2020 Census and Redistricting Thread: New Jersey

[Torie]

Do partisan Democrats even support universal fair redistricting? Given the anger about AZ and CO and the enthusiasm for the Hochulmander, I'm skeptical. It seems like Torie is the only Red av really decrying Democratic gerrymandering, and most of the rest just wanted fair maps when they thought the GOP would gain from redistricting and are just fine with gerrymandering when it goes the other way.

A big part of that is because the people who follow this on the Dem side know that it's only for that state, so thinking Democrats get burned in a state they control while Republicans are taking seat after seat in states they control creates a feeling of losing the battle. With the fact that national reform is further away than ever, there is a belief that Democrats have to gerrymander in the states they control to offset Republican gerrymandering. Really hard not to be sympathetic to that, particularly for a Democrat who came of age before 2010 and saw the effect partisan gerrymandering had on elections. There isn't anything necessarily wrong with believing it has to be national reform or bust, given the results so far at the state level.

If a national gerrymandering ban / commission bill had passed and every state was doing the same thing more or less, this wouldn't feel like such a zero-sum game with high stakes.

I get that for Illinois, where there are no laws being abused or defied. That is not the case in NJ and NY, where the legal process was gamed and abused, and the law defied (the NY CD map being utterly outrageous). Where the Pub were in control, and constrained by law, putting aside DeSantis, they respected the law in Florida (with a friendly court to boot), and did pretty well in Ohio. And then there is NC where a Dem court thinks the state copying and pasting of the federal equal protection clause outlaws Pubmanders, which is an abuse of judicial power imo.

[Torie]

NJ wasn’t gamed, really, it just has a bad process that leaves one party the winner and the Dems were a little more careless about it this time than Republicans were last time. As long as there is a tiebreaker, there will be a crappy outcome in NJ. Either one side wins or you get uncompetitive incumbent protection.

Wallace did not follow the law, so I disagree. He also it appears gamed the system by leaking, directly or indirectly,  the Pub map to the Dems.

The takeaway for me is that redistricting laws need very tight parameters, and I think VA points the way as how best to break the tie - each party subject to judicial confirmation makes nominations, and the two experts draw the map under instructions to try to compromise out their differences, and write a report explaining their choices.

Oh here is a link to the text of the decision. I find it unpersuasive, but maybe the law is so bad given the lack of the need to have a record and so forth that there is no effective remedy to the tie breaker going rogue.

https://newjerseyglobe.com/redistricing/n-j-supreme-court-dismisses-gop-lawsuit-on-congressional-redistricting/
 

[Torie]

My question is, what is a fair map?

That is the question, isn’t it? I think it has to be proportional to statewide results (within reason, this gets hard in unbalanced states), adhere to the VRA, be responsive to swings in the electorate from year to year. and pay some deference to political and demographic lines. Ultimately I would judge a map by whether it enables a responsive and representative democratic government.

You do realize that nationwide that gives the Dems an advantage right? That is because CA and NY are so heavily skewed to the Dems with IL headed that way, and on a smaller scale MA. In that regard,  Muon2 will tell you that proportionality is not linear, but exponential, so you take the percentage for the majority party over 50% and double it, to get the percentage split.

The "gold standard" state, MI, had to gerrymander n favor of the Dems, particularly for the legislative seats, to get to proportionality. The Grand Rapids seat is also a clear gerrymander in favor of the Dems. Myself, I prefer following neutral metrics, with proportionality only in play where there are two reasonable choices.  It also seems that when one party has the trifecta plus the court, particularly the Dems, the whole system breaks down, even with a commission, so there do need to be very tight rules, and a tie breaker mechanism like VA.

I am not optimistic about the future of redistricting. And with partisans in control in most places, the number of competitive seats is diminished, so both parties nominate and elect people that are more politically extreme than the majority of voters would prefer. The more extreme half of each party rules the roost, so basically the nation can be ruled by the most extreme 25% of the electorate.  The system is broken.

I appreciate your posts by the way, even if I typically disagree with them. 

[Torie]

You do realize that nationwide that gives the Dems an advantage right?

I don’t see why that should necessarily be true because of the situation in some of the larger states. There are a number of mid-sized and smaller states where this advantages Republicans, and there are more of them. Beyond that we get the issue that there is a slight Democratic majority in the country so a system favoring proportionality would favor Dems. And also that, given the situation you describe, it may favor Dems but the alternative (legislative rules) would favor Dems even more strongly because of California.

I am just saying what you propose nationwide gives the Dems more than their proportionate share. Other schemes give the Dems more of an excess than yours does.

I have posted this before, but when I write the new US Constitution, we are going to a parliamentary system, probably like the German one, where there are individual seats, but then seats are awarded to achieve proportionality based on the national vote.

In the meantime, I favor the Muon2 rules, where proportionality is a tie breaker between maps that otherwise score pretty equally. The amount of discretion would be minimized, because drawing the lines brings out the worst of the hack in our sadly flawed species.

[Torie]

I think Brittain33 is being honest here but national Democrats goals are Pr or efficiency gap in MI and PA and Wi but in states like CA minority seats come foremost. We would never see the south Brooklyn seat because it it probably takes away the north Manhattan seat even though a south Brooklyn seat is both great on COI grounds and great for partisan responsiveness

Based on a "challenge" by Sol, who opined that it seemed difficult to unite Woke White Park Slope with similar hoods in Manhattan, without messing up minority CD's and so forth, I drew a COI map that I think did the job, which just using neutral redistricting metrics and hewing to COI's, turns out to be pretty much of a Pub wet dream, but hey it gives them a proportionate share using the proper metric (1- ((.6076-.5) x 2) +.5) x 26 = 7.4048 seats).

https://davesredistricting.org/maps#viewmap::fe817ecb-ce7f-405e-8a00-6e2fcc802294

How did I do? 

[Torie]

I think Brittain33 is being honest here but national Democrats goals are Pr or efficiency gap in MI and PA and Wi but in states like CA minority seats come foremost. We would never see the south Brooklyn seat because it it probably takes away the north Manhattan seat even though a south Brooklyn seat is both great on COI grounds and great for partisan responsiveness

Based on a "challenge" by Sol, who opined that it seemed difficult to unite Woke White Park Slope with similar hoods in Manhattan, without messing up minority CD's and so forth, I drew a COI map that I think did the job, which just using neutral redistricting metrics and hewing to COI's, turns out to be pretty much of a Pub wet dream, but hey it gives them a proportionate share using the proper metric (1- ((.6076-.5) x 2) +.5) x 26 = 7.4048 seats).

https://davesredistricting.org/maps#viewmap::fe817ecb-ce7f-405e-8a00-6e2fcc802294

How did I do? 

Unrelated and tangential, but I'm surprised nobody has yet said your NY-11 needs to be nuked from orbit

COI my dear. That was the only way to pick up some more non-woke white people (Hispanics are an acceptable substitute if not needed for an Hispanic CD, along with Arabs* (lumping them in as Asians with the Chinese is a thought crime). To the south was water, to the west was NJ, to the north was a wall of persons of color and woke whites, so it was go east young man. And it was beautiful, and it was good, and there was great rejoicing.

*Have to keep the Arabs out of the orthodox Jewish CD of course.

[Torie]

You do realize that nationwide that gives the Dems an advantage right?

I don’t see why that should necessarily be true because of the situation in some of the larger states. There are a number of mid-sized and smaller states where this advantages Republicans, and there are more of them. Beyond that we get the issue that there is a slight Democratic majority in the country so a system favoring proportionality would favor Dems. And also that, given the situation you describe, it may favor Dems but the alternative (legislative rules) would favor Dems even more strongly because of California.

I am just saying what you propose nationwide gives the Dems more than their proportionate share. Other schemes give the Dems more of an excess than yours does.

I have posted this before, but when I write the new US Constitution, we are going to a parliamentary system, probably like the German one, where there are individual seats, but then seats are awarded to achieve proportionality based on the national vote.

In the meantime, I favor the Muon2 rules, where proportionality is a tie breaker between maps that otherwise score pretty equally. The amount of discretion would be minimized, because drawing the lines brings out the worst of the hack in our sadly flawed species.

Here's actually a mini example of what happens with Britainn33 standards. Let's look at 4 states. Let's also just equalize their turnout or whatever.

Oklahoma, Massachussets,Arkansas, and Wisconsin. Combine Arkansas and Oklahoma into 1 state for this purpose. With equalized turnout it's probably pretty close to 50/50


The current maps are 4 Safe R in AR, 5 Safe R in OK, 9 Safe D in MA. 4 Safe R in WI, 1 Lean R and 1 Likely R and 2 Safe D.

My ideal fair map would be 1 tossup in Oklahoma, 4 Safe R, 1 Likely R in AR and 3 Safe R,in WI it would be 3 Safe R, 1 Likely R , 2 Lean R and 2 Safe D.

What would Britainn33 standards result in?

Maybe 1 or 2 Likely D's in MA,  but still 9 very D districts. Oklahoma probably combines Norman and most of Oklahoma county to make a Likely D district.  Arkansas makes a Safe D district with Little Rock and the Delta. Wisconsin would probably be 3 Safe D , 1 Lean D, 1 Lean R and 3 Safe R.
So if we count Lean districts as .75, .25 for their respective parties. I will just count Likely's as a whole 1.

The current scenario has
14.75 R districts  11.25 D districts. We also do have to wait for Wisconsin's actual districts which could end up in 14.5/11.5


My fair map scenario is 14 R, 12 D.

Britain33's scenario is 15 D-11 R which is actually more extreme than the current delegation. Obviously still nothing crazy but still the most biased.

So infact as Torie said your standards/national Democrats standards would result in quite unfavorable maps for the GOP as national Democrats obviously designed it. I don't believe you have the same intent as the national Democrats but if your effects are the same. Just draw geographically good maps and you will get pretty responsive and equal maps nationwide. Small biases can exist but one can just suck it up. Incumbency bias's are probably more likely to result in an unfair result. Trying to muck around with this results in geographically worse districts and may not actually result in more fair maps.


I chose these 4 states because its an interesting exercise in geography. 1 Super Safe D state, 1 Super Safe R(combined) and 1 pure tossup. The very Safe D state has very good D geography while the tossup has very good R geography. Meanwhile the Safe R state has fairly neutral geography. I could also replace the 2 R states with TN but TN isn't nearly as R as these 2 states.

It's either that or just make an actual PR system like Germany does. Now what Democrats want is obviously more seats and maybe PR may result in that in a 2 party system  but they also obviously don't want to completely mess up the system either as PR very likely could end up in a multiparty system. So why not try to create something that not only gives them more seats than PR but doesn't radically change anything.

It seems also a priori obvious that you cannot get anywhere near strict proportionality in larger states heavily skewed to one party. All of those states are Dem. Meanwhile, in more closely balanced states, where the Pubs have a geographic advantage, that is systematically wiped out by abandoning neutral metrics to gerrymander to offset the geographic advantage.


So the Dems get far more than their proportionate stare in CA, NY, IL (maybe, can one gerrymander to get the Pubs up to 7 seats), NJ, and MA, with the smaller states a wash, overall. It's a formula for the Dems having a near permanent majority in the House given current party coalitions or a big wave election.

[Torie]

Torie, I think you're hanging a lot on California and Massachusetts geography that doesn't work with all those other states.

A proportional map in New York would have 10 R seats in a neutral year. If you do a mild R gerrymander you can get to 8-9 easily. 2 on LI, 2 in NYC, 1 in the Hudson Valley, 1 up North, the rest in central and western NY.

A proportional map of New Jersey would have 5 R seats. You can draw 4 R districts without breaking a sweat, 5 with a mild gerrymander.

A proportional map of Illinois would have 7 Republican seats. I haven't tried but I think 5-6 would be easy, 7 within reach.

And if those 1s and 2s in New York, New Jersey, and Illinois offend, then fine - let's assign a bonus to those states like OK, UT, and AR which are so far from the 50-yard line that they don't have to create any Lean D or neutral districts.

In addition, I think people are so accustomed to the Republican bias built into our current system and the preferences many have for compact and squared off districts (which dovetail well with historical segregation patterns / urban packing of Dems) that a neutral map feels like the worst-case scenario, and something favoring Democrats feels actively perverse and corrupt. You have not had to live with what we have had to live with for 20+ years of the game stacked against us. Consider that if a system has a small Dem edge, it's a reflection of the impossibility of achieving true neutrality over many years rather than a dealbreaker for system.  

You do know that before before the alleged 20 years, the gerrymandering was pro Dem for the 40 years before that right? The whole idea of chopping and erosity to achieve proportionality is a profoundly bad idea and will continue the downward slide of a functioning democracy away from where the two sides have some level of comity and do not hate each other, to where each side has  an uncomfortably high number of extremists/kooks and willing to do almost anything they can get away with to screw the other side. And there should be more swing districts, where there is a reward for those and room for those who are not the show horse firebrands. We need more room for the Katkos, and not less. And yes, Manchin and Sinema and Collins and Murkowski too, Senators of an ilk that will soon no longer be with us.

At the moment, what is happening is that the Dems in their gerrymanders and courts and commissions into gerrymandering to get to proportionality, and erasing in large numbers CD's friendly to moderate Pubs, so what will be left is a more pure caucus of hardline Pub kooks. You shall see. Be patient.

In any event for NYS, it is quite doable to get to 7 seats, with 8 you have to work, and to get to 10 Pub seats (10.5 or so actually is the proportionate number), with swing seats counting a half seat, near impossible. As a realistic matter counting swing seats as a half seat without something insane, 8 is about the max. NYS is becoming an almost bizarre state with so many orthodox Jews voting near unanimously Pub now and growing so rapidly. Dan and I visited Kyras Joel yesterday btw. It is right out of a sci fi novel. It is like entering through the wardrobe, on one side is rural England, and on the other side is Narnia just like that - zap.

The NY map below is actually a clean pro Pub map, that does hew to neutral redistricting principles taking cognizance of COI's. That is the kind of map that is appropriate to draw to even the scales some, for the minority party.

https://davesredistricting.org/join/2772529a-d162-4387-b07d-f5d9589005d2

[Torie]

If the algorithm used is disclosed, I think computers producing maps is a good starting point. Then the humans can evaluate them. The most frustrating aspect of where we are now, is the deafening roar of partisan spin and courts gone rogue. And nobody seems to agree on the most appropriate way to measure partisan bias. What I do is ignore the partisan numbers, draw a clean map adjusted for the VRA and each group getting a reasonable share of the spoils, and then look at the partisan split. I am highly suspicious of all the complex mathematical tests of partisan bias, and just go with the Muon2 formula, which uses  an exponential rather than linear function of what percentage share of the districts a party should get based on the statewide partisan split. If the split is ballpark, I'm done. If it is not I assess how to get it more ballpark, without making the map a mess. It is a balancing test. Competitive districts are also a plus.

[Torie]

Well, drawing CD lines in my current home state of NJ is always a challenge, but I decided to do the exercise based on neutral metrics, and that hack tie breaker retired judge really did snatch a seat from the Pubs, and in a grotesque manner to boot. I guess the Dems knew what was in the guy's brain in advance or something.

[Torie]

Well, drawing CD lines in my current home state of NJ is always a challenge, but I decided to do the exercise based on neutral metrics, and that hack tie breaker retired judge really did snatch a seat from the Pubs, and in a grotesque manner to boot. I guess the Dems knew what was in the guy's brain in advance or something.

It’s possible to give Republicans a second North Jersey seat too.

Not based on neutral metrics. I generally follow the Muon2 rules absent rather compelling circumstances (e.g.,, appending the Hispanic part of Newark to the Hispanic Palisades of Hudson County to get the Hispanic percentage up, which I would not do otherwise because avoidable large chops of municipalities are disfavored). But yes, a Pub gerrymander can squeeze out another CD, which I suppose could be "justified" as hewing to proportionality (7D-5R), which has been invoked to support what would otherwise be Dem gerrymanders (PA and MI and NC (until redrawn), come to mind).

[Torie]

Counties trump munis in most places That said, in NJ to chop the Newark muni is rather compelling to up in Hispanic percentage in the Hudson County based CD, and the black percentage in the Essex County based CD. And thus it is still justified, even if Munis rule ala New England over counties. It does not do much violence to the map in any event, and none to the division of the partisan spoils.

[Torie]

Well, drawing CD lines in my current home state of NJ is always a challenge, but I decided to do the exercise based on neutral metrics, and that hack tie breaker retired judge really did snatch a seat from the Pubs, and in a grotesque manner to boot. I guess the Dems knew what was in the guy's brain in advance or something.

What were the 2020 numbers on that 7th?

Per my map, Trump 2020 won by three points.

https://davesredistricting.org/join/af980daa-0afb-4e98-b88a-3d745c7455fd