# Can you devise a district drawing rule that would eliminate gerrymandering?

**URL:** https://boards.straightdope.com/t/can-you-devise-a-district-drawing-rule-that-would-eliminate-gerrymandering/654667
**Category:** Factual Questions
**Created:** [April 2, 2013, 9:31pm UTC](https://boards.straightdope.com/t/can-you-devise-a-district-drawing-rule-that-would-eliminate-gerrymandering/654667 "2013-04-02T21:31:36Z")
**Posts on this page:** 20
**Page:** 3

<div class="post-metadata">

### Author: ![Really\_Not\_All\_That\_Bright](https://avatars.discourse-cdn.com/v4/letter/r/e8c25b/32.png) [@Really\_Not\_All\_That\_Bright](https://boards.straightdope.com/u/Really_Not_All_That_Bright)
#### Post date: [April 3, 2013, 5:23pm UTC](https://boards.straightdope.com/t/can-you-devise-a-district-drawing-rule-that-would-eliminate-gerrymandering/654667/41 "2013-04-03T17:23:16Z")

</div>

Yes, it’s fair in that way. It’s completely unfair in the sense that everyone’s vote will count for a differing amount.

---

<div class="post-metadata">

### Author: ![Clothahump](https://avatars.discourse-cdn.com/v4/letter/c/51bf81/32.png) [@Clothahump](https://boards.straightdope.com/u/Clothahump)
#### Post date: [April 3, 2013, 5:24pm UTC](https://boards.straightdope.com/t/can-you-devise-a-district-drawing-rule-that-would-eliminate-gerrymandering/654667/42 "2013-04-03T17:24:34Z")

</div>

Use Post Office Zip code boundaries.

---

<div class="post-metadata">

### Author: ![Steve\_MB](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/steve_mb/32/5339_2.png) [@Steve\_MB](https://boards.straightdope.com/u/Steve_MB)
#### Post date: [April 3, 2013, 5:28pm UTC](https://boards.straightdope.com/t/can-you-devise-a-district-drawing-rule-that-would-eliminate-gerrymandering/654667/43 "2013-04-03T17:28:05Z")

</div>

> [@Great\_Antibob](#):
>
> I have no clue how practical it is, but it is pretty simple to feed a computer a program with some basic guidelines - total population, population density, wealth, race, religion, whatever, and have it generate N districts that fit your criteria, which may include keeping districts reasonably compact (no splitting Austin, TX into 4 districts, one of which extends to Mexico and another to Houston, for example) and maintains some kind of overall diversity guidelines.

A simpler version would be an algorithm that provides some numerical measure of total district compactness (e.g. area of district divided by inhabited area within the smallest circle containing the entire district, combined for all districts). Any and all comers submit maps; the one with the best score wins (thus, interested parties can slip in a bit of preference for their favored political, ethnic, etc faction, limited by the need to avoid losing the bid).

Edit: **Chronos** had a similar proposal.

---

<div class="post-metadata">

### Author: ![bleach](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/bleach/32/3106_2.png) [@bleach](https://boards.straightdope.com/u/bleach)
#### Post date: [April 3, 2013, 6:03pm UTC](https://boards.straightdope.com/t/can-you-devise-a-district-drawing-rule-that-would-eliminate-gerrymandering/654667/44 "2013-04-03T18:03:55Z")

</div>

> [@Clothahump](#):
>
> Use Post Office Zip code boundaries.

I think you proposed this [before](http://boards.straightdope.com/sdmb/showpost.php?p=12854015&postcount=35), and you got these responses:

> [@Anti-gerrymandering measures on Florida ballot](https://boards.straightdope.com/t/anti-gerrymandering-measures-on-florida-ballot/551396/39):
>
> Zip code areas are based on volume of mail, nothing to do with the number of eligible voters who live there. There are many office buildings that are a complete zip code on their own, but don’t have a single person living in the building. And we have a small town in Minnesota that does the processing for most coupons and mail-in rebates in the country – it is a very small town, about 1,000 houses – yet it has over 20 zip codes assigned to it.

> [@Anti-gerrymandering measures on Florida ballot](https://boards.straightdope.com/t/anti-gerrymandering-measures-on-florida-ballot/551396/40):
>
> And even if zip codes were uniform with population, they’re still too small. There are over 40,000 zip codes in the US, and only 435 House districts, which means that the average House district will have close to 100 zip codes in it. That’s enough resolution to allow for plenty of gerrymandering.

---

<div class="post-metadata">

### Author: ![septimus](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/septimus/32/410_2.png) [@septimus](https://boards.straightdope.com/u/septimus)
#### Post date: [April 3, 2013, 7:40pm UTC](https://boards.straightdope.com/t/can-you-devise-a-district-drawing-rule-that-would-eliminate-gerrymandering/654667/45 "2013-04-03T19:40:06Z")

</div>

We’ve discussed this in several other threads, without reaching any firm conclusion other than the desirability of non-partisan districting like California’s.

Gerrymandering usually manifests itself by the non-gerrymandering side winning by large margins. For example, [Democrats won only 4 of 16 Congressional seats in Ohio 2012, but these were 4 of the 5 _least contested_ districts](http://ballotpedia.org/wiki/index.php/United_States_House_of_Representatives_elections_in_Ohio,_2012). The GOP seat margin of 12-4 was despite a voter margin of only 51.4 - 46.6.

Based on this, I made a suggestion which I think is worth repeating: _Give added benefit, e.g. seniority, to Congressmen who are elected by the highest margins._ In the Ohio example, the 4 Demos, “super-elected” because of gerrymandering, would have superior status.

---

<div class="post-metadata">

### Author: ![The\_Lurker\_Above](https://avatars.discourse-cdn.com/v4/letter/t/dc4da7/32.png) [@The\_Lurker\_Above](https://boards.straightdope.com/u/The_Lurker_Above)
#### Post date: [April 3, 2013, 8:03pm UTC](https://boards.straightdope.com/t/can-you-devise-a-district-drawing-rule-that-would-eliminate-gerrymandering/654667/46 "2013-04-03T20:03:06Z")

</div>

> [@septimus](#):
>
> Based on this, I made a suggestion which I think is worth repeating: _Give added benefit, e.g. seniority, to Congressmen who are elected by the highest margins._ In the Ohio example, the 4 Demos, “super-elected” because of gerrymandering, would have superior status.

Or extra (fractional) votes in the legislature based on margin of victory. So say a 50%+1 margin gets you 1.0 votes, a 75% margin gets you 1.5 votes and 100% gets the legislator 2.0 votes. With computers the vote counting in the legislature shouldn’t be too cumbersome.

Get the scaling right and it should it reduce the temptation to gerrymander in the first place, and also encourage voters in safe districts to get out and vote to increase/decrease the power of the winner.

---

<div class="post-metadata">

### Author: ![Senegoid](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/senegoid/32/6606_2.png) [@Senegoid](https://boards.straightdope.com/u/Senegoid)
#### Post date: [April 3, 2013, 8:04pm UTC](https://boards.straightdope.com/t/can-you-devise-a-district-drawing-rule-that-would-eliminate-gerrymandering/654667/47 "2013-04-03T20:04:17Z")

</div>

Of course, we discussed and read about gerrymandering when I took American Government in college. Gerrymandering has taken place to distort voting districts since the early days of the nation, so the problem (as it currently is) isn’t just some recent artifact of Supreme Court decisions (as implied by one post above).

The practice has happened every place on Earth and every time in political history, where voting is done by districts. The only particularly unique thing about American-style gerrymandering is the name “gerrymander”.

Okay, how about this suggestion for political divisions without human intervention:

Let an entire State be divided into virtual districts by population, so that each district will contain roughly equal numbers of voters. But they will not be geographic districts. Instead, a computer program will assign every voter state-wide to one of these “districts” at random.

So, for example, if each district has 10000 people, then the 10000 people in each district will be scattered randomly across the state instead of all gathered into one geographic area. And likewise for each other district.

An implication of this would be that every district would be approximately the same as every other district, rather than having some areas be mostly Democrat and some be mostly Republican. And the political composition of each of the districts would roughly match the overall political composition of the entire State.

(So, come to think of it, why have district-based elections at all?)

---

<div class="post-metadata">

### Author: ![bleach](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/bleach/32/3106_2.png) [@bleach](https://boards.straightdope.com/u/bleach)
#### Post date: [April 3, 2013, 11:39pm UTC](https://boards.straightdope.com/t/can-you-devise-a-district-drawing-rule-that-would-eliminate-gerrymandering/654667/48 "2013-04-03T23:39:04Z")

</div>

I’m not sure that I’m following your plan **Senegoid**. Your second-to-last sentence sounds like it would be similar to a popular vote.

It would be difficult for a candidate to meet with constituents if his constituents are randomly scattered across the state. Wouldn’t a candidate have to compete state-wide for an office then? Or are you suggesting that the top XX vote-getters in the state will represent the state at the Federal level?

I would suggest that districts are important because of unique problems that each district faces. Different political issues are important to different strata of people. If a candidate can’t represent his local district, then what’s the point of having a representative government? People tend to homogenize into enclaves exactly because they like to be around like-minded people (e.g., **[The Big Sort](http://www.amazon.com/The-Big-Sort-Clustering-Like-Minded/dp/0618689354/ref=sr_1_1?ie=UTF8&qid=1365032055&sr=8-1&keywords=big+sort)** by Bill Bishop).

It’s likely that I’m misunderstanding your scheme.

---

<div class="post-metadata">

### Author: ![septimus](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/septimus/32/410_2.png) [@septimus](https://boards.straightdope.com/u/septimus)
#### Post date: [April 4, 2013, 12:23am UTC](https://boards.straightdope.com/t/can-you-devise-a-district-drawing-rule-that-would-eliminate-gerrymandering/654667/49 "2013-04-04T00:23:21Z")

</div>

> [@Senegoid](#):
>
> So, for example, if each district has 10000 people, then the 10000 people in each district will be scattered randomly across the state instead of all gathered into one geographic area.

Note that in the Ohio example, where “The GOP seat margin of 12-4 was despite a voter margin of only 51.4 - 46.6,” the GOP would now win probably(_) all 16 seats … unless there was a small sentiment shift, then the Demos would win all 16. (_ - A particularly popular Dem would have a chance, but GOP might run their most popular candidate against him. Or are candidates to be randomly assigned also?)

What makes districting complicated is that “flocking birds of a feather together” is _one of the goals_ – it’s just bad when that gets manipulated “unfairly.” In other threads, disadvantages have been mentioned for making all seats at-large.

BTW, a key virtue of the seniority plan I mentioned is that it could be implemented just with a Congressional resolution – no Constitutional change required. (Admittedly such a resolution would not be easy to pass.)

---

<div class="post-metadata">

### Author: ![TriPolar](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/tripolar/32/3008_2.png) [@TriPolar](https://boards.straightdope.com/u/TriPolar)
#### Post date: [April 4, 2013, 12:39am UTC](https://boards.straightdope.com/t/can-you-devise-a-district-drawing-rule-that-would-eliminate-gerrymandering/654667/50 "2013-04-04T00:39:57Z")

</div>

> [@Senegoid](#):
>
> Let an entire State be divided into virtual districts by population, so that each district will contain roughly equal numbers of voters. But they will not be geographic districts. Instead, a computer program will assign every voter state-wide to one of these “districts” at random.
> 
> So, for example, if each district has 10000 people, then the 10000 people in each district will be scattered randomly across the state instead of all gathered into one geographic area. And likewise for each other district.
> 
> An implication of this would be that every district would be approximately the same as every other district, rather than having some areas be mostly Democrat and some be mostly Republican. And the political composition of each of the districts would roughly match the overall political composition of the entire State.
> 
> (So, come to think of it, why have district-based elections at all?)

So now you end up with a state that is heavily rural or heavily urban only representing that majority population. This is especially bad in populous states where other groups also become unrepresented.

---

<div class="post-metadata">

### Author: ![morgensd](https://avatars.discourse-cdn.com/v4/letter/m/82dd89/32.png) [@morgensd](https://boards.straightdope.com/u/morgensd)
#### Post date: [April 4, 2013, 1:25am UTC](https://boards.straightdope.com/t/can-you-devise-a-district-drawing-rule-that-would-eliminate-gerrymandering/654667/51 "2013-04-04T01:25:56Z")

</div>

> [@bleach](#):
>
> I would suggest that districts are important because of unique problems that each district faces. Different political issues are important to different strata of people. If a candidate can’t represent his local district, then what’s the point of having a representative government? People tend to homogenize into enclaves exactly because they like to be around like-minded people (e.g., **[The Big Sort](http://www.amazon.com/The-Big-Sort-Clustering-Like-Minded/dp/0618689354/ref=sr_1_1?ie=UTF8&qid=1365032055&sr=8-1&keywords=big+sort)** by Bill Bishop).
> 
> It’s likely that I’m misunderstanding your scheme.

What bleach said. it may be a cliche, but the adage about all politics being local has some truth. If I was asked to join a virtual district I would most likely look for one that includes others in my local area so that we would be likely to be aligned on local issues such as zoning, property taxes and the like.

I like the approach in California. Not that our fine legislators in Indiana will ever consider it.

morgensd

---

<div class="post-metadata">

### Author: ![Learjeff](https://avatars.discourse-cdn.com/v4/letter/l/94ad74/32.png) [@Learjeff](https://boards.straightdope.com/u/Learjeff)
#### Post date: [April 4, 2013, 2:07am UTC](https://boards.straightdope.com/t/can-you-devise-a-district-drawing-rule-that-would-eliminate-gerrymandering/654667/52 "2013-04-04T02:07:25Z")

</div>

Re: **Senegoid, tripolar, and septimus**

Right – as I said above, lack of districts leads to underrepresentation of minorities and doesn’t provide local representation.

---

<div class="post-metadata">

### Author: ![sachertorte](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/sachertorte/32/3727_2.png) [@sachertorte](https://boards.straightdope.com/u/sachertorte)
#### Post date: [April 4, 2013, 2:31pm UTC](https://boards.straightdope.com/t/can-you-devise-a-district-drawing-rule-that-would-eliminate-gerrymandering/654667/53 "2013-04-04T14:31:40Z")

</div>

> [@dracoi](#):
>
> Here’s the problem I see: any formula is going to slightly favor one party over another, at least at the time that it is passed. It’s a given, really. It would be impossible to change the rules and yet also maintain the status quo.
> 
> So… the formula to prevent gerrymandering is just more gerrymandering.

I would commission a formula but legislate its enactment for sometime in the future (e.g., 2030). That way, futzing with the formula based on the current situation is more risky. I don’t know, maybe I’m being naive.

---

<div class="post-metadata">

### Author: ![GoodOmens](https://avatars.discourse-cdn.com/v4/letter/g/d26b3c/32.png) [@GoodOmens](https://boards.straightdope.com/u/GoodOmens)
#### Post date: [April 4, 2013, 3:08pm UTC](https://boards.straightdope.com/t/can-you-devise-a-district-drawing-rule-that-would-eliminate-gerrymandering/654667/54 "2013-04-04T15:08:02Z")

</div>

“All Congressional districts must be congruent, regular hexagons.” That would do it.

I’m not saying that it’s a PRACTICAL rule.

---

<div class="post-metadata">

### Author: ![Kent\_Clark](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/kent_clark/32/105_2.png) [@Kent\_Clark](https://boards.straightdope.com/u/Kent_Clark)
#### Post date: [April 4, 2013, 3:23pm UTC](https://boards.straightdope.com/t/can-you-devise-a-district-drawing-rule-that-would-eliminate-gerrymandering/654667/55 "2013-04-04T15:23:03Z")

</div>

Everyone proposing some kind of geometric boundary is forgetting the [14th Amendment](http://www.archives.gov/exhibits/charters/constitution_amendments_11-27.html#14).

> [@](#):
>
> Representatives shall be apportioned among the several States according to their respective numbers, counting the whole number of persons in each State, excluding Indians not taxed.

Any plan that doesn’t account for equal population for each district won’t pass Constitutional muster.

Carry on.

---

<div class="post-metadata">

### Author: ![Saint\_Cad](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/saint_cad/32/18907_2.png) [@Saint\_Cad](https://boards.straightdope.com/u/Saint_Cad)
#### Post date: [April 4, 2013, 5:49pm UTC](https://boards.straightdope.com/t/can-you-devise-a-district-drawing-rule-that-would-eliminate-gerrymandering/654667/56 "2013-04-04T17:49:02Z")

</div>

First of all, allow for at-large elections.

But more to the OP. take the ratio of the distance from the centroid of the district to the most distant boundry point to the distance from the centroid to the closest boundary point.

The lower the geometric mean of the ratios of all of the districts the better the apportionment with perfection at a gemetric mean of 1 (all districts are circles).

---

<div class="post-metadata">

### Author: ![yabob](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/yabob/32/2821_2.png) [@yabob](https://boards.straightdope.com/u/yabob)
#### Post date: [April 4, 2013, 5:58pm UTC](https://boards.straightdope.com/t/can-you-devise-a-district-drawing-rule-that-would-eliminate-gerrymandering/654667/57 "2013-04-04T17:58:04Z")

</div>

> [@Saint\_Cad](#):
>
> First of all, allow for at-large elections.
> 
> But more to the OP. take the ratio of the distance from the centroid of the district to the most distant boundry point to the distance from the centroid to the closest boundary point.
> 
> The lower the geometric mean of the ratios of all of the districts the better the apportionment with perfection at a gemetric mean of 1 (all districts are circles).

Actually, if you are going to start with a list of centers, the classic technique is Dirichlet Tessalations:

> **[Voronoi diagram](https://en.wikipedia.org/wiki/Voronoi_diagram)**
>
> In mathematics, a Voronoi diagram is a partition of a plane into regions close to each of a given set of objects. It can be classified also as a tessellation. In the simplest case, these objects are just finitely many points in the plane (called seeds, sites, or generators). For each seed there is a corresponding region, called a Voronoi cell, consisting of all points of the plane closer to that seed than to any other. The Voronoi diagram of a set of points is dual to that set's Delaunay triangul...

I don’t think the result would be very satisfactory.

As other posters have suggested, I think there are too many mutually incompatible goodness measurements for districts to make any purely automated border drawing acceptable.

---

<div class="post-metadata">

### Author: ![yabob](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/yabob/32/2821_2.png) [@yabob](https://boards.straightdope.com/u/yabob)
#### Post date: [April 4, 2013, 6:12pm UTC](https://boards.straightdope.com/t/can-you-devise-a-district-drawing-rule-that-would-eliminate-gerrymandering/654667/58 "2013-04-04T18:12:41Z")

</div>

I think I’ve raised the suggestion before of simply electing all the reps at large. In order to make this work, I would:

1. require that each candidate make a clear statement of who they view their constituency to be, which may or may not be geographic. A candidate can say something like “I want to represent the people of color” or “I wish to serve the LGBT community” rather than “I will represent the interests of {geographic area}”.

2. it works better if we also allow voters to split their vote in some way, since any individual is a member of several groupings. The gay shop owner in Carmel may be interested in the LGBT candidate, or the one concerned with the interests of the California coast communities. The Evangelical Christian farmer south of Fresno may be interested in the Christian Family Values candidate, or the one pledging to represent the interests of the San Joaquin valley.

---

<div class="post-metadata">

### Author: ![shiftless](https://avatars.discourse-cdn.com/v4/letter/s/e9c0ed/32.png) [@shiftless](https://boards.straightdope.com/u/shiftless)
#### Post date: [April 4, 2013, 6:23pm UTC](https://boards.straightdope.com/t/can-you-devise-a-district-drawing-rule-that-would-eliminate-gerrymandering/654667/59 "2013-04-04T18:23:04Z")

</div>

I have worked on some very similar problems professional. The issues become very complex very fast. We are talking about nice square, heavily populated blocks in New York city compared to huge, sparcely populated areas in Alaska. In a grid or other set sized district the population will vary wildly. Trying to group pop into as compact a polygon as possible is easy for the first polygon, gets harder for its neighbor and harder still when juggling 20 districts. If we are going to respect state boundries the task of equal representation is impossible to begin with.

But in the end, whatever algorithm we use will be designed by a human and therefore a human will decide which element of a choice is more important than another. And even if that were possible to solve, we would just create a political system that played specifically to the current redistricting algorithm.

---

<div class="post-metadata">

### Author: ![yabob](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/yabob/32/2821_2.png) [@yabob](https://boards.straightdope.com/u/yabob)
#### Post date: [April 4, 2013, 6:36pm UTC](https://boards.straightdope.com/t/can-you-devise-a-district-drawing-rule-that-would-eliminate-gerrymandering/654667/60 "2013-04-04T18:36:27Z")

</div>

And I have another odd suggestion:

If we wish to retain geographically defined districts, how about what I might call “Fuzzy districting with correction”. It might work something like this:

The voter receives a ballot containing not just their district, but all the bordering ones. They get to check one of the districts and vote in it.

Districts are redrawn after each election, using the results concerning how many voters located in a given district “jumped ship” to an adjoining one. If a whole lot of people in district A chose to vote in district B, you must move the border to include some of the A district in B. If practically everybody in a district jumps ship, that illustrates that the district should not exist anymore, and its voters should be distributed among the bordering districts. If a lot of people are voting in a particular district, it indicates that it ought to be split in two.

The rationale:

Gerrymandering of districts may be done by distributing a block of voters around several districts to assure that they cannot make an impact in any of them. Through grassroots efforts, that block is likely to decide that one of the districts is the one they really belong in, and will defeat the efforts.

The classic “Gerrymander” refers to an oddly shaped district drawn in such a way as to artificially concentrate a group within it - a group which would not have substantial numbers in any reasonably shaped districts. If this happens, you are likely to create a district which borders on many others. You have the potential for a lot of voters jumping into it, and forcing the gerrymandered district to split.

I’m thinking that over time, the districts will iteratively become reasonably defined with respect to the various interest groups involved.

[Previous page](https://boards.straightdope.com/t/can-you-devise-a-district-drawing-rule-that-would-eliminate-gerrymandering/654667.md?page=2)

[Next page](https://boards.straightdope.com/t/can-you-devise-a-district-drawing-rule-that-would-eliminate-gerrymandering/654667.md?page=4)
