538 district maps series - California

[bagelman]

Now we arrive at the most populated state

https://projects.fivethirtyeight.com/redistricting-maps/california/

[Anti-Bothsidesism]

This was harder to choose than I'd have thought.

[Nyvin]

For the "Favor Democrats" map...the San Diego districts look REALLY neat and tidy!


The Sacramento district look like crap though.

[America's Sweetheart ❤/𝕿𝖍𝖊 𝕭𝖔𝖔𝖙𝖞 𝖂𝖆𝖗𝖗𝖎𝖔𝖗]

I like to compact one, followed by the highly competitive one.

[Starry Eyed Jagaloon]

It's hard to judge without a zoom in on Los Angeles, but I generally like the algorithmic ones.

[Tintrlvr]

Compact on county lines is mostly very good except that I really, really hate that it has two districts crossing the Coast Range north of the Bay Area. It's totally unnecessary and a massive mistake from a communities of interest perspective. The max minority-majority districts map looks decent at a glance, too, though it's hard to tell without zooming in.

[King Lear]

I voted for the compact county-borders map, because I think all districts should be drawn in a fair, Non-partisan matter, and I think it’s the most aesthetically pleasing.

[Former President tack50]

^

Similar except I voted for the algorithmic compact map because for this one the county borders one doesn't really look nicer to me

[muon2]

In 2012 the muon rules weren't fully formed when we took on CA. The major difference was the UCC rules for metro areas weren't defined yet and districts were made more precise on population due to "microchops". But the rules evebn then were requiring VRA districts, and minimizing chops (except for microchops) and reducing erosity. Essentially a blend of whole counties and minority districts from 538. SoCal looks like it might be pretty much the same even with the updated rules, so here's that post.

This is the southern region of CDs. The counties closely match Torie's wall and is only 1019 over population for 31 CDs. The counties can be divided into three groups: Kings-Tulare-Kern-Mono-Inyo-SanB (5 CDs), LA (14 CDs), Riverside-Orange-SanDiego-Imperial (12 CDs), and each of these groups is close enough in population to a whole number of CDs that they could all be drawn within 0.5% of the ideal size without crossing out of their group of counties.

In this map, populations are with 100 of the ideal (except 24 at -119) so the microchop rule is used on four fragments in LAC (22, 24, 26, 27) and one in Riverside (25). There is one extra fragment due to the VRA requirements for CD 23 at 65.2% HVAP and it is over 50% HCVAP.

CD 53 uses the excess in Riverside in Coachella to avoid any part of the city of San Diego and get 65.5% HVAP. That allows all of the city of San Diego to be only in either CD 50 or 52.

The minority CDs are as follows:

CD 23: 65.2% HVAP
CD 27: 63.1% HVAP
CD 29: 63.9% HVAP
CD 32: 61.3% HVAP
CD 34: 64.5% HVAP
CD 35: 51.0% AVAP
CD 36: 68.4% HVAP
CD 37: 43.3% BVAP/44.9% HVAP (BCVAP majority)
CD 38: 67.5% HVAP
CD 39: 72.1% HVAP
CD 42: 51.3% HVAP
CD 46: 65.0% HVAP
CD 53: 65.5% HVAP

[Gass3268]

The current borders are the best and it's process should be replicated nationwide.

[Oryxslayer]

This one is tough. I really like the current map, and it works fine. However, the issues that I have with it are solved in other maps in the series - like the long beach seat reaching into OC or the Imperial-SD link when a HVAP seat can be drawn safely in SD county. However, the other contender, the minimize cuts map, doesn't respect race and doesn't appear to be prioritizing local lines in this instance. So current map.

Also, it is rare, but I think Moun has actually produced a map worse than the current one.

[muon2]

This one is tough. I really like the current map, and it works fine. However, the issues that I have with it are solved in other maps in the series - like the long beach seat reaching into OC or the Imperial-SD link when a HVAP seat can be drawn safely in SD county. However, the other contender, the minimize cuts map, doesn't respect race and doesn't appear to be prioritizing local lines in this instance. So current map.

Also, it is rare, but I think Moun has actually produced a map worse than the current one.

I didn't think mine was so different from the current map except that I don't mix LA and OC counties except for one small part of Brea, and the fragment from Riverside connecting south is at Coachella instead of Temecula. I'm curious what part you found worse?

[Torie]

This one is tough. I really like the current map, and it works fine. However, the issues that I have with it are solved in other maps in the series - like the long beach seat reaching into OC or the Imperial-SD link when a HVAP seat can be drawn safely in SD county. However, the other contender, the minimize cuts map, doesn't respect race and doesn't appear to be prioritizing local lines in this instance. So current map.

Also, it is rare, but I think Moun has actually produced a map worse than the current one.

I didn't think mine was so different from the current map except that I don't mix LA and OC counties except for one small part of Brea, and the fragment from Riverside connecting south is at Coachella instead of Temecula. I'm curious what part you found worse?

Granted the map adopted does the same thing, but what was the rationale again for combining Imperial County with the Hispanic hoods along the San Diego area coast?

[Oryxslayer]

This one is tough. I really like the current map, and it works fine. However, the issues that I have with it are solved in other maps in the series - like the long beach seat reaching into OC or the Imperial-SD link when a HVAP seat can be drawn safely in SD county. However, the other contender, the minimize cuts map, doesn't respect race and doesn't appear to be prioritizing local lines in this instance. So current map.

Also, it is rare, but I think Moun has actually produced a map worse than the current one.

I didn't think mine was so different from the current map except that I don't mix LA and OC counties except for one small part of Brea, and the fragment from Riverside connecting south is at Coachella instead of Temecula. I'm curious what part you found worse?

Coachella cut - its better to ignore cutting riverside from the south altogether and send the SD seats into OC through the 49th. LA will pick up the extraneous pop around SB/Riverside.  The cut into Antelope from the north - it only does this right now because of the military bases. 22-24 overall could probably do with a redraw, but that isn't really part of SoCal. I can't tell from here, but it looks like you cut a bunch of towns in LA - what is the point of sending the Long Beach seat and cutting the LA Port? Palos Verdes is better with the towns to the north of it and the other White communities along the West Shore. 33/30 need a redraw, the hollywood hills should be in one full seat.I can't tell, but is the 31st connect through the San Gabriel mountains?

[muon2]

This one is tough. I really like the current map, and it works fine. However, the issues that I have with it are solved in other maps in the series - like the long beach seat reaching into OC or the Imperial-SD link when a HVAP seat can be drawn safely in SD county. However, the other contender, the minimize cuts map, doesn't respect race and doesn't appear to be prioritizing local lines in this instance. So current map.

Also, it is rare, but I think Moun has actually produced a map worse than the current one.

I didn't think mine was so different from the current map except that I don't mix LA and OC counties except for one small part of Brea, and the fragment from Riverside connecting south is at Coachella instead of Temecula. I'm curious what part you found worse?

Granted the map adopted does the same thing, but what was the rationale again for combining Imperial County with the Hispanic hoods along the San Diego area coast?

The Commission's justification was to have one CD with a common feature of the Mexican border. My recollection is that it was hard to get HCVAP over 50% (HVAP > 65%) in SD alone without some serious gerrymandering that chopped up cities and sent tentacles all over the place.

[Torie]

That is an extraordinarily high HVAP percentage (65%). Does San Diego have an unusually low percentage of Hispanic citizens? I think one can get into the high 50's without undue erosity if I recall correctly.

[muon2]

That is an extraordinarily high HVAP percentage (65%). Does San Diego have an unusually low percentage of Hispanic citizens? I think one can get into the high 50's without undue erosity if I recall correctly.

I'd have to reread the commission's report. There are tables that can be used to guide estimation of HVAP to HCVAP. My recollection is that the non-citizen rate is quite high in the suburbs south of SD city.

[Torie]

That is an extraordinarily high HVAP percentage (65%). Does San Diego have an unusually low percentage of Hispanic citizens? I think one can get into the high 50's without undue erosity if I recall correctly.

I'd have to reread the commission's report. There are tables that can be used to guide estimation of HVAP to HCVAP. My recollection is that the non-citizen rate is quite high in the suburbs south of SD city.

OK. I hated that Imperial Valley stretch out myself.

[Torie]

This one is tough. I really like the current map, and it works fine. However, the issues that I have with it are solved in other maps in the series - like the long beach seat reaching into OC or the Imperial-SD link when a HVAP seat can be drawn safely in SD county. However, the other contender, the minimize cuts map, doesn't respect race and doesn't appear to be prioritizing local lines in this instance. So current map.

Also, it is rare, but I think Moun has actually produced a map worse than the current one.

I didn't think mine was so different from the current map except that I don't mix LA and OC counties except for one small part of Brea, and the fragment from Riverside connecting south is at Coachella instead of Temecula. I'm curious what part you found worse?

Coachella cut - its better to ignore cutting riverside from the south altogether and send the SD seats into OC through the 49th. LA will pick up the extraneous pop around SB/Riverside.  The cut into Antelope from the north - it only does this right now because of the military bases. 22-24 overall could probably do with a redraw, but that isn't really part of SoCal. I can't tell from here, but it looks like you cut a bunch of towns in LA - what is the point of sending the Long Beach seat and cutting the LA Port? Palos Verdes is better with the towns to the north of it and the other White communities along the West Shore. 33/30 need a redraw, the hollywood hills should be in one full seat.I can't tell, but is the 31st connect through the San Gabriel mountains?

Muon2 tries to avoid city chops, and the VRA and what is a VRA performing Hispanic district, puts a lot of constraints on how the lines work, along with the mix between blacks and Hispanics "fighting" over the same pie often. Muon2 and I ignore the subjective communities of interest game, which is well, "gamed" to death. The Commission had a wealth test (aka class warfare test), which it applied inconsistently, but applied it to draw a CD that went from the beach cities all the way to rich Hancock Park (beautiful stately homes there mostly built before WWII, well worth driving by), near downtown LA.

[muon2]

That is an extraordinarily high HVAP percentage (65%). Does San Diego have an unusually low percentage of Hispanic citizens? I think one can get into the high 50's without undue erosity if I recall correctly.

I'd have to reread the commission's report. There are tables that can be used to guide estimation of HVAP to HCVAP. My recollection is that the non-citizen rate is quite high in the suburbs south of SD city.

OK. I hated that Imperial Valley stretch out myself.

IIRC the commission didn't even try to justify it with the VRA, since they would have to be able to demonstrate that there was an HCVAP majority in a compact area, and rural Imperial is a long way from urban SD. Yet it's perhaps the only way to guarantee a Latino seat there, so expediency may have played a role.

Maybe I'll revisit it with modern estimates to see if it is made obsolete in 2020.

[muon2]

This one is tough. I really like the current map, and it works fine. However, the issues that I have with it are solved in other maps in the series - like the long beach seat reaching into OC or the Imperial-SD link when a HVAP seat can be drawn safely in SD county. However, the other contender, the minimize cuts map, doesn't respect race and doesn't appear to be prioritizing local lines in this instance. So current map.

Also, it is rare, but I think Moun has actually produced a map worse than the current one.

I didn't think mine was so different from the current map except that I don't mix LA and OC counties except for one small part of Brea, and the fragment from Riverside connecting south is at Coachella instead of Temecula. I'm curious what part you found worse?

Coachella cut - its better to ignore cutting riverside from the south altogether and send the SD seats into OC through the 49th. LA will pick up the extraneous pop around SB/Riverside.  The cut into Antelope from the north - it only does this right now because of the military bases. 22-24 overall could probably do with a redraw, but that isn't really part of SoCal. I can't tell from here, but it looks like you cut a bunch of towns in LA - what is the point of sending the Long Beach seat and cutting the LA Port? Palos Verdes is better with the towns to the north of it and the other White communities along the West Shore. 33/30 need a redraw, the hollywood hills should be in one full seat.I can't tell, but is the 31st connect through the San Gabriel mountains?

Muon2 tries to avoid city chops, and the VRA and what is a VRA performing Hispanic district, puts a lot of constraints on how the lines work, along with the mix between blacks and Hispanics "fighting" over the same pie often. Muon2 and I ignore the subjective communities of interest game, which is well, "gamed" to death. The Commission had a wealth test (aka class warfare test), which it applied inconsistently, but applied it to draw a CD that went from the beach cities all the way to rich Hancock Park (beautiful stately homes there mostly built before WWII, well worth driving by), near downtown LA.

Indeed. County and city lines define communities of interest that one can't argue on subjective grounds. Subjective CoIs lead to biased plans - not necessarily partisan bias, but bias nonetheless.

To follow up on Oryxslayers thoughts let's consider the following regions within SoCal (2010 pop):
LAC is 22K short of 14 CDs;
OC/SD/Imperial is 46K short of 9 CDs;
Riverside/SB/Inyo/Mono is 40K over 6 CDs;
Kern/Kings/Tulare is 29K over 2 CDs.

A 0.5% maximum deviation is 3.5K, so to minimize county splits these four regions should be grouped so that the deviation can be spread across all CDs staying within the max deviation. For example, LAC could spread its 22K shortfall over the 14 CDs, while the remaining regions spread their 23K overpop among 17 CDs (grouping A). Alternatively, LAC+Kerns/Kings/Tulare can spread a 7K excess over 16 CDs, while the rest spread a 6K shortfall over 15 CDs (grouping B). Both ways minimize county chops equally. Grouping A does have the challenge of meeting the larger population spread which likely means more city chops.

With that said, I interpret Oryxslayer's preference this way: Have the OC/SD/Imperial region take 46K from LAC. Then LAC takes 68K from SB. Finally SB picks up 28K from Kern around Ridgecrest (grouping C). That involves one more chop than either of the two region groupings I described in the previous paragraph.

The other provision in the muon rules is to protect UCCs, and wasn't formed when I drew the linked map. There are two UCCs here: LAC/OC with 18.25 CDs, and Riverside/SB with 6.01 CDs. This suggests that Riverside/SB could be packed neatly, but that would append Mono/Inyo onto Kern and there would be a bigger piece of Kern that would then attach to LAC.

In grouping A the LA UCC meets both the pack and cover, and the SanB UCC is 1 short of the ideal pack. In grouping B the LA UCC and SanB UCC are both 1 short of the ideal pack. In grouping C both UCCs are also 1 short of the ideal pack. Oryxslayer's suggestion is only one chop more, and might make that up on city chops or erosity. It would be worth looking at it from that perspective.

[Torie]

What is the reason that the map was not in the LA area turned clockwise, so CA-35 is shoved out of (or more out of) Brea, CA-26 is shoved out of Claremont, and CA-53 chops less into Riverside County? Avoiding municipal chops or the VRA?

[muon2]

What is the reason that the map was not in the LA area turned clockwise, so CA-35 is shoved out of (or more out of) Brea, CA-26 is shoved out of Claremont, and CA-53 chops less into Riverside County? Avoiding municipal chops or the VRA?

I don't have notes from 2012 that give me a definitive answer, and the map predates codification of the muon rules (microchops were still in play then). I suspect that the VRA played a role along with city chops.

The map would be different with the current muon rules. I put forward the county grouping schemes above to see if commenters had a preference.

A: LAC intact, OC only links to SD, SD or Imp link to Riv, SB link to Kern;
B: LAC link to Kern, OC only links to SD, SD or Imp link to Riv;
C: LAC-OC link, LAC-SB link, Riv only links to SB, SB link to Kern;
D: SB+Riv intact together, LAC-OC link, LAC-Kern link, Kern-Inyo-Mono link

[Torie]

What is the reason that the map was not in the LA area turned clockwise, so CA-35 is shoved out of (or more out of) Brea, CA-26 is shoved out of Claremont, and CA-53 chops less into Riverside County? Avoiding municipal chops or the VRA?

I don't have notes from 2012 that give me a definitive answer, and the map predates codification of the muon rules (microchops were still in play then). I suspect that the VRA played a role along with city chops.

The map would be different with the current muon rules. I put forward the county grouping schemes above to see if commenters had a preference.

A: LAC intact, OC only links to SD, SD or Imp link to Riv, SB link to Kern;
B: LAC link to Kern, OC only links to SD, SD or Imp link to Riv;
C: LAC-OC link, LAC-SB link, Riv only links to SB, SB link to Kern;
D: SB+Riv intact together, LAC-OC link, LAC-Kern link, Kern-Inyo-Mono link

That perhaps is closer to the map that I preferred. It's been a long time now.

[muon2]

What is the reason that the map was not in the LA area turned clockwise, so CA-35 is shoved out of (or more out of) Brea, CA-26 is shoved out of Claremont, and CA-53 chops less into Riverside County? Avoiding municipal chops or the VRA?

I don't have notes from 2012 that give me a definitive answer, and the map predates codification of the muon rules (microchops were still in play then). I suspect that the VRA played a role along with city chops.

The map would be different with the current muon rules. I put forward the county grouping schemes above to see if commenters had a preference.

A: LAC intact, OC only links to SD, SD or Imp link to Riv, SB link to Kern;
B: LAC link to Kern, OC only links to SD, SD or Imp link to Riv;
C: LAC-OC link, LAC-SB link, Riv only links to SB, SB link to Kern;
D: SB+Riv intact together, LAC-OC link, LAC-Kern link, Kern-Inyo-Mono link

That perhaps is closer to the map that I preferred. It's been a long time now.

So which of A, B, C or D do you prefer? I'm hoping Oryxslayer weighs in, too.