Mid-2014 county population estimates out tomorrow, March 26

[Torie]

Bonus question. Based on the maps above, which CD do you think has had post the 2010 census the most robust population gain, based on the census estimates as of July 1, 2014?  How about the second most? Which CD has had the most stagnant population growth?

[Torie]

Bonus question. Based on the maps above, which CD do you think has had post the 2010 census the most robust population gain, based on the census estimates as of July 1, 2014?  How about the second most? Which CD has had the most stagnant population growth?

Most growth: VA-10
2nd most: close call between VA-07 the various versions of VA-08 here?
Most stagnant: Has to be VA-09, right?

The most interesting scenario may be if VA-12 doesn't happen.  VA-10 and VA-11 would have to majorly contract.  Ironically, any court remap that created a 2nd safe Dem majority-minority CD in the SE would make 2022 a lot easier for a GOP trifecta in an 11 district world.  8R-3D would probably be impossible then, unless Rigell and Comstock are willing to take their chances in EVEN to D+low seats (note there is a significant Reverse Blue Dog phenomenon in suburban 2X Obama legislative districts, including even some Republicans holding 60% Obama seats).  But a preexisting 7R/4D map with all of the SE Dem areas accounted for would mean a lot more wiggle room to pull 7 and 1 further into NOVA and Comstock further out.

Good guesses, but you knew, no, when I asked for the second most dynamic growth CD, that it was a trick question, right? The second highest growth rate CD, is none other than that Pub nightmare CD, the almost inner city CD, VA-11. Alexandria and Arlington both had around a 25% growth rate, while Fairfax has had only an 8% or so growth rate. The figures below are for my third good government map (I found some population errors, so I replaced the first and third maps, and the second map, which I will not bother to redraw, is off a bit):




VA is almost certainly going to get a 12th CD unless its population growth comes close to stopping dead in its tracks. It already was close to getting a 12th CD last time.

Interestingly, Gillespie I think came close to carrying VA-08 as drawn in the good government map. He ran really, really well, in Prince William County, and quite well in Fairfax. He probably carried the Fairfax County portion of the CD by maybe 3 points, while losing Prince William by about the same margin (while Romney lost Prince William by 16 points). Since Prince William is a higher percentage of the CD, he lost it, but just by a hair. Where Gillespie lost was in the Richmond suburbs, not in NOVA. In the Richmond area,  he didn't make much progress from the Romney percentages. So in an off year election, with the seat open, the Pubs just might have a chance with the right candidate (and 2022 will be an off year election). Of course, that assumes Prince William ceases to trend Dem - which is a big assumption. Addendum:  well it turns out Gillespie lost the Fairfax portion of the CD 51-49 (21,278 Gillespie, 22,098 Warner), so it's basically a 51-49 Dem CD at least for that election. Odd Gillespie did so well in Prince William. Maybe he's from there.

[Torie]

a) giving the Eastern Shore to VA-2 instead, since it only has road connections to VA Beach, and/or

You need to do it the way I did to get the black percentage up in  VA-03 and VA-4. VA-02 needs to stay south of the bridge to suck up all the heavily white precincts down there, to keep them out of the two black CD's.  I suspect this approach will unite the Pubs and blacks in the next redistricting round.

b) splitting 6 and 10 such that they don't cross the Shenandoahs.  

That sounds like a community of interest comment of some sort. I just go with compactness myself, and within erosty constraints go with county collections that minimize chops. It just so happens that collection of counties for VA-10 just about avoids a chop. Also, the counties directly south of VA-10 east of the mountains are low population, so the CD would creep considerably farther south.


Anyway, it is one state where a good government map also makes pretty good political sense for both parties. That is what I found interesting about the exercise.

[Torie]

As part of my ongoing project of just how much gerrymandering is worth come 2022, as opposed to “good government” maps, we move on to PA. Below is first the good government map (most of the CD’s are exact projections, but not PA-03, 9, 10 and 17, which are not sensitive to partisan machinations).  PA-05, a Pub CD in north central PA, bites the dust, and by virtue of that elimination, what we find, is that assuming the Pubs abandon the insane enterprise of keeping PA-07 in the Pub column via erosity and chops run wild, we find that a Pub gerrymander is worth may 1/3 of a CD. Why? 

Well, part of it, is that Philly burbs (not so much with Delaware County), along with the Lehigh Valley, trended so hard Pub in 2012.  So in the good government map, PA-06 is 1% Dem PVI, but that CD as drawn trended about 5 points to the Pubs in 2012, so based on 2012 numbers, it’s 4% Pub PVI. The other CD in play, PA-15, is about 3% Dem PVI in 2008, but it trended about 2.75 points Pub, so it’s now about 0.25% Dem PVI – in other words, dead even, and easy for Dent to hold as long as he wants to remain in office.  True, he might not be around in 2022.

The second map is the Pub gerrymander – which is all about PA-15, and moves it about 2 points in the Pub direction. So in the gerrymander “lite” map, it’s tilt Pub rather than dead even – say worth a third seat seat max – one half a seat if the Pubs want to go ugly, and have the trifecta.

With respect to PA, this is not a wholly academic exercise. Governor Wolf is more likely to get reelected in 2018, and thus either there is a bipartisan deal, or it goes to the courts. Go Wolf. There is no reason other than insane partisan greed, to gerrymander PA come 2022, with some hideous map. The point of this exercise, is if, over the Fruited Plain as a whole, gerrymandering and ugly maps does net much, why go there?  Why?

[Torie]

Have the new population stats been loaded into DRA?

No, one needs to take the 7-1-14 census population estimates, extrapolate them thought to 4-1-2010 based on the annualized percentage change, and use the new county population numbers, and prorate for chopped counties. Using a spreadsheet, one can move county numbers from CD to CD, until the map "works" population wise, or gets close, without undue erosty, and then do the chops. At least that is the way I do it.

[Torie]

My map tried to max the BVAP's, and they were at 41.5% and 45.5%. It's not legally mandated, but it is probably what will be done politically. I was unable to avoid a chop for VA-09, but maybe my figures are still off some. There are a lot of counties to count down there.

[Torie]

My map tried to max the BVAP's, and they were at 41.5% and 45.5%. It's not legally mandated, but it is probably what will be done politically. I was unable to avoid a chop for VA-09, but maybe my figures are still off some. There are a lot of counties to count down there.

I guess I'd be a little concerned that the chops into both VB and Portsmouth look like race-based swaps at the expense of split counties/ICs.

Perhaps you haven't given as much growth to Roanoke and Blacksburg as I project in CD 9. I have Roanoke county (+Roanoke and Salem) at 224.9K and Montgomery (+Radford) at 120.7K. Together with Pulaski they are over half the CD. In any case I expect that these estimates will shift over the next half decade.

My map is a race based gerrymander. It's legal, because there is no packing. Your map will at once enrage both the Pub and the black community I strongly suspect.

My figures are 120.4 and 223.6.  I just take the current population growth percentage, take it to the power of 1/4.333 to get the annualized percentage, and then take that annualized percentage to the 10th power. It may be that whatever you did gets one to systematically higher numbers. For your Prince William plus Stafford plus Fredericksburg CD, I ended up however with 8K too many people. My population per CD is 731,024.

I don't like the erosity of your grey CD, but then we do have different philosophies about that. I am willing to chop when the erosity gets to be too much, absent a darn good reason to do otherwise. But then you already know all of this all too well about me!

Anyway, these maps are more about how legislatures interested in good government maps due to political power splits, or courts, will draw the lines, as opposed to following your unique rules (and up to a point mine). (In regard, the Jimtex definition of metro areas is totally unique, and I don't think most folks would view Stafford and Spotsylvania as part of the DC metro area.) Among other things, I suspect courts are more interested in "art" than you are.

[Torie]

Oh, that jut out as you call it is small beer. I did have trouble with the lines between VA-09 and VA-05 I admit. Anyway, do you like the version below better? It causes VA-06 to lose its more box like shape, but whatever. It does stop VA-05 from being inserted between VA-09 and VA-06, which is a point loser under your system I understand.

[Torie]

My eye tells me that this new version of 6 wouldn't be much worse than the old one for many conventional versions of compactness, and is probably about the same on the muon erosity scale. Since the new version reduces the erosity of 5 (and improves its compactness) it would be preferred by any system IMO. My guess is that it might get better still by rotating populations in Nelson, Louisa and Nottaway.

You are probably right. One map change leads to another. The only problem with your suggestion, is that Fluvanna gets boxed in on three sides by other CD's, which is a feature that I dislike quite a bit. Granted, Nottoway has the same problem, but its shape makes it look a bit more aesthetic that doing a box in of Fluvanna to my artistic eye.

[Torie]

After a lot more work, here are my two options for Virginia, with the correct population projections – at last.  I was reminded – painfully so – that for the ideal CD population, you cannot just take the statewide projected population, and divide by the number of CD’s. Rather you need to sum the counties’ projected populations, and divide by the number of CD’s, because as counties grow at different rates, the overall statewide projection will change a bit, and tend to be a bit higher, as faster growing counties become a higher percentage of the population as of July 1, 2014, than they were as of April 1, 2010.  Or something like that.

With the 2k or so higher population number per CD, it turns out that the population of VA-10 is almost perfect with no chops. It’s about 850 people too low in population.

And the first map also loses a second chop, where without a second chop, VA-02, 03, 04 and 07 collectively have almost a perfect population, collectively off by only 13 people (12.76 to be exact based on the population projections).  The lines of the second map incurs the cost of not taking advantage of this most fortuitous accident of a near perfect population count for the above referenced four CD's. However, the additional chop is a most aesthetic one of Louisa County, and while it generates a bit more erosity by moving Nottoway into VA-04, it has the virtue of moving the black percentage of VA-04 up by 60 basis points from 41.5% to 41.1% BVAP, which would please the black political establishment in VA.

VA-03 is 41.5% BVAP, which is probably enough to elect a black in the Dem primary, but it is cutting it a tad close since VA has open primaries, unlike Florida, where they are closed, with only party registrants eligible to vote in Dem primaries.  VA-03’s lines however look just so beautiful, and it’s compact, so I elected to go with the lower black population percentage. My thinking was the same when I gave up the chop of Chesterfield, which lowers the black population percentage of VA-04 by about 3 or 4 points, depending on which of the two maps one chooses.

Everything is a balancing test, including in the case of these maps, giving the finger to Muon2’s cover and pack rules for the DC metro area.   In that regard, I might note however, that the DC MSA includes Clarke and Warren Counties, even if Jimtex’s metric does not, so in his map if one chooses to hew to the official definition, VA-01 will need to chop into Stafford County, losing his microchop that is hidden along the Stafford-Spotslyvania County line.

Oh, in looking at the MSA's, I see Muon2's has a cover penalty for the Richmond MSA (using the standard definition), which it so happens I do not in my first map. So my first map loses two chops (assuming you don't count micro-micro chops), and a cover penalty, for three extra points, while incurring a 2 point penalty for my cover and pack penalty for the DC metro area. Thus, assuming Muon2 has VA-10 take in Warren and Clarke Counties, I still win by one point on the chops. He needs Jimtex's rules to win on the chops (presumably the Jimtex metric loses the counties that get Mon2's map into cover trouble). Screw Jimtex!

 [/URL]

 

[Torie]

Counties in the MSAs are divided by the Census into central and outlying counties. The MSAs had too many rural outlying counties that made it in solely due to relative numbers of commuters in and out, not based on the density of populated areas. Some outlying counties had population centers that were deservedly in the metro, so using central counties only was unsatisfactory. We spent weeks haggling over the UCC definitions to balance those two points and come to a consensus agreement. I'm not going to abandon them to aid one map over another

Torie, what do you have for your projected VA CD now?

732,463.1096 per CD.

I am well aware of Jimtex's rationale. It even makes some sense in theory. But this exercise was not about maxing the Muon2 score, but rather what seemed like the best map that would fly politically, and keep both sides reasonably happy, while still being a good map. (This is more than a theoretical exercise if Gov. McAuliffe gets reelected, as seems more likely than not.) Your map runs the risk of jettisoning two black congressmen, including one existing one, and also violates the bridge or full time ferry rule (seasonal ferries do not count, and I checked that issue myself, before just abandoning the rule (not the law in VA obviously), due to other considerations, just as you did). Anyway, even under your rules, putting aside our mutual cheat on the ferry issue, I think I win on the erosity front, so it is not as if I am knocked out of the box from a pareto optimality standpoint. The bottom line is that your map will never, ever, be enacted into law, as I am sure that you probably agree.  Mine might.

[Torie]

I will deal with the number crunching later in due course. What on God's Green Earth causes you to believe a commission under tight Muon2 metrics will be drawing the map in VA after the next census?

Absent that most miraculous development, the thing is, the Pubs keep their seats but one with my map, assuming Forbes survives the redraw until then, which is a big assumption, and the blacks get another seat, while the new seat is tilt Dem, but doing that makes the 4th Nova CD safe Pub, while at the moment, absent the talented incumbent Pub, it's under risk. That is the kind of map the politicians like and understand, presents not much risk, keeps the incumbents happy, and for the new seat, it is close to a fair fight. That is what in my world is called a saleable compromise map. If the Court follows the Muon2 map, you suddenly have a host of marginal CD's, inconveniencing and terrifying a host of politicians, and I doubt either party will be pleased with that. They would rather divide the spoils in a more predictable manner.

[Torie]

I don't like the erosity of your grey CD, but then we do have different philosophies about that. I am willing to chop when the erosity gets to be too much, absent a darn good reason to do otherwise. But then you already know all of this all too well about me!

Anyway, these maps are more about how legislatures interested in good government maps due to political power splits, or courts, will draw the lines, as opposed to following your unique rules (and up to a point mine). (In regard, the Jimtex definition of metro areas is totally unique, and I don't think most folks would view Stafford and Spotsylvania as part of the DC metro area.) Among other things, I suspect courts are more interested in "art" than you are.

Your art has always differed from mine. For example I find that your NW peninsula out of CD 5 and jut and chops around Richmond to add far more erosity than the clean L-shape of my CD 7. I do hope you are not abandoning the idea of measurable metrics to judge maps.

edit: The Census agrees with jimrtex that Fredricksburg and the two adjacent counties are part of the DC metro.

edit2: I think I found the math discrepancy. First the April to July shift is 0.25 of a year. Then the formula to get the annual rate is (estimate/census)^(1/4.25) - 1. The projection is census*(rate + 1)^10. I also had to correct for the fact that Bedford city was separate from the county in 2010, but they are together for the county estimate.

Right you are. The census excel list does not have Bedford City listed. So with that correction, my numbers for the VA total and per CD are 8,806,268.366 and 733,855.6972, respectively. So we are still off from each other.

Anyway, now my perfect population quadrant of CD's, is now 1,504 short, and VA-10 has 1,127 too many. Ah the pain of differential growth rates when doing the exponential power thing. My spreadsheet does not balance to the master one for some reason (it's off by about 75 people), but whatever. I have to do special pastes for the lines with county cuts, and apparently I missed something somewhere.

[Torie]

I don't like the erosity of your grey CD, but then we do have different philosophies about that. I am willing to chop when the erosity gets to be too much, absent a darn good reason to do otherwise. But then you already know all of this all too well about me!

Anyway, these maps are more about how legislatures interested in good government maps due to political power splits, or courts, will draw the lines, as opposed to following your unique rules (and up to a point mine). (In regard, the Jimtex definition of metro areas is totally unique, and I don't think most folks would view Stafford and Spotsylvania as part of the DC metro area.) Among other things, I suspect courts are more interested in "art" than you are.

Your art has always differed from mine. For example I find that your NW peninsula out of CD 5 and jut and chops around Richmond to add far more erosity than the clean L-shape of my CD 7. I do hope you are not abandoning the idea of measurable metrics to judge maps.

edit: The Census agrees with jimrtex that Fredricksburg and the two adjacent counties are part of the DC metro.

edit2: I think I found the math discrepancy. First the April to July shift is 0.25 of a year. Then the formula to get the annual rate is (estimate/census)^(1/4.25) - 1. The projection is census*(rate + 1)^10. I also had to correct for the fact that Bedford city was separate from the county in 2010, but they are together for the county estimate.

Right you are. The census excel list does not have Bedford City listed. So with that correction, my numbers for the VA total and per CD are 8,806,268.366 and 733,855.6972, respectively. So we are still off from each other.

If you have the numbers that go into the three counties I listed, I can perhaps reverse engineer our discrepancy.

Btw, I notice your Hampton Roads metro area incurs a pack penalty by virtue of VA-02 taking in Northampton. So that offsets my pack penalty for the Richmond metro area, assuming your map avoids that, which I assume it does given that erose grey CD of yours.

By the way, given your strict hewing to the Muon2 metrics, what is your thinking regarding jettisoning the year round ferry or bridge rule?  Isn't that a cheat? It seems that is about the same thing as doing a traveling chop to make a map work. If your exercise is to slavishly follow your rules to see what kind of maps it generates, then as you say, one needs to be consistent. Anyway, it only seems fair that if you do that, than I get to jettison the Jimtex metro area definitions, which while having a justification, are esoteric, when I find it convenient to do so. Or maybe doing it either way is OK for cover and pack penalties.

I am not trying to twist the knife here, but given all the hard work you are doing, it strikes me as an issue that you might think about, if only as a discipline, so that your portfolio of maps all play by the same rules. Also you seem to let the chop metric dominate over the erosity metric, when generating your portfolio. In that regard, putting aside your grey CD, a lot of erosity points are involved with respect to the macro chop of Fairfax by either VA-10 or VA-08. It may be that picking one CD or the other to make the macro-chop might save a sufficient number of erosity points, that it makes it a good bargain  to incur the concomitant cover and pack penalty points (if VA-08 is the superior choice from an erosity standpoint).  Maybe you might think about drawing two maps for each state, one erosity sensitive, and one chop sensitive (within reason of course). That might help folks get an idea of how your rules play out when balancing off and mix and matching the two competing metrics. Just some thoughts by me.

[Torie]

Would you post a link to where the correct numbers are please?  I am not sure I understand Jimtex's comment. What does proximity versus contiguity mean? How is proximity defined if you can spread outside the UCC within a proximate zone with penalty, if that is the drift here?

I gave some thought to the issue of minority influence or minority CD's that are not required by Section 2, and came up with the following overall scheme. Tell me what the holes or downsides are.



1. Upon there being sufficient data from the decennial census, a computer (I assume a computer can do all of this) generates the Eligible Pareto Optimal Maps (EPO Maps) within a reasonable period, say 90 days.

2. EPO Maps is defined as all maps otherwise legal maps under Federal Law that have up to four chops over the minimum number of possible chops while being pareto optimal. (My thinking is that there are five such maps, one with the least erosity for each chop level, although it could possibly be less if adding a chop actually increases erosity.)  If however there is more than one such map with a given level of chops that has the same erosity score, then all such maps are EPO Maps.)  If the total number of EPO Maps is an even number, then another map shall be added that either has one additional  chop over the balance of the EPO Maps with the least erosity,  or another map with the same number of chops as one of the existing EPO Maps, if adding another chop actually increases erosity more than the second best erosity scoring map that has one fewer chops.

3. At any time prior to the Deadline Date defined below, the Legislature, provided a majority of the members thereof of both of the two largest parties in such Legislature so vote, can adopt an EPO Map as the the map that shall become law (Final Map).

4. If by July 1 of the year following when the census is taken, the Legislature has failed to adopt a Final Map, then by lot one of EPO Maps shall be discarded. If the Legislature within 10 days thereafter  fails to adopt a Final Map, a second EPO Map shall be discarded by lot, and if within an additional 10 days period thereafter, the Legislature still fails to adopt a Final Map, a third EPO Map shall be discarded, and so forth, until just one EPO Map remains. The last EPO Map so remaining  shall become the Final Map, unless the Legislature adopts a Final Map by the end of the year following the year that the census was taken (the Deadline Date).

5. At any time prior to the Deadline Date, the Legislature may adopt a Final Map that would otherwise be an EPO Map, except that in order to increase the minority VAP percentage in one or more CD's to a figure that is within 30% VAP to 50% VAP, the otherwise EPO Map is modified to effect such a VAP percentage, but otherwise constitutes an EPO Map (i.e., such a map has either the minimum number of chops necessary, or if there are more chops, has a lower erosity score than a map with fewer chops). Such a map shall be deemed an EPO Map.


That's it. By the Deadline Date, the Legislature with a majority of both parties agreeing, adopts an EPO Map or a MEPO Map (see below), and if they fail to do so, you end up with last map left standing by lot that is an EPO MAP as the Final Map.

The idea is a machine generates the maps, and the Legislature bargains for one of the EPO Maps, but running the risk that if they fail to select a map, a map by chance will be selected for them, and they don't know which one, although as each 10 day elimination period goes by, they get a better and better idea of what the fall back option will be if they fail to cut a deal on one of the EPO Maps, including  a map adjusted to increase minority percentages as described above (MEPO Map). If a MEPO Map pleases both the Pubs and the relevant minority group or groups, and a majority of the Dems agree to accommodate the minority, a MEPO Map will be passed, but not otherwise. It might pass but only because both parties like that map better than what will be the default Final Map if they fail to agree, so the lot elimination process may need to play out to some extent or totally, before it's possible to reach an agreement.

The process encourages deal making within tight parameters, it's all objective, and there is a end game if all else fails, where a Final Map will become law, that scores quite well, if not necessarily the highest scoring map, in the sense that it may have a lot more chops for not that much less erosity, or a lot more erosity in exchange for not that many fewer chops, but it still is good enough for government work.  And it gives space for a minority friendly map, if the Pubs like it (they usually do), and the Dems are willing to go along, because their minority base demands it.

I would also urge that population deviations with 0.5% shall be deemed legal, so the minimum theoretical number of chops will be reduced, but that is a codicil. It seems obvious to me to reduce the size of chops short of macro chops that cause the erosity score to collapse, but maybe not to others.

What do you think of this design?

[Torie]

The link to the correct numbers was in my earlier post.

My idea since the OH competition has been similar to what you have. I like the bargaining idea which I hadn't considered. I have to think about the Survivor aspect of the proposal, which is certainly novel. However, I don't think a computer as map maker works.

Redistricting is an example of a NP-complete problem known to computer science. It basically says you can't guarantee that any algorithm will converge to an optimal solution or set of optimal solutions in polynomial time - that means the time grows exponentially as the state becomes more complex and counties require chops. It also means you can't readily defend in court that the computer will beat a neutral mapper.

I favor making the computer the judge of what is EPO, a task well suited to NP-complete problems. Anyone, including someone using a computer algorithm, can submit a plan to be judged. A human commission would support the judging algorithm to guard against hacks and to make sure all legal niceties are followed including the VRA.

After the black box generates the maps, there could be a ninety day period for anyone to submit a map that beats the black box. Would that work?

[Torie]

I'm still concerned as to how a computer draws a VRA-compliant map for IL, NY, or CA. In IL at the legislative level it would be extremely difficult to code that. Notwithstanding any of that, why not open up the process to the public in parallel with any computer algorithm that runs. It seems to me that would work in favor of the negotiation aspect in your proposal.

Sure. If the black box generates a bunch of maps that are inferior to existing submissions, its work can be tossed. In fact, if submissions can be made in a form that can just be dumped into the black box, submissions could go directly into the black box.

And there are the knotty issues of if a 50% VAP CD can be drawn of a contiguous minority population, do you have to meet that (if a lower percentage will elect a candidate of the minority's choice, and if a lower percentage just what is it (which alas will vary depending on the voting habits of those who are not of the minority to be accommodated within any proposed Section 2 CD), and what is the process adopted to set up that percentage that is subject to variance depending on the precise location of the CD), and what to do about the Central Valley of CA, where it is not clear to me that any Section 2 Hispanic CD's are required, or if required whether you need to go over 50% CHVAP.  In Illinois, the  same issue exists as to whether one needs to go over 50% CHVAP, but at least last time, it was clear one needed two black CD's and one Hispanic one, and if one decides on the VAP percentages, I would think one could code that into a computer that constraint easily enough given where the minorities live, which is quite compact. It would be nice to get Federal Courts to decide these matters in advance in declaratory relief actions, but I digress. A slew of such cases should hash out the gaps in the law that remain in a hurry, or at least a lot of them. We also don't have consistency about CVAP versus VAP between Circuits.

The whole system will collapse if there are not clear constraint parameters on this set up in advance.
I would suggest that whomever is hired to code the black box not be the firm that coded the Obamacare insurance software.

[Torie]

As part of my new found religion of jidading against VRA induced chops, here is my revised map for NYC, which probably looks remarkably similarly to Muon2's (the 26 CD map for the 2020 census). I think I may be more militant than he now against such chops - ships passing in the night. NY-05 is on the cusp of whether or not it is required by the VRA, assuming the test is 50% BCVAP, with only citizens counted. I think that needs to be hit before the CD is required to be drawn, even though when drawn it can go below that trigger number if it will still elect the candidate of the minority's choice. My CD is 46.4% BVAP, which should be enough to elect a black if that percentage holds in 2020). In political reality of course, the CD will be drawn anyway, whether legally required or not.

NY law now has a bipartisan gerrymandering regime. If a deal cannot be cut, the courts draw the lines, it appears to me. There is no "default map" mechanism, but then how could there be, if the map requires bipartisan approval (basically a two thirds vote of both the Senate and Assembly). So it will be interesting to see how this plays out. Will the Pubs insist on a second Pub seat in Brooklyn?  If they don't get it, will the court give it to them? The lost CD seat, assuming the state loses a seat, will be a Pub seat upstate, which will tend to make a bit safer the remaining Pub seats. From the Dem perspective, will they end up having the lines drawn by the courts to end the Pub control of the State Senate is the question that comes to my mind. The Dems would give up some Dem Assembly seats (but would still have about 60% or so of the seats), in exchange for taking control of the State Senate. A bipartisan deal will certainly not draw the Long Island CD's as drawn, where Peter King has the most Dem seat (3.5% Dem PVI, maybe more like 4% now if not more), and both NY-03 and NY-04 have no Dem PVI at all. Both seats are just about dead even circa 2008.

[Torie]

Well the map below (the screen shot is skewed because the map drawing utility froze on me) I think playing by Muon2’s rules (and putting aside the exact line of NY-04 in Queens to hew to neighborhood lines which might have some impact on the exact line but really not the partisan balance) I suspect is hard to beat. Broome and Herkimer are micro-chopped, along with Harrison in Westchester, and Catskill in Greene, and I think I played the road cut erosity measure game correctly.  That is what drove the location of my chops in Catskill in Greene County and Cortlandt Town next to Peekskill in Westchester.



Given the court default map if the Court gets close to the Muon2 line drawing regime, it really does seem like the parties will do a bi-partisan gerrymander if NY loses a seat in the next census, unless the Dems really are willing to inconvenience 3 Democratic Congressional incumbents, in exchange for taking over the State Senate (assuming, which is a big assumption, that they have not already taken it over after the 2020 elections).  Per the chart below, by doing a bipartisan gerrymander, the Dems can make the 3 Dem incumbents safe rather than sitting in toss up seats.  So the King seat becomes a Pub pack, and we have a Rochester to Syracuse Dem pack seat upstate, just like Louise Slaughter always wanted (but Syracuse in lieu of Ithaca). That seat can be named in her honor.  ☺  If that looks too grotesque for even NY gerrymanders, something more marginal can be done, to get NY-24 closer to tossup status at least. Alternatively, just leave the good government map as is, except have NY-05 chop deeply into Nassau to pick most of  the areas with some black population, thereby forcing NY-04 more deeply into Queens, thereby making it safe Dem, while NY-03 moves to lean Pub. NY-17 and NY-18's lines could be played with, to make one seat lean Pub, and the other lean Dem, as another option (maybe by NY-18 taking Orange County, and losing much of its Westchester salient). The thing is, is that while good government types love a host of marginal seats, the politicians of both parties tend to hate them. And in NY, that means only one thing: good government loses, almost always.
 

In this chart, I score the partisan effect by assuming each party loses half a seat out of the box with the elimination of the NY-19 Gibson toss up seat.  So the net Pub gain from it all is less than the net Dem loss.

[Torie]

MN was a state on the bubble to lose a seat in 2010, but they dodged a bullet and kept 8. The forecast for 2020 is that MN will lose a seat, though they are still on the bubble. I'll assume they drop to 7 as I use the 2014 data to project county totals in 2020.

The Twin Cities UCC will have about 4 1/4 seats in a seven seat map. This plan preserves the cover count of that UCC and keeps deviations within 5%. The Minneapolis area (orange) has 2 districts.

How close to the cusp vis a vis the population projections is the map below in play as between the blue and teal CD's, which has less erosity I wonder. In both maps, MN-01 moves discernibly more Pub, but it moves more in that direction in the map below.



Is this about the right split of Hennepin based on the population projections? I notice that Minneapolis is one of those inner cities that is now growing faster than not only the state as a whole, but also the suburban areas, just like NYC, which is interesting. Philly is also growing faster than its host state now. Folks are moving back to the city in some places it seems.

[Torie]

Thanks. It looks like the map has a cover and pack problem for the Twin Cities metro area however.

Maybe this iteration pencils out.

[Torie]

You also swapped out Morrison in addition to Mile Lacs.   The switches moves MN-07 40 basis points in the Dem direction. That's how the game worked out.

I for some reason errantly thought there was a cover penalty, which there is not, so your map wins the pareto optimality test by incurring just the pack penalty.

The single most controversial aspect of your rules seems to me that one incurs a horrific erosity penalty for macrochops, which perhaps in the mind of some, gives too much weight to avoiding macrochops. They are worth it (where otherwise not unavoidable), if, and only if, doing the macrochop is the only way to minimize the chop count, in which event, at best, the map would only be competitive, rather than the winning map, because there certainly will be another map, with a extra chop, but less erosity.

There is obviously no right or wrong answer here, but it is a policy choice. If say, the rule instead were that a macrochop counts for 2 chops rather than one, but no special erosity penalty other than what normally obtains, then a macrochop would leave a map competitive if it avoided two chops elsewhere, counting as chops pack and cover penalties, and winning if it happens the macrochop does not activate an additional road, and turns out to the be the map that minimizes erosity, or tying if it avoids one chop elsewhere. Whether the normal erosity penalties for macrochops is sufficient to avoid undue gerrymandering, is another issue to consider. How ugly can a macrochop get while avoiding an extra town or city chop, or activating an additional road, is the issue. Do you have any thoughts on that Muon2?

Not that you didn't already know most of the names of counties, towns and cities in MN like the back of your hand, but I suspect that by virtue of our obsession, we are both in about the 99.99% percentile when it comes to knowing, and being able to identify, counties and many of the subdivisions across the Fruited Plain. If some kind of trivial pursuits game were constructed involving having to name them, we would be near unbeatable.

[Torie]

"Its weakness is when a rural district goes into an already chopped urban county such as your MN CD-2."

My MN-02 was not a rural CD. I take your point about the hostility of rural CD's intruding into urban areas, that does not obtain with respect to intra metro area CD's. Whether there is any merit to drawing a distinction is yet another matter. Indeed, I macro chopped precisely because I wanted more pure urban focused CD's.

[Torie]

"Its weakness is when a rural district goes into an already chopped urban county such as your MN CD-2."

My MN-02 was not a rural CD. I take your point about the hostility of rural CD's intruding into urban areas, that does not obtain with respect to intra metro area CD's. Whether there is any merit to drawing a distinction is yet another matter. Indeed, I macro chopped precisely because I wanted more pure urban focused CD's.

I understand your point, I blanked on the CD-6 into Wright as opposed to the CD-2 into Hennepin. I know the area well and except for the exurb Rogers, most of your CD-2 in Hennepin is rural, so that helped me lose sight of your objective.

Erosity and compactness measures generally don't know the population make up of the geographic units in the district. The exceptions are compactness measures that use the distances to the center of population, but they aren't particularly good at measuring erosity. We spent a lot of time with Grand Rapids and Detroit area districts to find a scoring balance in macrochops that reasonably approximated what the eye suggested. I'm not sure it makes sense to go back to the drawing board.

Did you look at swinging all those chops east so that CD-2 macrochopped Washington and then CD-4 went into Anoka?

No, I didn't. I would still have the macrochop issue that we are discussing however I presume. Anyway, it's an alternative scoring system to keep in mind. It still penalizes macros, but just in a different way, that does not swamp, and dominate, the erosity score. In smaller counties, there will not be macrochops, and with the extra chop point, an incentive to avoid them anyway. The only reason to do a macrochop into an urban area, would probably be to avoid a pack or cover penalty, or both, anyway, which facilitates, rather than degrades, the rural-urban divide.

[Torie]

One does not credit for keeping the Red River basin together, but it does appear by doing so, erosity goes down in the map above.

If based on your experience, nuking the erosity score for macrochops keeps them much cleaner, when necessary, that would be a major factor militating in favor of your approach. Being practical is job one.

I don't have a macrochop in Hennepin anyway, if the 33,000 chop number is based on projected population. The quota is about 40,000. Hennepin is projected to grow about 10% over the decade. So my map, or revised map, adjusting for the 10% population growth, has no macrochop there. I want a rescoring! My map should do what you did in Washington County to avoid a muni chop there too I suppose.