner rather than later. As Virginia goes, so goes Geotgia.

That is how I seem the situation anyway, but one thing I am certain of, is that the Georgia Pubs could care less what I think. Ha!

Here is a "new and improved" revised map btw with only one excess county chop for what it is worth.

[Torie]

If you draw the lines, so that the perimeters of the three CD's that you mention contain the correct population for three CD's within them without change, then you may be right, even though it is wrong.

For example, here is a hideous (and evil) Pubmander that eviscerates the the SW corner black performing CD:



And here is one that adjusts the lines between just two CD's, that gives Biden a somewhat narrow margin.



And here is one where the SW corner black performing CD adjusts the lines between it and both of its adjacent CD's.

[Nyvin]

No real reason to post this, just thought it looked nice







https://davesredistricting.org/join/664342a1-f3ad-4993-9fe2-3828ddb87e3a

Seems almost criminal not to give Cobb and Dekalb their own districts, they're pretty much perfect for them.

Dekalb and Fulton share the urban area of the city of Atlanta and its inner ring suburbs. Pretty reasonable argument to split it although keeping them whole makes just as much sense. Cobb has the argument to be split between South Cobb and the highway cities compared to North Cobb. This would allow one to form a 4th black majority district as Sol did earlier in this thread.

I am not saying your map doesn't work fine in the Atlanta metro(GA02 is streching it pretty far though)

Keeping Atlanta whole vs Dekalb whole is a worthy debate,  I'm fine with either, the Dekalb portion of Atlanta is so small I think Dekalb district works better.

GA-2 is, over all else, just a Non-Atlanta Black Belt COI,  so the shape of it doesn't really matter.   The black rural population just needs to have representation one way or another.   That's one district where compactness shouldn't play any role.

[Born to Slay. Forced to Work.]

I was testing out the knew precinct splitting tool () by making a GA senate map and its soo good. I was kinda vaguely drawing to help democrats and it turned out to be a pretty even proportional map with the balance of power lying in a handful of swing districts in the southern part of the state and suburbs of Georgia. I really like how the districts look especially how they tend to follow natural barriers such as roads and rivers, as well as how it creates a competitive map.



https://davesredistricting.org/join/42a0d7a3-b956-4bd4-92e4-a10f1083d2dc

[S019]

9-5 GA with 4 black Atlanta seats and preserving the Southwest VRA seat

https://davesredistricting.org/join/4b904aee-9df7-4fbd-9173-0fa4748d3cbf 9-5 GA

the 10th could possibly be a dummymander down the line, but the seat is extremely Republican outside of Gwinnett, so it should last, in my opinion. McBath's seat was converted into a Republican seat that could flip back by the end of the decade, but the rurals will outvote the Democratic part of the seat for now.

[beesley]

I was testing out the knew precinct splitting tool () by making a GA senate map and its soo good. I was kinda vaguely drawing to help democrats and it turned out to be a pretty even proportional map with the balance of power lying in a handful of swing districts in the southern part of the state and suburbs of Georgia. I really like how the districts look especially how they tend to follow natural barriers such as roads and rivers, as well as how it creates a competitive map.

Great map!

[President Punxsutawney Phil]

9-5 GA with 4 black Atlanta seats and preserving the Southwest VRA seat

https://davesredistricting.org/join/4b904aee-9df7-4fbd-9173-0fa4748d3cbf 9-5 GA

the 10th could possibly be a dummymander down the line, but the seat is extremely Republican outside of Gwinnett, so it should last, in my opinion. McBath's seat was converted into a Republican seat that could flip back by the end of the decade, but the rurals will outvote the Democratic part of the seat for now.

Props for making a 9-5 less reliant on baconstripping Northern Georgia than is usually the case.

[Nyvin]

9-5 GA with 4 black Atlanta seats and preserving the Southwest VRA seat

https://davesredistricting.org/join/4b904aee-9df7-4fbd-9173-0fa4748d3cbf 9-5 GA

the 10th could possibly be a dummymander down the line, but the seat is extremely Republican outside of Gwinnett, so it should last, in my opinion. McBath's seat was converted into a Republican seat that could flip back by the end of the decade, but the rurals will outvote the Democratic part of the seat for now.

Props for making a 9-5 less reliant on baconstripping Northern Georgia than is usually the case.

That map is still pretty baconstripped (but a marginal improvement than the usual, yes.)

[President Punxsutawney Phil]

9-5 GA with 4 black Atlanta seats and preserving the Southwest VRA seat

https://davesredistricting.org/join/4b904aee-9df7-4fbd-9173-0fa4748d3cbf 9-5 GA

the 10th could possibly be a dummymander down the line, but the seat is extremely Republican outside of Gwinnett, so it should last, in my opinion. McBath's seat was converted into a Republican seat that could flip back by the end of the decade, but the rurals will outvote the Democratic part of the seat for now.

Props for making a 9-5 less reliant on baconstripping Northern Georgia than is usually the case.

That map is still pretty baconstripped (but a marginal improvement than the usual, yes.)

What is the smallest amount of baconstripping possible in a genuine 9R-5D map?

[Torie]

https://davesredistricting.org/join/b9b2fcb9-67e0-4b92-962f-3342adde1935