2020 Census and Redistricting Thread: Alabama

[Torie]

Assuming Alabama keeps its seven seats, and assuming further that the blue district is 50% BCVAP (it isn't quite, but the numbers are out of date and this is meant as a hypothetical), does the law require that one has to draw the blue CD under the VRA, or alternatively, and indeed maybe as a further requirement, given the hypothetical 50% BCVAP CD "VRA "trigger" is in play, a lower than 50% BCVAP CD that basically is Jefferson County, assuming it is black performing?

The issue is whether the  50% BCVAP VRA trigger happens with a hypothetically erose CD, that chops counties, with some white rural territory in between. It is not clear in my mind just where the "line" is drawn on such hypothetical CD triggers. It is a salient question, because Wasserman says the Dems plan to sue if they do not get two black performing CD's in a seven CD Alabama map.

[Torie]

Assuming Alabama keeps its 7th seat, a lawsuit is born. I was able using the 2018 numbers to draw two CD's, that are both over 50% black in population, as well as over 50% BCVAP. I now accept my accolades. Suitable for framing, no? Please do not get overly effusive. That would make me blush. Thank you.

In other news, as much as it may shock you, the remaining CD's are not politically competitive.  

[Torie]

Even with 6 seats, it is possible to draw 2 Dem districts - even without splitting Mobile.
https://davesredistricting.org/join/3935470e-4f7f-44ee-aec7-0d913809fe45

And here's a 7 seat map on a similiar concept.
https://davesredistricting.org/join/0e7a07ee-8cf0-4d10-8926-50be3efcbfeb

Both of  your CD's are below 50% BCVAP, which is where the heavy lifting is, to get over that.
But I notice that while you are using 2018 population numbers, you are using 2010 VAP numbers. You might switch to 2018 VAP numbers on the DRA to see if that gets both of your CD's over the 50% BCVAP hurdle. If not, no cigar unless the census numbers, or to put it more accurately, whatever algorithm will be used by the courts to estimate such numbers, gets you over the top, or the VRA law further evolves.

[Torie]

Even with 6 seats, it is possible to draw 2 Dem districts - even without splitting Mobile.
https://davesredistricting.org/join/3935470e-4f7f-44ee-aec7-0d913809fe45

And here's a 7 seat map on a similiar concept.
https://davesredistricting.org/join/0e7a07ee-8cf0-4d10-8926-50be3efcbfeb

Both of  your CD's are below 50% BCVAP, which is where the heavy lifting is, to get over that.
But I notice that while you are using 2018 population numbers, you are using 2010 VAP numbers. You might switch to 2018 VAP numbers on the DRA to see if that gets both of your CD's over the 50% BCVAP hurdle. If not, no cigar unless the census numbers, or to put it more accurately, whatever algorithm will be used by the courts to estimate such numbers, gets you over the top, or the VRA law further evolves.

I don't care about BCVAP, personally. Districts like NC-01 never had an absolute need to be majority black CVAP anyway (I'm aware that what can be done to individual districts can vary quite a bit depending on specific case law). I agree that VAP can be of validity, but CVAP is no cigar for me.

It's a legal issue, under the VRA, as to what triggers the legal requirement for a performing minority district, not a policy issue. I discussed this at length with Muon2 recently. It may be true that the law is not crystal clear on this matter, but it is an issue. Another issue, is when you hit the trigger, are you then required to avoid a gerrymander and erose CD's that have more of a minority population than you need to make the CD minority performing? I think the answer to that is probably yes. So the irony is that having drawn two CD's that effect the trigger, you then cannot actually draw them, because they are erose and gerrymandered black packing CD's. So you then, e.g., under the VRA, end up having to drawing a black performing CD nested in Jefferson County, which in this case is what hewing to appropriate redistricting principles as to what is good policy would dictate anyway. The dance is a most complicated two step.

Have I made this all perfectly clear? Yes, I thought not. Good luck!  

[Torie]

You can actually get two majority black districts with 6 districts, though it requires splitting Huntsville.

That purple thing of yours is almost certainly not compact enough to effect the trigger.

A couple of more generalized comments about what will probably the most interesting VRA issue/case of the cycle, and thus my interest. The black population where it counts may be higher than what is now 5 year old data, per Muon2 as to the staleness of the data. Second some of the "precincts" in Alabama have like 10K people in the DRA shape files. When we get real precincts, lines can be more tailored to keep "unnecessary" whites out. So there is probably some wiggle room when we get down to having real data and real precinct lines, to get up to the 50% BCVAP trigger. It is unlikely however that one will be able to get there with a 6 CD map. These are my surmises anyway.

Finally, if there is a way to make a Jefferson County dominated CD that is not a performing black CD, that does not make the state map look like a gerrymander per se overall, to mitigate their VRA exposure,  they would be well advised to draw it (not sure that is possible, but whatever). An example would be to de-nest a Jefferson County CD at the margins, say in order to avoid a county chop elsewhere, or a municipal chop of some importance. The idea is to locate chops where it most benefits them, without degrading too much its overall score based on neutral redistricting principles. I played that game with my Pubmander lite map of Missouri. So many games, so little time.

Below is the kind of direction that I am talking about. The blue CD is close to a swing CD (which the Pubs might be wise  to settle for nevertheless under the theory that pigs get fat and hogs get slaughtered), designed so that the map does not appear to be facially a gerrymander designed to screw the blacks. It would need to be very carefully done, so the green CD is not excessively black, i.e. a black pack. How to most effectively do it will depend on the final numbers, with much smaller units to work with. One "precinct" I found had 61K people. That is not very helpful when it comes to game playing.

[Torie]

You can actually get two majority black districts with 6 districts, though it requires splitting Huntsville.

That purple thing of yours is almost certainly not compact enough to effect the trigger.

I mean, if a Mobile to Montgomery district is sufficiently compact, then who's to say a Birmingham to Huntsville district isn't? It's a stretch to be sure, but not a massive one.

Well, OK. Snake things like that appear considerably different to me than a prong jutting south a bit from where a CD has a bunch of contiguous counties, covering territory that all the way is substantially black, rather than a much longer snake covering territory where blacks are very thin on the ground. But hey, what floats my boat may sink yours. Again, we are talking just about trigger mechanisms here under the VRA, not what must be drawn, or even perhaps what may ultimately be legally drawn. The VRA is packed with quick sand traps every step of the way in fact patterns such as presented in Alabama. It is kind of a legal nightmare really - at least when it comes to Alabama. The existing Alabama map btw looks as illegal as hell to me even at the time it was drawn. That is puzzling to me in and of itself. Gross gerrymandering to black pack is a good way to get yourself in legal trouble in a hurry. Just ask the Virginia Pubs about that one.

[Torie]

And if two black performing CD's are required because two are triggered per the 50% BCVAP test, that is the kind of map that in general might well be required. But in reality, it is less likely the triggers will be found, if you have a map that does not look much like a gerrymander per se.

[Torie]

And if two black performing CD's are required because two are triggered per the 50% BCVAP test, that is the kind of map that in general might well be required. But in reality, it is less likely the triggers will be found, if you have a map that does not look much like a gerrymander per se.

What's the trigger? That two 50% BCVAP could be drawn, so some attempt must be made to draw them, even if the districts don't actually reach that mark?

When two triggers are in play, you have to draw two performing black CD's, but not have excess blacks in the CD that are not needed for a performing CD, pursuant to a gerrymander. But if not a gerrymander, excess blacks are OK.

That's my take of what it all comes down to, anyway.

[Torie]

And if two black performing CD's are required because two are triggered per the 50% BCVAP test, that is the kind of map that in general might well be required. But in reality, it is less likely the triggers will be found, if you have a map that does not look much like a gerrymander per se.

What's the trigger? That two 50% BCVAP could be drawn, so some attempt must be made to draw them, even if the districts don't actually reach that mark?

When two triggers are in play, you have to draw two performing black CD's, but not have excess blacks in the CD that are not needed for a performing CD, pursuant to a gerrymander. But if not a gerrymander, excess blacks are OK.

That's my take of what it all comes down to, anyway.

I'm still confused what a trigger is exactly though. Can you explain what you mean when you say that?

As I said above, you need to be able to theoretically draw two "compact" CD's that are 50% BCVAP. Where the rubber meets the road, is what "compact" means, for purposes of the trigger. What I don't think it means, is hewing to county and municipal lines. But theoretically being able to draw something erose, which has a bunch of white territory that it has to cross, is more problematical, as sufficing to meet the "compact" prong of the trigger. But aside from governmental lines not meaning much for purposes of this test, nobody really knows exactly where the line is drawn legally as to the "compactness." That is one reason, that drawing a map which looks reasonable, will make it less likely that the "compactness"  lodestar will be given an expansive definition, in my opinion. Conversely, if you draw something that looks gross, it is more likely for a court to think, hey if you live by the sword, you die by the sword, and give a broader definition of what is "compact," for purpose of the trigger. If you go gross, we can too, for purposes of finding the trigger is there, even though you may well not actually be able to draw a CD that actually looks that way, because it has excess blacks in it, not needed for it to be performing.

So for me, I first try to draw the theoretical trigger CD's, before abandoning them. The legal mind works in mysterious ways.

[Torie]

I just drew a nice AL map that has two 50%+ BCVAP CD's that does not even look particularly ugly. The best scenario for the news this afternoon is for MN to retain an 8th seat I think. AL retaining a 7th CD will most likely mean a second minority performing CD, by the time SCOTUS does its thing at least, assuming it does not reverse Gingles.

https://davesredistricting.org/join/457311a7-6b54-4fd1-a906-4102125c7082

[Torie]

I just drew a nice AL map that has two 50%+ BCVAP CD's that does not even look particularly ugly. The best scenario for the news this afternoon is for MN to retain an 8th seat I think. AL retaining a 7th CD will most likely mean a second minority performing CD, by the time SCOTUS does its thing at least, assuming it does not reverse Gingles.

https://davesredistricting.org/join/457311a7-6b54-4fd1-a906-4102125c7082

I still don’t understand why the Obama DOJ didn’t push for a 2nd VRA district here.

I suspect the numbers were not there to draw two "compact" 50%+ BCVAP districts that had no overlapping territory at the time (e.g., before the 2019 CVAP data was put into the DRA, it was a very close case whether Gingles triggered a second black performing CD). It may also be that the contours of the Gingles parameters had not been as clarified by SCOTUS as they are now. If the data base the Court uses matches the current data base in the DRA, either the Pubs are going to lose this case, or SCOTUS is going to modify Gingles. The districts are clearly sufficiently compact. Heck, the map does not even entail an extra county chop.

[Torie]

If I were the Alabama Pubs, in order to increase the odds that they will not lose what would appear to be a most meritorious Holder lawsuit, I would draw the map below, and I would do it before rather than after losing the Holder lawsuit because if they do it after, the odds would be higher than it would be held an insufficient after the fact remedy given that they demonstrated bad faith in the first instance.

You see, they did not chop Jefferson County, and then added the county which minimizes the chop into the third county, and the chop into the third county is the logical place to do it. In other words it was a map which hewed to neutral non race based redistricting principles. At least the Pubs get a somewhat competitive district which if it elects a Dem, the Dem will be a moderate Dem. What are the odds that the Pubs will take my advice? Probably about as high  as I being elected POTUS. Sad!

[Torie]

Just for the Mobile-split-naysayers:


It's cleaner than the current map....

Two performing AA districts is 100% possible and should be required.   No excuses.

It is very probably required not because two performing black CD's can be drawn, but rather because two most probably deemed "compact" CD's can be drawn that are 50%+ BCVAP, thus triggering the need to draw two performing black CD's under the Gingles metric.

[Torie]

Question: say AL has two black seats drawn in 2020. In 2030, if the state drops to six seats, does it lose one of those black seats?

For purposes of the applying the Gingles test, it would if the census figures are identical to the 2020 census figures.

https://davesredistricting.org/join/457311a7-6b54-4fd1-a906-4102125c7082

[Torie]

If SCOTUS upholds the decision that most likely intimidates Louisiana Republicans into allowing a second black district too, which would be amazing.

Unlike AL, you can't draw two "compact" 50% BVAP CD's in LA, so no. Gingles does not trigger the drawing of a second black performing CD in LA.

[Torie]

One of the problems is that Alabama is oddly shaped. I've long advocated for giving everything west of Gadsden County in Florida's Panhandle to Alabama.

If SCOTUS upholds the decision that most likely intimidates Louisiana Republicans into allowing a second black district too, which would be amazing.

Unlike AL, you can't draw two "compact" 50% BVAP CD's in LA, so no. Gingles does not trigger the drawing of a second black performing CD in LA.

How does the "compact" criteria work in states that aren't compact or otherwise oddly shaped?

I am not sure I understand you question since it appears to be focused on the shape of states rather than districts, but the definition of "compact" is not that precise under the law.

[Torie]

If SCOTUS upholds the decision that most likely intimidates Louisiana Republicans into allowing a second black district too, which would be amazing.

Unlike AL, you can't draw two "compact" 50% BVAP CD's in LA, so no. Gingles does not trigger the drawing of a second black performing CD in LA.

Wrong. You can.

So it would appear as to the 50% element of the test, and we do have racially polarized voting, so that just leaves the compact element to litigate. 

https://davesredistricting.org/join/8bfa1863-ed2e-466e-9d9f-90b35e656962

[Torie]

Can someone post an actual map that has two 50% BVAP CD's with what they think minimizes erosity?

I revised mine a bit to get rid of the neighborhood in the NW corner of NO that is almost all white with about 20K people or so.

The "contest" here is to draw a map that is easiest* on the judicial eyes that has two 50% VAP black CD's. The idea is to maximize the odds thata court that matters would find that both CD's were "compact."

I think the odds are low that it is possible to  draw such a map that gets the odds into more than between one and two standard deviations from the mean on the losing side, with me the handicapper, but absent actual maps, it is all ultimately meaningless foreplay for me to handicap at all. At least in my opinion!  

https://davesredistricting.org/join/8bfa1863-ed2e-466e-9d9f-90b35e656962



*Job one of course to get even half way to first base with the wooing of your affection being to lose Shreveport as being part of the same zone of compactness as Baton Rouge.

[Torie]

Can someone post an actual map that has two 50% BVAP CD's with what they think minimizes erosity?

Oryxslayer's map is just upthread! Yellow is genuinely quite compact, and Green is crummy but not any worse than the current 2nd, and as he says it's probably legal to do a district based in New Orleans which is only plurality on VAP.

This one: https://davesredistricting.org/join/354c6bee-a5b8-4157-ace1-218130b9a4ce ?

It is prettier than mine, and given the zero deviations seems like a pro job. I note that it does have to combine the inner cities of Shreveport and Baton Rogue. What are the odds that a high court will find that yellow CD "compact" for purposes of Section 2 of the VRA?  Close to de minimus. JMO.

[Torie]

If SCOTUS upholds the decision that most likely intimidates Louisiana Republicans into allowing a second black district too, which would be amazing.

Unlike AL, you can't draw two "compact" 50% BVAP CD's in LA, so no. Gingles does not trigger the drawing of a second black performing CD in LA.

Can you help me understand how the situation in Louisiana is different from what happened in Virginia, where a packed African-American majority district was found to be unconstitutional and replaced with two districts, neither of which is 50% BVAP?

Most of Louisiana has racially polarized voting, but you could draw a <50% BVAP based on New Orleans which would perform for the candidate of the Black community's choice.

The issue is whether Gingles requires LA to draw two performing black CD's? You are asking that if it does, are you allowed to draw the CD's that constituted the trigger? That question remains unsettled, as well as just what more precisely "compact" means. But the current SCOTUS, the majority thereof, would rather DIAF than find there is enough compactness to trigger Gingles in this context, IMO.  

[Torie]

I suppose I’m asking why, all else being equal (not factoring in SCOTUS taking things in a different direction), the precedent in Virginia which led to creating a second district doesn’t apply to what looks like a similar situation in Louisiana. Virginia didn’t require two 50% BVAP districts as a prerequisite to unpack the district and draw two performing districts.

Or put another way, are you arguing that hypothetically the current Supreme Court would have overturned the lower court decision in Virginia from some years ago because their views are different from the old court’s.

The VA Pubs believed that under the VRA, you needed to have a 50% BVAP CD. The same was true of the OH Pubs when they had the Cleveland CD go down to take in black Akron. That misconception of the law caused the court in VA to nix the CD the Pubs drew as race driven, and a black pack CD. Now the Pubs know better than to try to get up to 50% through gerrymandering. Instead you go after Dems, and pack them, and if Gingles has been triggered, you also make sure the CD is black performing. If Dems and blacks are co-extensive, you have a tricky situation about intent, and effect and so forth, but that is a rare situation on the ground. In all  events, you would not want to exclude all those white Dems in the Garden District and the French Quarter from a black CD in LA.

Another way to put it, is that the legal dance is very different now. I also think the current SCOTUS will  be parsimonious when it comes to deeming these wild looking CD's that take in very disparate black nodes far away from each other as being found "compact" for purposes of triggering Gingles.

Another complication is that many states have their own set of laws, and in places like NC, activist Dem courts have used generalized language in the state constitution that was copied and pasted from the federal one, to reach a very different legal result from SCOTUS. Since I as knee high to a grasshopper, I have been very hostile to the copy and paste practice. That just gives activist judges more power to effect their favored policy goals, unnecessarily.

[Torie]

Sigh. If a state wants to go the proportionality route like Ohio apparently for the state legislature, then like Ohio the state should pass a statute. I was sharing my opinion about abuse of process as it were by activist judges. Given how hideous the Illinois CD map is, if Illinois has an initiative process, the citizens should rise up and try to copycat the Ohio law, and spend 50 million promoting it. That should keep the Dems busy for awhile. But also I don't think Illinois has the citizen initiative process.

[Torie]

It's funny reading through this thread and seeing everyone being deadset on either a 5-2 happening or not happening.

The reality is it's prolly a tossup right now, and depends upon what SCOTUS wants to do with it, though it def isn't a shoe-in case either way.

There's been a lot of focus on other Southern states with GOP maps, but I wonder what this precedent would do to existing Dem maps?  It would seem to force a Hispanic CVAP performing district in Las Vegas assuming one can be drawn?  

First, there is a GOP suit to this nature right now in NV. Second, while a Hispanic CVAP congressional seat is impossible - a lot of the population is intermingled with north Las Vegas AAs or South suburban Whites - you can almost certainly draw a coalition seat and make it so the Whites in said coalition seat are rural GOP republicans excluded from the D primary process. The Dems in fact will argue that this is what they did with CD4, even though it is just above 50% White by CVAP.

IMO, the Pub lawsuit is frivolous. Coalition CD's are not a legal thing, and Gingles is not triggered, both because of the 50% threshold, and also most probably, because there is not racially polarized voting when it comes to Hispanics.

[Torie]

Your post leaves me confused, but then maybe my post was confusing. There are substance issues and process issues, and legal issues from the past, and in the present. My discussion of the past was when the Pubs misunderstood the law, as it turned out - VRA law as interprested by SCOTUS. And there are state laws and federal laws.

My activist judge comment was about a state court taking a state constitution provision about equal protection of the laws or something like that, and then holding it outlaws gerrymandering or requires proportionality. And yes, gerrymandering is bad (although what is gerrymandering is often in the eyes of the beholder, and proportionality pursuant to an appropriate formula not linear should be in the mix along with respecting jurisdictional lines, compactness, COI's and all the rest, I agree. But not by a state court using broad feel good language never intended for that purpose. Do it by statute, or express constitutional amendment.

[Torie]

Maybe the thread below should be merged into this one. My wild guess is that there is about a 20% chance that the court will hold Gingles triggers 2 black seats, a 40% chance that it triggers 1 black seat, and a 20% chance Gingles triggers 0 black seats, either because Gingles is dead, or a revised test for the trigger finds nothing in AL meets it, i.e. under a revised standard, you need to crack a cohesive black zone in a urban area to deny blacks representation, and rural black belts just don't measure up anymore.

https://talkelections.org/FORUM/index.php?topic=481005.msg8459764#msg8459764