How To Reform Redistricting And End Political Gerrymandering

Voting Thread Open!

Hi all! This thread is going to cover districts, redistricting, what they are, what their purpose is, and how their implementation affects our democracy. First, a little background...For those of you who don't know, congressional districts are used to divide people up into groups for the purpose of having each group elect one from among themselves to act as a representative within government for the group as a whole. There are also districts and reps at lower levels of government which basically all serve a similar purpose...just...at a lower level. This discussion though will focus mainly on federal representatives in the house congress.

I'm sure most of you are already aware...that our federal congress is made up of two branches; the senate, comprised of two representatives from each state, and the previously mentioned house branch, the composition of which is determined based on a state's population. Essentially, for house representation, the U.S. constitution, in order to ensure some amount of fairness in representation, mandates that each state should be afforded a number of representatives in the house congress which is proportional to that state's population.

So...all fine and good so far right? Though honestly... all of that should have really been something that we all learned in elementary school (and if your school didn't teach at least that much, well then I'm sorry to say it, but you probably didn't have a very good elementary school )

...Anyways,...so we've established that each state is going to get a certain number of representatives based on the state's population. The question then becomes, how is each state going to go about picking who those representatives get to be?....As long as we insist on using the standard Plurality methods of voting in this regard our options for how to resolve the question are limited, and oh, do I ever have a whole separate topic set up in relation to that issue. But back to districts...basically, the idea our founders came up with for determining how states pick their congressional representatives, is to divide the geographic area of each state up into districts, with each state having one district per constitutionally allocated house representative and likewise each district being responsible for electing one member of the house congress.

Current Congressional District Map:

​

The original reasoning behind this is that people should get representation based on where they live, i.e. the idea that the best person to represent someone in congress will be a person who lives in the same general local area. And since different areas of a state may have vastly different needs from other areas of the same state, we want to make sure that each part of each state which is of a sufficient population gets at least some representation in congress, so thusly we have congressional districts so that individual representatives are chosen by local majorities.

Now, this all sounds good in theory, and does in fact make a lot of sense on paper too. However, in practice, we start to run into issues when it actually comes time to draw up and define the boarders for those congressional districts. And this, redistricting, happens every 10 years following the census. This redrawing process is necessary simply due to the natural shifts in relative population distribution that occur from year to year. And it is precisely at this juncture which is where we run squarely into the main problem of which this thread aims to address...a problem which has come to be known as gerrymandering.

BTW, this thread is part of a multipart series of solutions-based topics and ranked votes I've been setting up to delve deeper into a variety of topics and issues our country faces which I believe we can be distracted from finding solutions for due to the daily sensationalized noise in the news as of late. If you're interesting in reading more threads like this one, or just want to keep track of when the votes are taking place, feel free to check out the related threads below:

Solutions Oriented Approach to Restoring Meaningful Civil Discourse
What To Do To Reduce Partisan Dysfunction In Politics
Ranked Vote: Discussion Thread (includes schedule)​

OK, great, so with that out of the way, back to gerrymandering...So, as previously mentioned, those congressional district maps shown earlier have to be redrawn every few years during a redistricting process. Now, one major factor contributing to the problem is that the responsibility for drawing these new maps currently lies with partisan state legislators who are almost completely free to draw these new maps however they like. So, naturally, given such an important responsibility, these legislatures are of course incentivized to draw maps which favor their respective parties. This activity is what we call gerrymandering. And to the detriment of our democracy, state legislator across the country not only frequently act upon that incentive, but also actively look for ways to heighten the political impact of such politically drafted maps.

"gerrymandering is the act of defining a congressional district such that it maximizes a particular party's political power."​

There are basically two main techniques that politicians use when creating a gerrymandered map; cracking and packing. I.e. drawing the district maps such that ones opposition is split up (or cracked) in some areas of the map to lesson their political impact in close districts and subsequently grouping those opposition voters instead into packed districts where those opposition voters have so many votes that they are essentially guaranteed to win the packed district but end up with less representation overall as a result of their votes being segregated into a smaller number of districts.

https://en.wikipedia.org/wiki/Gerrymandering

​

The above image perfectly illustrates how the results of an election can vary drastically simply based on how the district lines are drawn. And I don't know about you guys, but the way I see things, if we have ourselves a congressional election in which 60% of voters overall vote for candidates from the green party and 40% of voters vote for candidates from the yellow party, and we somehow end up with either a congress in which 100% of the representatives are from the green party (i.e. they get significantly more representation than they should have) or a case in which the minority yellow party somehow ends up with majority representation in congress...then something has gone seriously wrong.

It's not just about which party gets the most power either, or what proportion of power, as congressional district gerrymandering is also often used not only to give advantages to one party over another, but also tends to be used to draw maps in such a way that favors incumbents in general regardless of party,...which...when you think about it, actually goes a long way towards explaining why so many politicians despite however despised they are yet remain in office nonetheless election after election after election. [/run-on]

"It makes it impossible to throw out congressmen no matter how much everyone hates them. And we hate them. It's the only thing we as a country agree on anymore, we really hate Congress, but gerrymandering means we probably love our own congressmen.

Gerrymandering takes a 13% approval of Congress and turns it into a 90% reelection rate for incumbents. Let me repeat. Despite only 13% of us being happy with Congress, congressmen won 90% of their races in 2012. In fact, we haven't had a reelection rate in the House lower than 80% since at the very least 1964 and likely since longer."​

​

Again, this state of affairs is a serious problem. This is not the way that democracy is supposed to work. It should be the case that voters get to choose their representatives. But it should never be the case that representatives get to be the ones to choose their own voters. Such a thing defies logic and flies in the face of true democracy...and unfortunately, such is exactly what we have today.

What does this mean? It means that,

1. We end up with a congress full of representatives that we don't like and can't get rid of,
   
2. We end up with congresses which are not truly representative of the voters,
3. We end up with less competitive races in many areas, which in turn leads to,
4. More political polarization, gridlock, and dysfunction, within the congress itself.

In the past, gerrymandering has occurred mostly along racial dividing lines in order to give certain racial groups more or less governing power, but today's technology has now advanced to such a level that maps can now be gerrymandered in a much wider variety of more target and specific ways in addition to race, making the skewed maps all the more effective at significantly altering the outcomes of elections. At times, including a few recent examples, things have gotten so bad that the Supreme Court has stepped in to overturn some gerrymandered maps. But it seems clear to me that this piecemeal case by case approach to fixing the most egregious gerrymanders out there have been largely ineffective at solving the issue for our nation as a whole. And some actions that the courts have taken have actually acted to exacerbate the issue. Certainly it seems then, that a much more comprehensive solution is needed...

I'm sure a lot of you know where the term gerrymander comes from, and have already seen the famous Salamander/Vulture image. Since this post is getting long already, I wont repost it here. And besides, as far as gerrymanders go...why post that sorry excuse for a gerrymander when we've come such a long ways since then? My friends...Behold! North Carolina's 12th and 4th districts. Yep, we've come a long way indeed...

​

In fact, North Carolina as a whole has some pretty messed up districts in general...

​

Not that I mean to pick on North Carolina or anything. OK maybe a little...but this particular issue is not limited to just one or a small handful of states. Nor is it limited to just one party. It is a bipartisan issue, with bipartisan contribution to the issue, which should have a fully non-partisan solution.

Ideally, the results of an election should result in a congress which is proportional to voters...so, going back to that green and yellow graphic above, one would expect that an election given those votes, if fair, should result in a congress that is 60% green party and 40% yellow...or at least something pretty close to that split. Some might say then, why not just have proportional representation in congress, and while certainly a good thing to have as an ideal guide, to simply say that alone and to say nothing else is over-simplistic as a an actual solution, since we still need some process to actually pick the individuals to fill those congressional seats.

So assuming we wanted to stick with the idea of having congressional districts (not that we necessarily have to)
There are several general rules for how a fair district map should be drawn:

• Districts should each contain about the same number of people
• Districts should be geographically compact (i.e. uncomplicated in shape)
• Districts should not be drawn to intentionally favor one party over another
• Districts should not be drawn to intentionally favor one class of person over another

But as for how to actually go about achieving a map that meets those standards...that part isn't quite as straightforward...

...or is it? I'll let you guys decide. In a latter post, I'll post my own ideas about how to achieve a greater level of fairness within the redistricting process, but very eager to hear what others have to say on the matter. Please post and discuss your ideas for how we might best tackle fixing the issue of gerrymandering, and once we get a good number of ideas we can vote on which ones we like best.

​

Remember, the focus of this thread is on finding/coming up with solutions. And we should of course be critical of the different ideas that get proposed, but if you're going to criticize something here, I just ask that you be ready to offer up a better solution in the other ones place. Thanks! Oh and and also...as always, any discussion of conservative republicans this, or liberal socialist democrats that, or communism fascism this that and the other... will be considered off-topic and out of bounds for this discussion. Basically, it doesn't matter who does it more, or who started it. This thread isn't about...and the focus when looking at this issue shouldn't be...in placing blame or pointing fingers, but...as mentioned before, the focus here should be in finding/coming up with solutions. So with that, I eagerly look forward to hearing yours.

-Meta

 

Ideas so far:
-Use an Automated Algorithm to Draw Maps Based on Key Principles of Fairness
-Algorithm: Square-Rectangle Method
-Algorithm: Shortest Split Line Method
-Algorithm: Ring Method
-Algorithm: Ring Method (Quadrant)
-Algorithm: Ring Method (Multi Segmented)
-Limit District Lines to County Borders
-Require Districts to Meet a Complexity Ratio Standard
-Require Districts to Meet a Proportionality Ratio Standard
-Use Bipartisan Councils
-Use Independent Commissions
-Party-list Proportional Representation
-Single Transferable Vote
-Mixed-member Proportional Representation

 

OK, so for my proposed solution...there's actually a few different ones that I want to suggest.
But for the most part, they all fall under one central class of solution type.
That being, to have some sort of automated algorithm determine the district maps,
rather than continuing to rely on partisan state actors to simply draw the maps however they please.

So having said that much, there are a ton of different algorithms out there for this particular problem. So if we were to agree that an algorithm should be used, which one should we go with? Well let's look back for a moment at the general guidelines for drawing fair district maps.

• Districts should each contain about the same number of people
• Districts should be geographically compact (i.e. uncomplicated in shape)
• Districts should not be drawn to intentionally favor one party over another
• Districts should not be drawn to intentionally favor one class of person over another

Meeting the standards of the second two bullets is simple enough, if we just create an algorithm without thinking at all about party or class of person. So that would just leave figuring out what to do about the first two.

-Meta

 

The first idea that comes to my mind, is that we simply take how ever many congressional seats a state has been allocated, and divide the state up into population proportionate divisions, with either vertically or horizontally drawn lines (whichever results in more square resulting divisions), and then continue to divide those resulting divisions in the same way until we have reached the required number of districts for the state. As for the divisions themselves, simple halves are actually pretty good, but thirds, fourths, fifths etc. can also be used as appropriate in order to achieve the most population proportionate and square districts possible.

What will result from this process will be districts which are firstly each of a similar population size and second are all either perfectly square or square-rectangular in shape (save for the borders with other states). And the borders between the districts will all be perfectly straight lines. This is basically an alternate take on something known as the split method, which I'll mention again latter. For now though, let's refer to this method as the square-rectangle method.

Why squares and square-rectangles you ask? To put it simply, it is simply because rectangles are simple and very easy to understand. And we would like them to be square or as square as possible simply because squares are the most compact form of rectangle, and as such will help us to best meet our second standard. Admittedly, squares themselves are not necessarily the most compact shape of all that we could use, as things like hexagons and circles are more compact. But again, the reason I'd favor squares instead is simply due to their simplicity and the ease at which they can be understood by others. They also make for a very simple process and much simpler algorithm when drawing the maps. This process is so simple in fact, that anyone who's even moderately proficient in Excel could program within an hour or two an Excel macro to create such maps on the fly.

Also, a requirement of straight borders BTW makes it exceedingly difficult for anyone to inject politics or any other unwanted biased factors into the drawing of the maps...though of course since the idea here is to have an algorithm do the work, having politics injected was already sort of an impossibility anyways, but still, the straight lines allow for all of us to see clearly with our own eyes the lack of political influence right there in the results.

-Meta

 

The next idea I want to talk about is something I think we should call the ring method. I prefer the square-rectangle method, but an argument can be made against that one and similar 'class-blind' methods that their 'class-blindness' can in some cases result in certain types of people not getting any representation depending on how their population is distributed across a state.

Most notably I think, there are some cases whereby if you are in a rural part of a state that has a mostly urban population, you might end up by chance with no state representative for the rural portion...and likewise there is a possibility that if you are an urban citizen in a state that is mostly rural, you may end up with no urban state rep. Obviously, it should go without saying that such considerations only apply for states with more than one allocated house seat. And I should also mention that if a state's rural and urban populations are about the same size, or if the minority portion is at least of a significant size (typically about 1/4th the population for states with two districts, or less for more populous states) then such a chance occurrence will never be an issue for that state, at least not if we're using the square-rectangle method.

Still, doesn't hurt to try to address that concern I suppose. Which is what got me to thinking about the potential for this ring method. Basically, it would work in much the same way as the square-rectangle method, except that rather than dividing the state with vertical and horizontal lines, the state would instead be divided proportionally using a series of concentric rings centered around a state's major population centers (i.e. the areas of a state where the population is most densely packed).

If we wanted, we could then even divide those rings up into quadrants to create the districts (using straight lines) as a variation on the method, or simply create enough rings such that the resulting entire areas between them would act as the districts.

Either way, what this would result in would be a set of districts in which every segment of a state, by population density type...from the inner-cities and its urbanites, to suburban and outer cities, all the way to the most rural portions of a state,...each of them would be pretty much guaranteed at least some representation in the house congress.

I do sort of like this method and its variations, because much like the square-rectangle method, it is simple and easy to understand. Though it can get a bit trickier to deal with when it comes time to actually implement such an algorithm. But if a state has only one major population center, its actually even easier to understand than the square-rectangle method, and not really hard to implement at all (and there happens to be at least one state which already approximates this). The complexity comes in though when we start looking at states with more than one major population center. Its not prohibitively complicated or anything, just a little more difficult for one to wrap their head around without seeing it.

Then too, one does also have to consider what actually gets to count as a major population center. Take DFW metroplex in Texas for example. Should that count as one population center? Or two? Given the very close proximity of the two cities, I'd say that counting DFW as one population center and putting a ring center-point right in the middle of them would make the most sense, but go to another state where close cities can be a bit further apart and you can begin to see how its possible for a bit of a dilemma to arise in figuring out how to best determine this.

But really...all of those worries are mere trivialities in the grand scheme of things. A simple inconsequential value or two that needs to be defined for our algorithm which in the end wont actually have much impact at all as long as we avoid extremes. That's not to say that the ring method doesn't have real problems though. We start to see them actually when we look back and compare the method to our basic guidelines...

• Districts should each contain about the same number of people
• Districts should be geographically compact (i.e. uncomplicated in shape)
• Districts should not be drawn to intentionally favor one party over another
• Districts should not be drawn to intentionally favor one class of person over another

No issue with the first one or the third one. As for the second...the inner most circles of a ring method-based district map would actually be the most compact form of district possible...which is good...but, each ring as we get further from the center would be less and less compact, and the outermost areas of the map wouldn't be very compact at all, geographically speaking. Though using the quadrant variation could help, or even a variation in which more divisions were added the further from the center we got. Even without those variations though, even then it could still be argued that while not compact geographically, each area between the rings would be compact in the sense that all the people within them would tend to have similar lifestyles to one another and by extension similar needs. So I wouldn't consider that failure too much of an issue in this case,

No...where the ring method really falls flat on its face is when we compare it with the fourth guideline, as it pretty much flies in the face of that one due to this method being essentially based around the idea that divisions will be made based on the type of area a person lives in (i.e. densely populated vs sparsely populated). Fortunately, the algorithmic nature of the method means that one class of people still wont ever gain any significantly unfair advantage over any other class, but it can be said that it gives some minority classes representation that they wouldn't otherwise have, and arguably shouldn't. And its ironic really, and kind of weird when you think about it, that this method's main flaw also happens to be at the same time the very reason we might even consider using it in the first place. Lol.

Anyways...I personally don't find that last fault to be too terrible, as there is somewhat of a justification for it, and at the very least, at least its not based on something silly like race. But, regardless of however significant or insignificant we might find those failings to be, I still think that overall even with them, the ring method is a huge step up in fairness from the wild-west anything goes way our districts are drawn today.

And speaking of the wild-west...does anyone else think that Illinois's 4th district looks like a horse or cow's head?.....Or maybe a dog?....German Shepard perhaps?....Doberman?....Scotch Terrier?

Let me know what you guys see.
(more ideas to come latter)

-Meta

 

Yup, I agree with this. And there is a quote which perfectly encapsulates the whole of that sentiment.
Will have to dig it up and post it latter. But anyways, my hope is that we can solve much of that issue,
by having an agreed upon algorithm which, as you said, is decided upon ahead of time and
based itself upon what we collectively decide on as equitable and fair.

Well, no, not really. Because its the people that actually do the voting, not the land mass.
So districts shouldn't be drawn based on how much land-mass they contain, rather on how many people are within each one.
What you want to do is make it so that for each x number of people that group of x gets a representative.

Otherwise, we'd end up with 'representatives' in congress who didn't represent anyone or anything other than empty space.
And I'm pretty sure we don't want that.

I agree that it isn't possible if we try to make all the districts the exact same physical size. But again, making districts the same physical size isn't what we're trying to do here. What we want instead, is for the districts to each contain about the same number of people, regardless of what land area the district ends up taking up in the end.

Perhaps where the confusion comes in, is in my use of the term compact?
So I should clarify...'compact' in the context of redistricting, does not mean small in size per se, or really anything like that.
What compact means, is that a district is uncomplicated in shape...that its total area is as congregated towards its center point as possible. i.e. that it doesn't have a lot of random tentacles spreading out in all different directions and isn't overly flat or elongated in any areas. This is why I say that a circle is the perfect district shape when only looking at the standard of compactness.

So in the previously mentioned square-rectangle method, we'd want the districts to be reasonably compact, and luckily, squares and rectangles while not as perfect as circles, aren't really all that much less compact. But note that that compactness does not mean that each district will necessarily be the exact same size. Particularly around large population centers (i.e. the cities) districts will naturally be smaller than in the rural areas in order to maintain the standard of each district containing about the same number of people. The same thing can be said, for the most part, of the districts we have today, or of districts created using the ring method, or really any method that follows the standard of keeping the districts population proportionate.

So great...now I have to figure out some way to retcon that explanation into the OP >_<

-Meta

 

Well, if we're still talking federal house representatives and only changing how the lines are drawn (e.g. not changing how many reps or districts there are total) then I'm pretty sure that almost every state, if not actually every one of them has more counties in them than they have congressional districts.

Local districts on the other hand (as in the ones used for electing local statehouse reps or other local officials)
is another matter, and another topic entirely...(well, I shouldn't say entirely...its related...just not the main focus)

-Meta

 

The total number of congressional districts in the country is 435.
The total number of counties in the country is 3,007.

That said, I do have something I want to say on counties...but I'll save that for latter...

-Meta

 

I think those are both great points! With the ideas I've posted so far, I was just going forward with the assumption that we would simply keep things mostly the same for the most part, and only change how the dividing lines of districts were drawn in order to make them more fair, while at the same time keeping the number of districts and representatives the same, and also keeping the house setup (apart from the redistricting process) the same as well. But I'm certainly open to discussing more fundamental changes to the overall system, and like you were saying a more fundamental change may be exactly the sort of thing we really need.

-Meta

 

I agree that a more fundamental change to our system could be beneficial to us in the long run. But...assuming that we are unable to agree on what such a more fundamental change should actually be, I do not agree that such change is necessary if all we then want to do is to just make the lines themselves under the current system more fair.

There is already a general sense out there of what the house should be given what our country's founders set up. The districts as they were originally set up were, and still are, intended to allow state's to have representation in congress based on their relative population. Given that, there's really no reason why we shouldn't be able to come to some sort of agreement on how to do that in the fairest way possible, tweaking nothing more than the finer details of the various processes used to get us there...

-Meta

 

Oh! Easy answer to that question. You Don't!

I mean really...when drawing up district maps its better if one simply does not consider or even think about such various classifications of people. Because that's where the bias starts to seep in. Remember that fourth bullet point? Remember that I actually called it a fault in the ring method that that approach even so much as brought population density into the equation? Though there are several extenuating circumstances in that case. You start to bring things like race and or income class into the mix, and you're really just asking for trouble. So it really is just better to leave those consideration out of the process entirely.

One cannot give every single minority out there a voice in a congress of limited representatives.
And again, one is simply asking for trouble if they even attempt such a thing.
And if you give preference to only one or a few, then other minorities are rightfully going to accuse you of bias.

For that reason, its really best to just define districts without those considerations, and then sit back and let democracy do as it was intended. The wishes of those various minorities will rise up to prevalence through the democratic process should enough individuals share their views. And if there aren't enough who agree with those minorities such that the minority gets to have its way, well then, I'm sorry, but that's kind of the way democracy and majority rule are supposed to work. Majority makes the rules, and the minority should be content in the understanding that while they don't get to run things they are at least afforded the constitutional/judicial protection of a few key and unalienable rights.

-Meta

 

Heh, our congress (as created under the current flawed set up) are incompetent, and may very well be incapable of handling something like this on their own. But I believe that we here at politicalforum certainly have what it takes to come to a consensus. Though I can't guarantee that even if we were to decide on something as a site, that we wouldn't still get the occasional screamer. That said, the odd screamer should not be considered as representative of the majority, or even as a representative of a significant minority for that matter. And perhaps given time, our population as a whole can come to some sort of consensus as well. At that point, all that would be left to do would be to tell congress what it is we want them to do,...and to immediately start kicking them out of office if they refuse.

-Meta

 

Oh I see what you're saying now, and yeah that does make sense. Limit the potential for bias by limiting where the politicians can draw the lines to the county boarders. It would work well for a lot of states. May get just a bit wonky on a few. I have pictures actually. Will post later...

-Meta

 

OK, so if you've read my previous posts in this thread you might have noticed a few key similarities between the ideas I've posted so far. That being that they are each defined algorithms and second being that I choose and favor ones with greater simplicity to them, and there is a good reason for those two things. Firstly, the goal here is to try to remove all the bias that politicians regularly insert into these district maps, or to at least remove as much as we can. Things like limiting where and or how they can draw the lines helps,...having a mathematical equation draw the lines helps even more...but if we really want things to work out...if we want people to have faith in the fairness and sincerity of it all...then those mathematical equations, or at least the basic ideas behind them, need to be simple enough for the average citizen to understand. As my buddy hawkingdo succinctly puts it,

"...are we really willing to let a machine draw our maps? To put the job in a black box and assume it knows best?...the problem with people is that they're corrupt. The problem with algorithms is that they're hard to understand..."
​

So with that in mind, while we're sitting over here thinking of different algorithms we can use to remove what is...essentially political corruption...we should remember to take simplicity into consideration along the way and give favor towards those solutions which are easier to understand. Well...per Einstein, that should really be a general rule when coming up with solutions in any circumstance and perhaps a rule for life as a whole, but I think that it is extra relevant for stuff like what we're dealing with.

BTW, that writer, hawkingdo, has come up with and or covered a lot of great ideas for how to address the issue of gerrymandering in the article below. A few were so good that I want to make sure that they get brought to light in this thread. There is also a pretty good video explanation of gerrymandering itself in there for anyone who's still a bit confused about how it all works and why its an issue.
https://observationdeck.kinja.com/i-solved-gerrymandering-sorta-1606089817

Coincidentally, I don't exactly agree with every single view the writer describes in the piece. Those who've read my posts from other threads can probably guess the points at which we diverge, but other than those its almost eerie just how much our views overlap...but then...maybe not...I mean, after-all we're all living in this same country, witnessing and lamenting over these same silly self-imposed problems...am I right? @perotista, you'd likely see some overlap with your views as well, as like your idea , hawkingdo also puts a large emphasis on limiting how the politicians draw the lines rather than removing the power from them entirely.

In fact, the main idea described is just that. An equation which doesn't draw the maps for us, but rather acts as a measure for how well a drawn map fits our compactness/lack of complexity standard from before.

(see next post)

-Meta

 

So yeah, these next few ideas aren't mine, but come from hawkingdo's piece on gerrymandering. I'm just going to attempt to summarize them, so read the piece for a fuller explanation.
https://observationdeck.kinja.com/i-solved-gerrymandering-sorta-1606089817

The best idea described there imo is the Complexity Ratio standard.
Basically, the idea behind this one is that, rather than having an algorithm draw the maps, we have an explicitly defined standard instead which all districts on all district maps are then required to meet.

The complexity ratio is a standard for the compactness of the districts. hawkingdo represents it with the equation:
Complexity = Circumference^2 / Area <= 40.
So essentially, the more complex a district is, the less compact it is, and the higher that complexity number. hawkingdo suggests we aim for districts with a complexity that is less than or equal to 40, which looking at the districts and their respective complexities, seems like a pretty good standard to me.


hawkingdo doesn't mention this part, but I think it makes sense that if we were to have a standard for compactness/complexity, thereby satisfying our second guideline, we would naturally also want to set up a similar standard to satisfy the first guideline (of each district having about the same number of people) by coming up with some value to act as the Proportionality Ratio standard. What should that standard be? Well, when using the square-rectangle method to draw district maps, it is well within reason to produce districts in which the population differs between them no more than 2%. So...just to give us a bit of wiggle room, I would say that a standard maximum variance for proportionality of say 4-5%...let's just say 4% would be pretty reasonable. Any thoughts on this?

-Meta