2020 Census and Redistricting Thread: California

[Greedo punched first]

Doesn't it have to do with citizenship and turnout? Doesn't the Central Valley have lower turnout and more noncitizens? Doesn't low turnout in midterms help Central Valley Republicans like it does in most places?

[SevenEleven]

Doesn't it have to do with citizenship and turnout? Doesn't the Central Valley have lower turnout and more noncitizens? Doesn't low turnout in midterms help Central Valley Republicans like it does in most places?

Obama and Clinton didn't get 55% on the backs of white voters. Yes, the turnout is low and there are a lot of non-citizens, but that is not the reason why Democrats have failed downballot in the Valley. Valadao did just as well there in 2012 and 2016 as he did in 2014.

Again, Valadao is no moderate. For him to be winning 58% consistently in a 55% Clinton district, there has to be other factors at play. The Bakersfield to Fresno connection does not work.

[Greedo punched first]

Doesn't it have to do with citizenship and turnout? Doesn't the Central Valley have lower turnout and more noncitizens? Doesn't low turnout in midterms help Central Valley Republicans like it does in most places?

Obama and Clinton didn't get 55% on the backs of white voters. Yes, the turnout is low and there are a lot of non-citizens, but that is not the reason why Democrats have failed downballot in the Valley. Valadao did just as well there in 2012 and 2016 as he did in 2014.

Again, Valadao is no moderate. For him to be winning 58% consistently in a 55% Clinton district, there has to be other factors at play. The Bakersfield to Fresno connection does not work.

Is this better?

[Oryxslayer]

What's considered an illegal racial pack?

When there is enough Pop in a region to produce a certain number of districts, but you pack one seat to more than capacity to dilute the neighbors. This is best demonstrated in south texas, when the most rather orientation ends up with two uber-hispanic seats along the border and one barely Hispanic seat just to their north. It's very vague in description, but you know it when you see it.

Another good case study is Virginia. The GOP maintained a tentacle that went from Richmond to Norfolk grabbing AAs along the way and making her neighbors whiter. However, there was enough AA voters in the region to elect two candidates of their choosing, but they were being packed into one seat. So the court threw out map.

It's why the GOP in 2010 aimed for just under their states AA% in the south when drawing the number of AA seats on their maps at various levels, since then they could try to dodge this line of attack.

What percentage would count as illegal?

There is no clear percentage. It's a you know it when you see it kind of thing. A African American seat in Mississippi would be fine and legal with 62% of a district being AA, but that would be laughed out in Virginia who is good with something 20 points lower. In CA, the South Valley needs a 71% Hispanic seat (by Pop) to elect a candidate of choice because Hispanics don't vote, but in the Inland Empire you can get away with 58% or so. It's fluid.

If that's so, then are you sure that 80-soemthing % Latino district in the Gateway Cities qualifies? After all, the Downtown district is still majority Latino by CVAP, so it isn't hard to argue that both will elect Latino candidates and the Gateway Cities district therefore is not reducing the electoral influence of the Latino community.

How do you know it doesn't though? The law assumes the worst without proof. Proof is easier obtained when there is a partisan divide between the groups. It's not just dems who need to abide by this - check out the Hispanic Percentage in each the Miami Hispanic seats. In the absence of proof, you must start with the must basic standards and work from there.

I don't but the way you described it makes me think it only becomes a problem if drawing compact districts with an exceptionally high minority % is only a problem if it means minorities miss out on control of districts they would otherwise have. It doesn't seem so clear to me that margins matter as long as they're all majority Latino. For example, drawing that 85% Latino district seems like it would be okay since the Downtown district is over 50%, but if it dropped Downtown to 40%, then it would become a problem.

If you have a good amount of time on your hands, I would go read Johnson v. De Grandy (1994), a subsequent case to the well known Gingles court ruling. Johnson is suprisingly relevant to LA here since it concerns Hispanics in Miami-Dade who are also dealing with the differences between VAP and total Pop. The big finding of the case was that maximization of minority seats, e.g. say you gerrymandering minorities to 51% in more seats than their justification-wide%, is not protected, encouraged, or subject to section 2 Gingles claims. However, relevant to our discussion is the arguments about proportionality. When discussing minority districts in a similar area (M-D), minority populations must have a proportional opportunity to elect candidates of their choice - they are not entitled to proportional outcomes. As long as districts respect Gingles test 1, that districts do not sprawl erroneously (remember Fajitas have been upheld and districts thrown out under this clause include NC13 and some Louisiana seats from the 90s), minorities should not be erroneously diluted. Near proportionality of minorities is opportunity after all, though Hispanics can elect non-Hispanics and visa versa.

With this stated, I think your suggestion of the LA river as a divider probably works best, or at least something a bit more North-South-ish. If we do N-S there is also a option of putting Montebello in the Asian seat and giving NELA to the twins. This is mostly an equitable trade, though CA26 still needs to cut LA for Hermon.

[SevenEleven]

Doesn't it have to do with citizenship and turnout? Doesn't the Central Valley have lower turnout and more noncitizens? Doesn't low turnout in midterms help Central Valley Republicans like it does in most places?

Obama and Clinton didn't get 55% on the backs of white voters. Yes, the turnout is low and there are a lot of non-citizens, but that is not the reason why Democrats have failed downballot in the Valley. Valadao did just as well there in 2012 and 2016 as he did in 2014.

Again, Valadao is no moderate. For him to be winning 58% consistently in a 55% Clinton district, there has to be other factors at play. The Bakersfield to Fresno connection does not work.

Is this better?

It's better than the current map. I prefer to make a Visalia-Fresno county Hispanic district as well though. What are the CA-18 stats?

[Greedo punched first]

Doesn't it have to do with citizenship and turnout? Doesn't the Central Valley have lower turnout and more noncitizens? Doesn't low turnout in midterms help Central Valley Republicans like it does in most places?

Obama and Clinton didn't get 55% on the backs of white voters. Yes, the turnout is low and there are a lot of non-citizens, but that is not the reason why Democrats have failed downballot in the Valley. Valadao did just as well there in 2012 and 2016 as he did in 2014.

Again, Valadao is no moderate. For him to be winning 58% consistently in a 55% Clinton district, there has to be other factors at play. The Bakersfield to Fresno connection does not work.

Is this better?

It's better than the current map. I prefer to make a Visalia-Fresno county Hispanic district as well though.

CA19 is a Fresno Hispanic district. CA20 is particularly compact.

[SevenEleven]

Doesn't it have to do with citizenship and turnout? Doesn't the Central Valley have lower turnout and more noncitizens? Doesn't low turnout in midterms help Central Valley Republicans like it does in most places?

Obama and Clinton didn't get 55% on the backs of white voters. Yes, the turnout is low and there are a lot of non-citizens, but that is not the reason why Democrats have failed downballot in the Valley. Valadao did just as well there in 2012 and 2016 as he did in 2014.

Again, Valadao is no moderate. For him to be winning 58% consistently in a 55% Clinton district, there has to be other factors at play. The Bakersfield to Fresno connection does not work.

Is this better?

It's better than the current map. I prefer to make a Visalia-Fresno county Hispanic district as well though.

CA19 is a Fresno Hispanic district. CA20 is particularly compact.

A more compact district will perform better, which is what my long-winded post was hinting at. What are your CA-18 stats?

[Greedo punched first]

Doesn't it have to do with citizenship and turnout? Doesn't the Central Valley have lower turnout and more noncitizens? Doesn't low turnout in midterms help Central Valley Republicans like it does in most places?

Obama and Clinton didn't get 55% on the backs of white voters. Yes, the turnout is low and there are a lot of non-citizens, but that is not the reason why Democrats have failed downballot in the Valley. Valadao did just as well there in 2012 and 2016 as he did in 2014.

Again, Valadao is no moderate. For him to be winning 58% consistently in a 55% Clinton district, there has to be other factors at play. The Bakersfield to Fresno connection does not work.

Is this better?

It's better than the current map. I prefer to make a Visalia-Fresno county Hispanic district as well though.

CA19 is a Fresno Hispanic district. CA20 is particularly compact.

A more compact district will perform better, which is what my long-winded post was hinting at. What are your CA-18 stats?

My CA-18 is Nunes' seat.

[SevenEleven]

Doesn't it have to do with citizenship and turnout? Doesn't the Central Valley have lower turnout and more noncitizens? Doesn't low turnout in midterms help Central Valley Republicans like it does in most places?

Obama and Clinton didn't get 55% on the backs of white voters. Yes, the turnout is low and there are a lot of non-citizens, but that is not the reason why Democrats have failed downballot in the Valley. Valadao did just as well there in 2012 and 2016 as he did in 2014.

Again, Valadao is no moderate. For him to be winning 58% consistently in a 55% Clinton district, there has to be other factors at play. The Bakersfield to Fresno connection does not work.

Is this better?

It's better than the current map. I prefer to make a Visalia-Fresno county Hispanic district as well though.

CA19 is a Fresno Hispanic district. CA20 is particularly compact.

A more compact district will perform better, which is what my long-winded post was hinting at. What are your CA-18 stats?

My CA-18 is Nunes' seat.

Ok but what's the ethnic breakdown and partisanship. It looks like all of Fresno is there so it would be pretty unfair to have a large city dominated by Clovis and a horde of mountain people.

[Greedo punched first]

Doesn't it have to do with citizenship and turnout? Doesn't the Central Valley have lower turnout and more noncitizens? Doesn't low turnout in midterms help Central Valley Republicans like it does in most places?

Obama and Clinton didn't get 55% on the backs of white voters. Yes, the turnout is low and there are a lot of non-citizens, but that is not the reason why Democrats have failed downballot in the Valley. Valadao did just as well there in 2012 and 2016 as he did in 2014.

Again, Valadao is no moderate. For him to be winning 58% consistently in a 55% Clinton district, there has to be other factors at play. The Bakersfield to Fresno connection does not work.

Is this better?

It's better than the current map. I prefer to make a Visalia-Fresno county Hispanic district as well though.

CA19 is a Fresno Hispanic district. CA20 is particularly compact.

A more compact district will perform better, which is what my long-winded post was hinting at. What are your CA-18 stats?

My CA-18 is Nunes' seat.

Ok but what's the ethnic breakdown and partisanship. It looks like all of Fresno is there so it would be pretty unfair to have a large city dominated by Clovis and a horde of mountain people.

It contains half of Fresno. Much of Fresno is in CA-19.

[Idaho Conservative]

Now the money question - the Empire. The goal here is to get a fourth performing Hispanic seat out of the region, since we are following the suburbanization of minorities. Reminder the preforming means protected seat because a majority of the electors share a demographic group and will be selecting a candidate who best suits the interests of their group. This 4th seat is to compensate for cutting a Hispanic seat in LA. All told, my aim is to get between 16 and 18 seats where Hispanics have domination of the electoral process, and are not just one of many actors a total number of seats that lines up with Hispanic CVAP% statewide.

The borders in Riverside and between the seats can change because of what ends up happening in San Diego. This should be stated in advance.

So with that out of the way lets talk about the four. The biggest problem appears to be the Victor Valley. Victor Valley is better paired with the Imperial valley, I agree with that 100%. However, I don't want west LA getting sucked into Santa Clarita and Ventura to the tune of 350ishK because that's how we end up with odd white packs or minority districts that cannot preform. If I was to restore the old CA08 in some capacity then that wouldn't elect a minority candidate, the desert's too white. It's also not like the commission minds using I15 to connect a district or drawing something weird in the region. Crossing the mountains here using the eastern Artery is just as fine as using the western one to link Santa Clarita to the Antelope Valley, if that is one's discretion. With that in mind then, the Victor Valley needs to stay within the grouping of four, but it doesn't need to be paired with SB city.

Now, there are some changes that I can make to the grouping of four. When changes come to SB, they will bring changes to CA35. For example, Yucapia/Calimesa and the Arrowheads are only in the group because of connections or pop, I would trade them for Apple Valley in a heartbeat. So if I was to adjust how the four districts were obtained from this region, what would people like to see? How would you like to see the communities paired?

Not sure that is possible or justifiable.  The IE is half hispanic and the citizen VAP is 41% (the actual electorate is likely closer to 1/3 hispanic).  There are roughly 6 seats in the IE, 2/3 seats for 1/3 of the electorate sounds like a racial gerrymander to me.  Also it could be a vra dummymander where in a midterm, hispanic choice candidates fall like dominoes.  I think the best case would be 3 seats.  2 in San Bernardino, and I combining the Coachella valley and Perris.  And those would certainly perform.

[SevenEleven]

Some things we might be getting wrong:

1. Two AA districts in LA. At best you have two weak pluralities. Perhaps try to consolidate one and give another access or influence. This loonses up the whole map.

2. Not crossing the mountains into the SFV. This also opens up a lot of opportunities, specifically from a ethnicity perspective.

3. OC-San Diego. This one goes out to me personally. I realized that even when I start out in LA County, my configurations take on the assumption of no OC-SD. This has already been done, so it is definitely on the table. I don't like it, but I need to be more open to it.

4. The Central Valley. You can draw two respectable 50+% Hispanic CVAP districts in the South valley without connecting Kern and Fresno. This is worthy.

5. Majority-minority districts in the Bay Area. I'm in favor of them and would like to see more. However, there's no real bloc voting in the bay, per Gingles, and it's easy to get away with pluralities and focus on COIs foremost.

6. Connecting Imperial to Coachella Valley. It's better than the current map, but make sure not to focus too much on the county lines if you do. Either go north into SB County or head west and take in the SD mountains.

7. Monterey County. As Oryx said, this county should largely be in a Hispanic seat. Splitting the county can often be preferable to keeping it whole, though if you do this, you must also attach San Benito.

[EastAnglianLefty]

What's considered an illegal racial pack?

When there is enough Pop in a region to produce a certain number of districts, but you pack one seat to more than capacity to dilute the neighbors. This is best demonstrated in south texas, when the most rather orientation ends up with two uber-hispanic seats along the border and one barely Hispanic seat just to their north. It's very vague in description, but you know it when you see it.

Another good case study is Virginia. The GOP maintained a tentacle that went from Richmond to Norfolk grabbing AAs along the way and making her neighbors whiter. However, there was enough AA voters in the region to elect two candidates of their choosing, but they were being packed into one seat. So the court threw out map.

It's why the GOP in 2010 aimed for just under their states AA% in the south when drawing the number of AA seats on their maps at various levels, since then they could try to dodge this line of attack.

What percentage would count as illegal?

There is no clear percentage. It's a you know it when you see it kind of thing. A African American seat in Mississippi would be fine and legal with 62% of a district being AA, but that would be laughed out in Virginia who is good with something 20 points lower. In CA, the South Valley needs a 71% Hispanic seat (by Pop) to elect a candidate of choice because Hispanics don't vote, but in the Inland Empire you can get away with 58% or so. It's fluid.

I'm not sure you are understanding the fundamentals of the southern Central Valley properly. David Valadao wasn't getting 57% in a 55% Obama district because Hispanics don't vote. How do you think Obama got 55% in CA-21 in the first place?

Trump got less than 40% and Valadao still won in 2016.

Oh some Hispanics are voters, no sh**t, I'm not dumb. Valadao was good with a handful of crossover voters. But compared to other areas? Well....

Taking a look at DRA now, Lets use Corcoran in Kings as an example. All of it's precincts and the one surrounding the city come to 23.4K pop. It's about 69% Hispanic by pop. All told there are about 2.7K votes in the town. You can get that many votes from 10K pops worth of Hispanic precincts in LA. The truth is that Hispanics in the region are statically likely to be not voters.

Now, why are they not voters? Well, there's a whole lot of reasons which I probably don't need to explain to you ranging from citizenship issues to disconnection from government and outreach.

Maybe we are talking past each other here and both understand whats going on just using different language....

Out performing Trump by 17 points is not a "handful of crossover votes" and shouldn't be handwaved.

It's true that many Hispanics do not or can not vote. That is a fact that can not be disputed. However, the number of Hispanics in the district being too low is not the reason that Democrats have failed to win some of these districts. Hillary didn't win by 15 points in CA-21 because Hispanics don't vote.

Before David Valadao was elected to the House in 2012, he served in the State Assembly. Now, if you look at that 2010 election you will notice it's a Republican hold. If you go back to 2008, you will see it won by Republican Danny Gilmore--a Republican gain. Yes, the Republicans actually gained a seat in California in 2008 of all years. You might also notice that Valadao won that seat in 2010 by over 20 points! In a year where California Democrats picked up two seats, one of which was the seat vacated by Juan Arambula, a Fresno Democrat who became an Independent in 2009. All of the aforementioned seats have consistently voted Democratic at the presidential level with pretty hefty margins.

Jim Costa almost lost his Berman-mander seat to Andy Vidak in 2010, a near unfathomable feat. The district was a 60% Obama district. He bailed on his previous district after the new maps were released and headed north to a less Hispanic district, a smart move. Andy Vidak went on to win a State Senate seat and hold it until 2018, with him and Valadao finally losing because Trump really didn't leave a good impression on Hispanics in a rather remarkable way.

Rudy Salas' Assembly district is about a point more Hispanic and about a point more Democratic than the current CA-21, yet he has held it pretty comfortably. A one point partisan difference is not going to turn a Democratic win into a 58% Valadao win. The problem with CA-21 is that it spans from Kern County to Fresno, taking in Kings, and there is no unifying theme besides "Latinos". Not all Latinos have the same interests, however. Much like what we see in Los Angeles, with pro-labor and pro-business Hispanics often facing each other in run-offs, with some fairly interesting results, the same phenomenon is noticable in the south Valley. Pro-labor, to these Hispanics, tends to mean pro-farming, pro-water, etc. A candidate like John Hernandez (Hispanic Chamber of Commerce) is not going to be received as well by these rather conservative Hispanics as an actual farmer, like Valdao or Vidak. Likewise, a pro-labor candidate such as Emilio Huerta is going to struggle heavily with those pro-business Hispanics. Add in a 40% or so block of inflexible white Republicans and it's a recipe for disaster.

The solution, as seen in Salas' Assembly district, is to keep it simple. Don't draw a district trying to make it as Latino as possible. Draw a locally-oriented district that respects COIs. CA-21 was a failure of the current map because it did not allow Valley Hispanics to elect a candidate of their choice. Not by virtue of having too few Hispanics, as the top-level results show, but because connecting groups that have no business being lumped together means a "candidate of choice" is literally impossible. IIRC, TJ Cox had the narrowest victory of any of the CA House pickups in 2018. A 55% Clinton district struggled more to elect a Democrat than some 55% Romney districts in Orange County.

If you were trying to expand this into a more general point which might conceivably be applicable outside the Central Valley, would it be fair to say that it's much more difficult to make a VRA district perform when the relevant community only votes 70-30 in favour of its chosen candidate, rather than 90-10, and that making the district more disparate actually promotes the former?

[Greedo punched first]

Visalia-Bakersfield Hispanic seat. I instead used a Bakersfield-Fresno connection for a white seat. CA-18 and CA-20 are both performing seats. Fresno is in CA-18, CA-19, and CA-21. Clovis and a tiny piece of Fresno proper are in CA-19.

[Idaho Conservative]

Visalia-Bakersfield Hispanic seat. I instead used a Bakersfield-Fresno connection for a white seat. CA-18 and CA-20 are both performing seats. Fresno is in CA-18, CA-19, and CA-21. Clovis and a tiny piece of Fresno proper are in CA-19.

Clinton margin in CA 20?

[Greedo punched first]

Visalia-Bakersfield Hispanic seat. I instead used a Bakersfield-Fresno connection for a white seat. CA-18 and CA-20 are both performing seats. Fresno is in CA-18, CA-19, and CA-21. Clovis and a tiny piece of Fresno proper are in CA-19.

Clinton margin in CA 20?

It's only Clinton +6 . I'm working on fixing that.

[SevenEleven]

You should try to put Bakersfield with the rest of Tulare County and Clovis, and push the Kings district into the Fresno farming communities.

[Greedo punched first]

You should try to put Bakersfield with the rest of Tulare County and Clovis, and push the Kings district into the Fresno farming communities.

Do I need to have at least two Central Valley performing seats?

[SevenEleven]

You should try to put Bakersfield with the rest of Tulare County and Clovis, and push the Kings district into the Fresno farming communities.

Do I need to have at least two Central Valley performing seats?

Due to demographic changes, I would say yes indeed.

[Greedo punched first]

I got two performing seats.

[SevenEleven]

Don't do Bakersfield to Fresno.

[lfromnj]

What are the counties that are defined as part of the CV?, for the 2 performing seats.

[Greedo punched first]

Should I put Inyo with Kern or San Bernardino?

[SevenEleven]

What are the counties that are defined as part of the CV?, for the 2 performing seats.

San Joaquin, Stanislaus, Merced, Madera, Fresno, Tulare, Kings and Kern.

[SevenEleven]

This alignment is better IMO. You can play with splitting Visalia to even out districts 21 and 22 if you prefer.

CA-22 64.8% Hispanic/51.6% CVAP/49.6% Clinton
CA-21 71.0% Hispanic/56.7% CVAP/52.9% Clinton
CA-16 53.6% Hipsanic/42.0% CVAP/54.1% Clinton
CA-10 46.9% Hispanic/34.6% CVAP/48.3% Clinton

Splitting Visalia right down the middle puts both CA-21 and CA-22 at 54% Hispanic CVAP and 51% Clinton.