How Democrats Fooled California’s Redistricting Commission

[muon2]

After I read the cases, if I am satisfied that a 50% HVAP Riverside CD need not be drawn as a legal imperative, then if to draw it involves as it appears more erosity, it is my opinion that the GOP members should have vetoed it, and that is how I would vote. In other words, in my mind, Riverside is back in play. I might withdraw my gift to the Dems here.  You do the minimum to satisfy the VRA if it otherwise violates appropriate redistricting principles, if it hurts your party.  To me that is rather basic.


My Asian tiger CD has no muni chops at all.

... except for Montebello.

In the meantime I'll see how smooth I can make a Riverside district. I'm still concerned that Riverside meets the Gingles test and something will have to be done there to answer to the VRA.

Yes, one precinct. I forgot that one.  "Answering the VRA" means a current legal mandate, and it appears that there is none for a mere 50% HVAP CD based on our chat. Moreover, since the 9th Circuit decision interpreted Gingles, it is the law, unless Bartlett changed it. Did Bartlett change it? I suggest not, although Kennedy as is sometimes his wont, when not grandiloquent, can be vague and imprecise. For example from Bartlett we have this schizophrenic gem. Given the policy thrust of the prose, as to whether at least theoretically, a minority if unanimous can elect they own candidate without anyone else's help, I see nothing that reverses the 9th circuit decision, and indeed it kind of goes there itself.  

"Unlike any of the standards proposed to allow crossover-district claims, the majority-minority rule relies on an objective, numerical test: Do minorities make up more than 50 percent of the voting-age population in the relevant geographic area? That rule provides straightforward guidance to courts and to those officials charged with drawing district lines to comply with § 2. See LULAC, supra, at 485 (opinion of SOUTER, J.) (recognizing need for "clear-edged rule"). Where an election district could be drawn in which minority voters form a majority but such a district is not drawn, or where a majority-minority district is cracked by assigning some voters elsewhere, then — assuming the other Gingles factors are also satisfied — denial of the opportunity to elect a candidate of choice is a present and discernible wrong that is not subject to the high degree of speculation and prediction attendant upon the analysis of crossover claims. Not an arbitrary invention, the majority-minority rule has its foundation in principles of democratic governance. The special significance, in the democratic process, of a majority means it is a special wrong when a minority group has 50 percent or more of the voting population AND could constitute a compact voting majority but, despite racially polarized bloc voting, that group is not put into a district. [*1246]" (emphasis added)

You see, the thrust here is 50% VAP, that is also 50% voting VAP, or at least eligible to vote, to wit, CVAP. Kennedy while  bouncing erratically in his prose  from population to voters, when he gets down to brass tacks refers to a "compact voting majority." "Voting majority" means "voting majority." not voting age population majority. In other words, folks not eligible to vote don't count. They ain't voters. How can you have a "voting majority," if it is illegal for you to vote?

Kennedy is just a mess isn't it - and in more ways than one.

The 50% HVAP CD in Riverside is dead in my map, unless you think I missed something Mike. The Romero 9th circuit decision is still the governing law, interpreting Gingles, without dilution. Sure drawing a 50% HVAP CD would be legal, but given that it violates other redistricting principles, e.g. going down to Perris, and f's the Pubbies, it needs to be vetoed. Why on earth would the Pubs vote for it? And doesn't it violate your own good redistricting principles, which are while more mechanistic than mine, similar to mine, which is that you do erosity and chops to and only to the extent the VRA demands it? And surely it is not appropriate to alter the shape of CD's on the assumption that  Romero will be reversed (highly unlikely it appears from Kennedy's Bartlett prose) is it?

I've sat through a number of panels listening to legal minds debate this very point. The first part of the citation makes a clear standard as to when Gingles applies. It's 50% VAP, and he says it's to provide "guidance to the courts". The later part is almost a moral statement that decries a "special wrong" that occurs when a district is not formed for the voting population. It doesn't seem to me that that's the clear direction part, and he knew the difference because it's noted elsewhere.

Interestingly, I found that the Pubs on the panels generally argued for the strict VAP definition, while the Dems argued for a CVAP view. Both sides agreed that this question is going back to SCOTUS.

In any case, I again note that the commission did put a 50% VAP district there after citing the Gingles factors at work in Riverside county. I think they actually drew a pretty good district without chops. I missed it due to the mismatch between block group and muni lines. I think I can rework mine to look nice and meet my reading that the first part is the directive. I can't fault you if you read the other part as controlling, you are in good company.

[muon2]

Here's my revision for the IE. CD 42 has dropped to 51.2% HVAP, but all the chops are gone except a small part of Riverside city by Woodcrest (at least as far as block groups permit). Torie may note that Laguna Niguel is now apart from Irvine.

[muon2]

Laguna Niguel may be part of Irvine, but separated from Dana Point? LOL. No.

No it's apart from Irvine, but with Dana Point, Laguna Beach and Oceanside. Did I read the map on the app wrong?

[muon2]

In any event, if giving the Hispanics a 50% VAP pound of flesh (well actually the pound of flesh goes to white Dems, but I digress) does not hurt the Pubs cause by more than a few basis points, and the CD looks as good as yours does (which isn't bad, but still chops the Riverside metro area more in order to race down to distinctly non metro Perris, which is down a canyon and over a hill), I am open to it - provided I get something else in return. I don't give away freebies.

OK, so I peeked at the answers, ie the PVIs using your metric. Here's the question, do you want all three Riverside seats? The county as a whole is R+5 by the usual 2008 and 2004 average, but it's only R+2.5 by your metric. Trying to hold all a three district split is a dangerous game. In my map with your metric I get the districts at D+7, R+3 and R+9. I'm curious what your plan gives - does it still split 2-1 or do you get three R+2 or 3 districts?

[muon2]

So I don't think the Commission itself thinks a 50% VAP Riverside CD is legally necessary. They just did it, to do it, to appease interests hostile to Pub interests, or whatever, with the Pubs asleep at the wheel.

Count me as skeptical. The commission district is 50.21% HVAP. They put Lake Matthews with a split in the Riverside CD and left Woodcrest out, even though it has no good connection to the rest of the district. Had they flipped those two communities the HVAP drops below 50%. Hmm, a coincidence? You tell me.

[muon2]

Here's my revision for the IE. CD 42 has dropped to 51.2% HVAP, but all the chops are gone except a small part of Riverside city by Woodcrest (at least as far as block groups permit). Torie may note that Laguna Niguel is now apart from Irvine.

I see you  chopped the BIG city of San Bernadino in half Mike. Nice Pub gerry! 

Yes I write that off in the same category as the splits of Fresno and Bakersfield - the VRA made me do it. And I've checked that my Chino Hills split really does reduce county chops by one. I also need advice to get a better handle on Riverside PVI's. What would you suggest if all I have in the app are 2008 Pres and 2010 Gov?

BTW, this is area is my counterexample to your distaste for my passage from Santa Clarita into north LA. Why is that bad, when you so willingly make what my eyes see as the same sort of passage across the SBD mountains to get the Yucaipa area? I make the former hop but it lets me avoid the latter hop. Isn't that an even trade?

Speaking of LAC, I have taken your advice to reduce some muni splits on the Asian district. My split of Chino Hills required the addition of parts of West Covina. But I noticed that you had a chop of El Monte that I could eliminate by increasing my West Covina chop. We both have a Pomona chop, so I think that puts me with only one additional chop (Chino Hills) compared to your plan. The whole shift only costs me 0.1% so the AVAP is 51.9%.

[muon2]

Oh, I got rid of my El Monte chop long ago. The only chop for the red tiger is that one precinct in Montebello to equalize population. I also don't like the Chino Hills chop or an OC district going there. That is just too cute by half. Don't worry. The balance of Chino Hills has construction activity going on again, and you know who whom it is designed, with fancy mother-in-laws quarters with its own entrance and separate small kitchen? You guessed it - Asians!

Yucaipa v Chatsworth?  First, two wrongs don't make a right. Second going down the connector road from Big Bear to the valley to nip off a couple of whole towns, to which they often go to shop and stuff not available up on the mountain, is not the same thing as chopping San Bernadino in half.  Surely you know there is zero chance any commission would ever agree to that. I also doubt you would get much support for moving into LA City to an area unconnected with Santa Clarita. No my CA-25 should go north precisely where I sent it. It is by far superior, and the only sensible chop. The rest all suck really. This is a case where you need to throw your computer program out of the window, and that little parameter that putting aside the VRA, county chops are always worse than an extra muni chop, or a county chop to put together munis that belong together (our little Silicon Valley disagreement).

For that matter, when I get finished (well when I think I am finished, until the next complaint comes along ), I will send you my data file, and perhaps you can do the same for me, although I can't seem to reopen anything with a block data base, after I do my first save of it.

I see from my map, that I may have to play with my northern end of CA-33 a tad. I think I see an extra muni chop there that is unnecessary. A couple of three of the  munis up there have these odd little shapes, which is kind of irritating.  So I need to pay with it a little bit, while minimizing erosity.

I can respect your priorities to go with what some of the public see as communities of interest and use that to place muni integrity ahead of county lines. My more detached view is to look at factors that are harder to be manipulated. We can agree to disagree here.

It's interesting looking at the commission's maps, and what they were willing to vote for. For instance they did come into LA from Santa Clarita with what seems a far worse chop than I used, and that solely so they could hop to Simi Valley. So that kind of LA chop has received votes from commissioners.

Likewise, putting Chino Hills with northern OC is what the commission did as part of a larger tri-county district. Of course they didn't split it, but made a split of Upland that seems inconsistent with so many other borders they carefully followed. I'm not convinced it was the only way out, but they certainly voted for it.

We should swap files at some point, I was going off your old work when I thought El Monte was still cut, and my screen resolution didn't reveal the change. My bad. Meanwhile, I've done a major cleanup on my south LA area. I reduced the chops into Long Beach, kept the Torrance district out of the LA city corridor, and kept Gardena whole so only Hawthorne is split in that area.

[muon2]

Oh, I got rid of my El Monte chop long ago. The only chop for the red tiger is that one precinct in Montebello to equalize population. I also don't like the Chino Hills chop or an OC district going there. That is just too cute by half. Don't worry. The balance of Chino Hills has construction activity going on again, and you know who whom it is designed, with fancy mother-in-laws quarters with its own entrance and separate small kitchen? You guessed it - Asians!

We should swap files at some point, I was going off your old work when I thought El Monte was still cut, and my screen resolution didn't reveal the change. My bad. Meanwhile, I've done a major cleanup on my south LA area. I reduced the chops into Long Beach, kept the Torrance district out of the LA city corridor, and kept Gardena whole so only Hawthorne is split in that area.

OK I'm looking at your image and I do still see an El Monte chop in te SW corner. Am I going blind from staring at too many of these maps?

[muon2]

Here's a reworked south side of LAC. There are no muni splits except for LA, Long Beach, and Commerce (and the Montebello nibble). All these districts are within 100 of the ideal population! If this looks at least as reasonable as my previous work, I could be convinced to switch. The BVAP in 34 should still equate to a CVAP over 50%. Here are the VAPs for groups over 10%.

CD 33 Downtown/South LA: 68.5% HVAP, 15.9% BVAP
CD 34 Inglewood/Compton: 44.9% HVAP, 43.3% BVAP
CD 35: Downey/Norwalk: 67.5% HVAP, 19.7% WVAP
CD 36: South Gate/Carson: 72.1% HVAP
CD 37: Torrance/Santa Monica: 55.4% HVAP, 20.7% HVAP, 16.5% AVAP
CD 38: Long Beach: 40.4% WVAP, 29.1% HVAP, 19.0% AVAP

[muon2]

Do we like this version of a Silicon Valley cut better?  It is the white/Asian v. Hispanic chop.

Interesting. What I would have done is got rid of the Hispanic parts of the 15th basically east of CA-87 in exchange for the areas of the 16th northwest of Campbell, Campbell itself, and the areas to the east of it as much is needed. I think that would create a nice east side district.

I don't see the COI in the east side extension of CD 15. Overall it looks like one Asian tiger (my preference) gives way to two Asian plurality districts. It's also hard to judge CD 16 without seeing the southern end. It looks like you'll be chopping Gilroy.

sbane's suggestion would leave neither district with an Asian plurality from what I see.

Is there any way to make a reasonable map that only splits SJ two ways?

[muon2]

Are we going to have a race-class based division of the Silicon Valley, or a geography based division?  You can't have both. Sure I can knock out CA-14 from San Jose, but then Mountain View goes back into CA-14, along with Campbell, and it will still be tight. CA-14 might have to go down to Morgan Hill over some crummy little mountain road to make it happen.

I am open here. I don't know the this part of CA very well at all (south of Palo Alto, and north of Los Gatos). How are we going to resolve this? I am getting "stereophonic" advice.

And yes, I try to balance geography and the minimize chops thing, and race and class where I can. And I focus more on communities, so the net Asian percentage in a CD is not the key, for example, but rather within a community that needs to be chopped. I mean Miltpitas and Fremont are heavily Asian, but have been chopped from now more Asian designed CA-16 from just a San Jose standpoint. Hispanics have been shoved into CA-15 in SJ, yet because Morgan Hill and stuff has to be in CA-16, the Hispanic percentage between the two CD's does not vary too much.

In sum, one cannot minimize chops of SJ, and do the class and race based metric here at the same time. One must choose. My prior map focused more on geography, but not entirely, since I still tri-chopped SJ, just more modestly. And that little jut of SJ to the west south of Cupertino, which looks to be a similar demographic, I am inclined to leave in CA-14, even if it represents a third CD moving into SJ. And CA-14 taking Campbell, which would also be necessary to keep it out of SJ, creates its own erosity (while taking those handful of SJ precincts south of Cupertino, reduces erosity).

So again, how will we resolve this?

Oh, the southern end of CA-16 is unchanged. Gilroy was chopped long ago. We have already had that discussion. And the east-side extension of CA-15 in this version is Hispanic, uniting SJ Hispanics.



Anyway, here is another version, which makes CA-16 40.4% AVAP.

I agree that it is a conundrum. Geography and demography are working at cross purposes in the SV. I was willing to go with a northern Asian tiger and let geography dictate the rest. Even so, in this edition I was left with a bit of SJ between Saratoga and Campbell that I couldn't help but put in a tri-chop. I didn't get a lot of support this year, though I would note that when I floated this concept in 2008 on the forum it got better reviews. I suspect we've gotten more sophisticated as we've watched the last year unfold. This was my original offering.

[muon2]

Here's a version that avoids the tri-chop of SJ and keeps the east side Asian district together at 50.4% AVAP. The SJ Hispanic core areas are kept intact as well (Alum Rock is in though the muni lines make it hard to see. The west side corridor along Ca-85 is maintained. Muni boundaries are respected elsewhere.

 

[muon2]

I put up another map in my post above, and then three more posts intervened. Assuming we do a SJ tri-chop (I just can't get off on Mike's maps here, they are too damned erose for starters), is that a reasonable "chop" of the Gordian knot?

The challenge for erosity is dealing with the shape of the Latino core of SJ. Here's a more compact version that has the SJ tri-chop. How compact do you want?

[muon2]

I think you need to work with my map, and the CA-14, 15 and 16 merry-go-round Mike, or we will be talking past one another. I am not going to trash my entire NoCal map design. Given the outer perimeters of those three districts as drawn, how do you divvy up the spoils? That is the question with my map. But yes, that aside, we are getting closer.

I will happy to do the same working with your map design, although given your very tight metrics, there really aren't many choices out there, are there? The computer drives your map, with mere humans having relatively few choices. Am I wrong about that?

Your challenge is fair and I accept. I'll do analysis of your map to see if there are a reasonable set of options to choose among.

[muon2]

I think you need to work with my map, and the CA-14, 15 and 16 merry-go-round Mike, or we will be talking past one another. I am not going to trash my entire NoCal map design. Given the outer perimeters of those three districts as drawn, how do you divvy up the spoils? That is the question with my map. But yes, that aside, we are getting closer.

I will happy to do the same working with your map design, although given your very tight metrics, there really aren't many choices out there, are there? The computer drives your map, with mere humans having relatively few choices. Am I wrong about that?

Your challenge is fair and I accept. I'll do analysis of your map to see if there are a reasonable set of options to choose among.

I have constructed the following map to illustrate my analysis. I begins by following the 48 K chop out of Alameda, so it should be compatible with the Torie plan.



The lime green area is what seems to be the consensus core of CD 15. The only variable might be how far south to extend from Milpitas. I used the natural division that occurs where the Hispanic population dominates, and without breaking Alum Rock or East Foothills. This core area has a pop of 450 k with 48.8% AVAP. Adding the 48 K from Alameda gives a population that requires an additional 205 K to complete the district.

I can identify three basic choices to complete CD 15, two of which are shown in the map. The purple area is downtown and the Hispanic core of SJ including Alum Rock which wraps around some of those core blocks. This area is 56.7% HVAP and seems like it should stay together in a single district. The yellow areas sit between the CD 15 core and the CD 14 core shown in red. The third option would be to extend south from East Foothills into the heavily Asian areas, but that is a non-starter in Torie's plan.

Either the purple or yellow option would work with the CD 15 core, and both choices have AVAPs in the low to mid 20's. That means that CD 15 would be at best an Asian plurality district, and it brings up the issue of how to chop into Alameda. The Fremont chop in blue has a high Asian pop, but that doesn't seem relevant given the direction of the district as a whole. It is an erose peninsula, and the Asians in Fremont get split no matter how one cuts it. I would suggest consideration of the green chop instead. It has a lower AVAP, but make a less erose match to the rest of CD 15.

I think that resolving the shape of CD 15 first will lead more naturaly into the best division between CD 14 and 16.

[muon2]

Excellent idea on the alternate chop in Fremont assuming the green part has the same population as the blue district. The green chop of Fremont and the city of Newark are distinctly middle class. The blue parts are completely upper middle class to upper class with a median income around 120-130k easily. If we are going with a class map, the green chop of the Fremont area is better.I also think the 15th should pick up the yellow areas instead of the purple Hispanic areas. I think the Hispanic areas go better with the areas to its south. The Asian areas to its south, especially north and west of Capitol expressway around US 101 are fairly working class.

That would be my inclination as well, but I thought I'd lay out the case both ways. If Torie concurs, I'll proceed to my analysis of the division CD 14 and 16.

[muon2]

I'm with you on your pics Torie. I saw the same thing in my analysis.

I'm confused on the extension SW of 87. There's a little pocket of Hispanics, sure, but it goes through so many Anglo areas the sum is only about 45% HVAP, and a significant piece of one of those chops into residents of Burbank. In exchange you could keep the Milpitas foothills intact as well as Alum Rock, since my satellite images of roads make me think those areas really do belong with the valley rather than a long link along the foothills to the south.

I also take it that you weren't interested in my Newark alternative, even with sbane's glowing endorsement.

[muon2]

I'm with you on your pics Torie. I saw the same thing in my analysis.

I'm confused on the extension SW of 87. There's a little pocket of Hispanics, sure, but it goes through so many Anglo areas the sum is only about 45% HVAP, and a significant piece of one of those chops into residents of Burbank. In exchange you could keep the Milpitas foothills intact as well as Alum Rock, since my satellite images of roads make me think those areas really do belong with the valley rather than a long link along the foothills to the south.

I also take it that you weren't interested in my Newark alternative, even with sbane's glowing endorsement.

I will eviscerate your most creative Newark option soon, very soon, but I need to cook the steaks for my guests now.

In the meantime, putting aside the 87 thing which I don't understand, and maybe the map below "solves" it, I think we are down to the class warfare theme, and the race warfare theme. I assume the class one gets the nod, since the chop of Sunnyvale is rather vicious, even though it gets rid of the SJ trichop. The class warfare map still takes in the SJ western salient, but that is because there are no upper middle class areas to take, and that is the cut which reduces erosity.




  

Yes, that's better, particularly in the foothills, but the part of 15 in the area bordered by 87/880/280 is only 45.9% HVAP. Is it worth the type of erosity I'm usually accused of for such low gains and an extra split?

Also, are we now just working on a socio-ethnic map? I thought we still wanted to reduce splits and erosity as much as play with classes.

[muon2]

I assume you mean the first map, and that little green jut there. Collectively it is only 49.5% Hispanic?  If so, I will find my Hispanics elsewhere.

Try looking along the east side of the 101 just south of your district. (wrong version of your map) It would help if I knew the ground rules for this map.

[muon2]

I assume you mean the first map, and that little green jut there. Collectively it is only 49.5% Hispanic?  If so, I will find my Hispanics elsewhere.

Try looking along the east side of the 101 just south of your district. (wrong version of your map)

I think the answer for your current map is just along the Capitol Expy just east of Monterey Hwy.

It would help if I knew the ground rules for this map.

[muon2]

It looks to me like many of the Mission district folks in Fremont are the very same sort you are so willing to excise in the Milpitas foothills. I'm still with sbane, but I'll postpone further discussion about erosity definitions for now.

There is a place where we can perhaps have our cake and eat it, too. I looked at the minimum selected block groups in yellow to make Milpitas and Alum Rock whole and found they are only 21.7% WVAP which is lower than your district-wide average. I've shaded a pale silver area (including one block group way west that's not in Sunnyvale) that has WVAP of 37.1% which is much whiter than your district as a whole. Same pop, check it out.

edit: I see I inadvertently grabbed a block of East Foothills in yellow as well. Ignore that, and you can return a matching block elsewhere.

[muon2]

What is the population of the huge Newark-surrounding precinct, and where do they presumably live - in its southeastern part? Maybe we could have our cake and eat it too in regards to issues of de facto and, as it were, de jure erosity as well? (Though the part angling around to beyond the bridge would always look ugly.)

The block group is population 3906 and is all located at the Milpitas border. Torie's plan would need about 6 K from Fremont in addition to Newark in CD 15. Any little pocket in addition to the aforementioned block group would do.

[muon2]

I see the two white plurality precincts you wanted to grab Mike in west downtown SJ, wading through some lightly populated Hispanic precincts to get them.  It is a reminder to look at precinct size. Anyway, I decided to just go for the whole hog. In for a penny, in for a pound.  Sorry boys, I just can't do Newark. It is just so wrong. I could not sleep at night having crossed that bridge too far. You don't want that do you?

Are we done with SJ now?  By the way, that county airport precinct east of Mountain View is not an American dream precinct. It is tiny 75K condos wedged between the freeway and the airport!  (Maybe those are pied a terre's, where the pilots "do" the stews, except that I suspect it is not the SJ commercial airport. I wonder what airport that is.) But the adjacent precinct already took a bite out it, the airport creates a natural barrier, so let the deed be done.  There is only one precinct CA-14 can lose in return, and only one, there on the east edge of the now famous west SJ jut of which we have spoken, and speculated so much, and it is too large. So that precinct will need to be chopped in half. In the meantime, CA-15 has 1,400 folks too many, and CA-14 1,400 residents too few.

Thanks guys, and particularly Mike for all your hard work on this. I appreciate it very much, and yes, it made the map better.

A much better product, but I get 706,024 for the district when I draw one to match. Is there a block somewhere in yours which is actually assigned to another district? It would be 1695 in pop if that's the cause. Unfortunately that pushes it even higher and you may have to shed some off the SE corner.

I don't know if we're done, however.

I think you need to work with my map, and the CA-14, 15 and 16 merry-go-round Mike, or we will be talking past one another. I am not going to trash my entire NoCal map design. Given the outer perimeters of those three districts as drawn, how do you divvy up the spoils? That is the question with my map. But yes, that aside, we are getting closer.

I will happy to do the same working with your map design, although given your very tight metrics, there really aren't many choices out there, are there? The computer drives your map, with mere humans having relatively few choices. Am I wrong about that?

Your challenge is fair and I accept. I'll do analysis of your map to see if there are a reasonable set of options to choose among.

I had my original design with 52% AVAP.



And my recent modification with 49.3% AVAP, and that should still have over 50% ACVAP.



My initial parameters require a bigger chunk in Fremont/Newark, and I insist on a northern Asian tiger. There should be some options available, and I welcome advice.

[muon2]

So if I follow sbane' comments, I think this is the plan of mine he likes for the Asian district. I've adjusted the boundary between the central SJ district and the western SV to put the Campbell in the former and the New Almaden valley in the latter as suggested by Torie. The most Hispanic areas of Evergreen stay with central SJ, and it's now a HVAP plurality district. Did I pick up the comments accurately?

All the pictured districts are within 200 of the ideal population.



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.

[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.

Based on the Commission's data, the ACVAP is about 10% lower than the AVAP in that district. I also note that they connected Cupertino instead of Mountain View whihc is a big boost to AVAP, but that means that Los Altos only connects to Saratoga in the same district by way of the Pac coast and Santa Cruz.