How Democrats Fooled California’s Redistricting Commission

[muon2]

edit: I think I did misread sbane's comments, but I really don't want to put Mountain View in that district. I was trying to provide somewhat more separation between the Hispanic and Asian areas and that throws them together more.

Here is what I would do with your cut in Fremont and Newark. It makes it easier to draw a rational Asian heavy mostly middle class district. The 15th is 44.3% AVAP and the 16th (Cyan) is 37% HVAP.

You removed the area of Fremont I included north of Newark. I need that for population balance in Alameda county and it's heavily Asian as well. I assume that you would adjust your map to remove blocks in SJ south of Santa Clara.

[muon2]

It's about 15,000 people it looks like? Putting it in and taking out more of San Jose north of Campbell leads to it becoming 45.4%AVAP.

But, as Torie has implied the WCVAP would be significant higher than the ACVAP. WCVAP increases by about 4/3 compared to WVAP. The Latino citizen rates are higher than the Asian rates in that area.

[muon2]

It's about 15,000 people it looks like? Putting it in and taking out more of San Jose north of Campbell leads to it becoming 45.4%AVAP.

But, as Torie has implied the WCVAP would be significant higher than the ACVAP. WCVAP increases by about 4/3 compared to WVAP. The Latino citizen rates are higher than the Asian rates in that area.

That's fine with me. It's an influence district but really it's mostly a division based on class.

One reason I mixed in Evergreen was to cement the influence. The citizen rate among voting age Asians in the lower valley is less than 80%, but it rises to 94% in Evergreen.

[muon2]

It's about 15,000 people it looks like? Putting it in and taking out more of San Jose north of Campbell leads to it becoming 45.4%AVAP.

But, as Torie has implied the WCVAP would be significant higher than the ACVAP. WCVAP increases by about 4/3 compared to WVAP. The Latino citizen rates are higher than the Asian rates in that area.

That's fine with me. It's an influence district but really it's mostly a division based on class.

One reason I mixed in Evergreen was to cement the influence. The citizen rate among voting age Asians in the lower valley is less than 80%, but it rises to 94% in Evergreen.

By lower valley do you mean the Asian areas right south of Milpitas? What about the Asian areas along the 101 (so a little farther from the hills) south of 680? Where are you getting this data from?

I extrapolated from the Commission's data matched by the MALDEF submission. The district including Sunnyvale, Santa Clara, Northern SJ, Milpitas, and Fremont/Newark has an AVAP of 50% and ACVAP of 40% so the effective share is 80% due to citizenship. The real citizenship rate is lower but since our district covers much the same area the shift in percentage is appropriate to measure. Similarly the Evergreen and areas south of 680 are in a commission district where the fractional dropoff in the CVAP fraction was much less.

[muon2]

It's about 15,000 people it looks like? Putting it in and taking out more of San Jose north of Campbell leads to it becoming 45.4%AVAP.

But, as Torie has implied the WCVAP would be significant higher than the ACVAP. WCVAP increases by about 4/3 compared to WVAP. The Latino citizen rates are higher than the Asian rates in that area.

That's fine with me. It's an influence district but really it's mostly a division based on class.

One reason I mixed in Evergreen was to cement the influence. The citizen rate among voting age Asians in the lower valley is less than 80%, but it rises to 94% in Evergreen.

By lower valley do you mean the Asian areas right south of Milpitas? What about the Asian areas along the 101 (so a little farther from the hills) south of 680? Where are you getting this data from?

Similarly the Evergreen and areas south of 680 are in a commission district where the fractional dropoff in the CVAP fraction was much less.

Perhaps that is because the district also contains many Hispanics and so the drop from AVAP to ACVAP is not as great in this district because citizenship rates of Hispanics is even lower.

No the difference is in the Asian rates alone. The northern commission district I mentioned (17) has an ACVAP pop of 145,669 and AVAP pop of 267,863 or 54.3%. The district that includes Evergreen (19) has an ACVAP pop of 102,286 and AVAP pop of 143,387 or 71.3%. The populations are substantially different on their own.

[muon2]

It's about 15,000 people it looks like? Putting it in and taking out more of San Jose north of Campbell leads to it becoming 45.4%AVAP.

But, as Torie has implied the WCVAP would be significant higher than the ACVAP. WCVAP increases by about 4/3 compared to WVAP. The Latino citizen rates are higher than the Asian rates in that area.

That's fine with me. It's an influence district but really it's mostly a division based on class.

One reason I mixed in Evergreen was to cement the influence. The citizen rate among voting age Asians in the lower valley is less than 80%, but it rises to 94% in Evergreen.

By lower valley do you mean the Asian areas right south of Milpitas? What about the Asian areas along the 101 (so a little farther from the hills) south of 680? Where are you getting this data from?

Similarly the Evergreen and areas south of 680 are in a commission district where the fractional dropoff in the CVAP fraction was much less.

Perhaps that is because the district also contains many Hispanics and so the drop from AVAP to ACVAP is not as great in this district because citizenship rates of Hispanics is even lower.

No the difference is in the Asian rates alone. The northern commission district I mentioned (17) has an ACVAP pop of 145,669 and AVAP pop of 267,863 or 54.3%. The district that includes Evergreen (19) has an ACVAP pop of 102,286 and AVAP pop of 143,387 or 71.3%. The populations are substantially different on their own.

That's pretty interesting but I still don't see any need to racially gerrymander the Silicon Valley. Did the commission find evidence of racially polarized voting? Otherwise I don't see why it would be necessary. If we can draw a rational district and get the Asian population up then that is great. That is what I drew out with your cut in Fremont/Newark. It was 45%AVAP. I don't see why we need to make it even more robust by creating that erose appendage into Evergreen.

From what I read, the commission didn't look at the question there, and with two significant minority populations it surprises me. They did go ahead and make a 50% AVAP district, similar to mine but with Cupertino instead of Evergreen. My original plan used Cupertino as well, but I didn't like the split of the west side, requiring a road link back to Santa Cruz.

Here's my final revision (I think) to match the Commission's AVAP. There are minor shifts in Fremont and in Evergreen and along the 101.

[muon2]

From what I read, the commission didn't look at the question there, and with two significant minority populations it surprises me. They did go ahead and make a 50% AVAP district, similar to mine but with Cupertino instead of Evergreen. My original plan used Cupertino as well, but I didn't like the split of the west side, requiring a road link back to Santa Cruz.

Here's my final revision (I think) to match the Commission's AVAP. There are minor shifts in Fremont and in Evergreen and along the 101.

Well, I still don't agree with the Evergreen split but I will try to work with it. At least remove the parts of the Asian district that crosses the 101 (or US-101 as Xahar might prefer to say) and if possible get it east and south of Capitol Expressway as well. To make up the population you should pick up those areas right south of where your CD-17 stops in Evergreen and pick up the population there up to the 101 and stop there.

The area across 101 is 58% AVAP. The area south of my border in Evergreen drops to 37% AVAP, so that doesn't work. I could take a small area in SJ between Santa Clara and I-280, but that doesn't strike me as all that attractive. I am able to chop up Fremont with some more erosity and meet my goals in the map below. Let me know if that is more acceptable.

[muon2]

This is my updated map for the Bakersfield region. I followed Torie's lead and reduced the muni splits to Bakersfield and Tulare. I kept the HVAP at 65.2% (down from my previous 65.3%). That's the minimum in that area that still gives 50% HCVAP, needed as this is a section 5 and section 2 area.

[muon2]

This is my updated map for the Bakersfield region. I followed Torie's lead and reduced the muni splits to Bakersfield and Tulare. I kept the HVAP at 65.2% (down from my previous 65.3%). That's the minimum in that area that still gives 50% HCVAP, needed as this is a section 5 and section 2 area.

Is 65.2% based on something real, or just Maldef yammering? If it is real, I need to find 30 Hispanic basis points somewhere, probably involving another chop. This is a very interesting CD, by the way, but I digress.

That's the factor I get from MALDEF's table. I've checked their HVAP to HCVAP in other districts and it matches the commission's tables. They probably have the same data set for CVAP. I used MALDEF here because their district is much closer to the shape and area of ours.

[muon2]

From what I read, the commission didn't look at the question there, and with two significant minority populations it surprises me. They did go ahead and make a 50% AVAP district, similar to mine but with Cupertino instead of Evergreen. My original plan used Cupertino as well, but I didn't like the split of the west side, requiring a road link back to Santa Cruz.

Well, I still don't agree with the Evergreen split but I will try to work with it. At least remove the parts of the Asian district that crosses the 101 (or US-101 as Xahar might prefer to say) and if possible get it east and south of Capitol Expressway as well. To make up the population you should pick up those areas right south of where your CD-17 stops in Evergreen and pick up the population there up to the 101 and stop there.

The area across 101 is 58% AVAP. The area south of my border in Evergreen drops to 37% AVAP, so that doesn't work. I could take a small area in SJ between Santa Clara and I-280, but that doesn't strike me as all that attractive. I am able to chop up Fremont with some more erosity and meet my goals in the map below. Let me know if that is more acceptable.

Yes, the area across the 101 is pretty Asian but lower middle class or working class. The areas within Capitol expressway are similar and it would be better if you substituted them for those more affluent areas to the south of the appendage. How far would it drop the AVAP? And it's not as if you have to stick to any specific number since the VRA is not involved here.

I think I already have some working class areas by including Newark, so including the area within the Capitol Expy doesn't seem out of place. I understand that the VRA may not be involved, but I would feel more secure if the commission had made that finding. They didn't and they went ahead with a 50% AVAP district. I've taken a cautious approach in making VRA-related decisions, so I'm going to continue that approach here as well.

[muon2]

It took a while, but I can now compare the two alternatives to the Commission's map. DRA only has the 2008 Pres and 2010 Sen data, and I wanted to build PVI's to compare the three plans to a known standard. I took the 2001 map and created PVIs using Atlas data to form a 2008 baseline for each area of the state. The two-party Dem base line varied from 52.1% in SF to 57.4% in Anaheim.

Then I matched up the plan districts to the core areas, so I could use an appropriate baseline to get a PVI. The long step was constructing the Commission's plan in DRA so I could compare it with our two plans. For each plan I counted the number of strong (PVI 6+), lean (PVI 2-5) and even (PVI 0-1) districts. As with other states I can get a partisan differential counting lean as half a point.

Commission: 27 SD, 9 LD, 1 E, 5 LR, 11 SR; Partisan differential +18.
Torie: 27 SD, 7 LD, 3 E, 4 LR, 12 SR; Partisan differential +16.5.
Muon2: 29 SD, 4 LD, 2 E, 8 LR, 10 SR; Partisan differential +17.

Based on the state PVI, an ideal fair map would be +8 and have 14 seats either lean or even. All three plans are suitably competitive, and one can see that the ideal partisan fairness isn't achieved due to the spread of Pubs across the state in a way that resists grouping in districts without gerrymandering (think MA as the extreme example). For reference, the bipartisan gerrymander in place now is +14.5, so any of these plans is more Dem by comparison.

[muon2]

SD will be an interesting case study. SD + Imperial have a PVI of nearly 0. Together the population is 244 K less than 5 CDs and there are a number of identifiable interests that can be grouped together. Such an even starting point lends itself to different political results. Here's how we did on the 5 CDs.

The commission has R+5, R+15, D+11, D+1, D+8. Total R+0.
Torie has R+11, R+11, D+8, D+1, D+10. Total R+3.
Muon2 has R+5, R+10, D+8, D+8, R+3. Total R+2.

As a partisan on the commission, do you push for one of these outcomes over the others? Can you know the political tendencies? The total PVI for the five districts depends on the areas added. The third district on the above lists is potentially VRA, so do you push for a certain type of VRA district to help the surrounding ones?