[tunesmith]

More competitive? Or more representative of the overall population? Because those two things are at odds.

Say you have a state that is split 55/44, and has nine districts. Do you want nine districts each at 55/44, meaning 9-0? That's more competitive, but less representative of the state. Or do you want five districts for one, and four for the other? That's more representative of the state, but less competitive. Redistricting is hard.

(I suspect my scenario is a false choice if you wrestle with the math enough, but I'm not sure. I prefer the Wisconsin test that we all thought Justice Kennedy would decide in favor of, but then he took the coward's way out, probably corrupt too.)

[tialaramex]

If you worry about this, the reform you want is multi-member districts. Instead of voting for one representative people are picking several and the districts are bigger. This is how EU parliamentary elections work for example, a region gets say 5 slots and it needs to send 5 representatives to fill those slots, elected democratically but the EU doesn't tightly constrain the method used.

There are a few ways you can do this, Jefferson developed one of them so that's got a nice pedigree. It is also a good way for any third party to start to get some attention. Getting 20% of votes together to have your candidate as the fifth member for a district with 2 Dems and 2 Republicans is going to be easier than finding a plurality of voters for a single rep when the two big parties are both saying that's a wasted vote.

[crdrost]

With that said, before you start writing your state legislature about “Multi-member districts TODAY!” you should also know that they have a very difficult history of being shot down by the Supreme Court. This was pointed out to me by a number of good folks on Politics.StackExchange[1].

But yes, the route to avoiding a second US Civil War is only incidentally through the “eliminating gerrymandering” nodes of the graph—the 50/50 saturation of the vote into two spineless political parties happens in both the House (gerrymandered to hell) and the Senate (not gerrymandered at all) and is a consequence of something more fundamental. States do need to switch to the proportional system, but there is a decent chance that the Supreme Court might destroy this out-of-hand, in which case a Constitutional amendment may well be necessary.

[1] “Would Switching to a Proportionate House Require a Constitutional Amendment?” https://politics.stackexchange.com/questions/42551/would-swi...

[thechao]

That response fails to distinguish between multi-member at large districts which are Real Bad^TM, and some form of ranked mechanism which are Real Good^TM. Multi-member ranked districts are both more representative and more competitive than our current system.

[jfengel]

Or doing away with districts entirely, which amounts to the same thing. Right now we elect people as if geography were the only concern that could link voters. That makes it prone to ignoring voters whose concerns differ from that of their immediate neighbors. And very susceptible to manipulation by those who get to define "neighbors".

Geography will always play into it, especially in our odd hybrid system where we insist on devolving many laws to the states (such that even obvious crimes like murder, and apparently-unrelated notions like real estate or health care, end up with 50 separate laws with small but crucial differences). But there are lots of ways to change the system so that you're not solely linked by geography and hoping that, if you're in a local minority, somebody else will elect somebody to represent your view.

[legitster]

This is SUPER important. People think redistricting is a panacea, but it actually sounds like a super tough nut to crack. Even states that have non-partisan districting enshrined in law are going through their own controversies and it doesn't (at first blush) appear to be solving anything.

99pi/538 did an excellent series on the topic I would recommend.

[padobson]

Came here to say something like this.

Chief Justice Roberts' opinion[0] touches on the legislative evolution of the problem. The advent of districts was instituted by congress to encourage proportional representation - and it worked! Gerrymandering is actually an improvement over the older system that led to situations like GP is describing.

Gerrymandering is not the problem, it's a symptom. The problem is first-past-the-post elections. Ranked-choice voting or some other proportional system isn't a panacea either, but it's much better than judges re-drawing political districts instead of elected representatives.

[0]https://www.supremecourt.gov/opinions/18pdf/18-422_9ol1.pdf

[djsumdog]

I agree entirely. Ranked voting or preference with instant runoff (Australia) or MMP (New Zealand) are all much better at representing the population.

They're not perfect of course. People in power will always try to adjust the system to benefit them and their supporters (read: The Dictators Handbook). Even in Australia where voting is mandatory, there is a top and bottom of the ballot and only the top is required (if a party doesn't have a certain percentage, that party is moved to the bottom).

[mikeash]

This seems like a “perfect is the enemy of the good” sort of thing.

Yes, drawing good districts is really hard. Non-partisan redistricting doesn’t solve everything.

But surely it’s a hell of a lot better than allowing whatever political party is in power in a particular year to draw districts to their own advantage, allowing them to rule long after they lose the support of the majority?

You’ll sometimes end up with minority rule in any district based system, but at least it’ll be accidental rather than intentional and self-perpetuating.

[clmckinley]

The problem with this analogy is that the 55/44 split is not equally distributed across a state. In virtually every case there are pockets of one or the other. As long as you keep the goals of uniform, compact and consistent sizes they should do a much better job of getting closer to the 4/5 you are looking for or maybe a couple of safer and a couple of swing districts. What happens now is that the party in control (which in 2010 was largely R) packs and cracks to get an unnatural advantage (looking at you Wisconsin)

My bigger question is when the political split is more like 80/20 case should the 20 be guaranteed representation?

[runarberg]

> My bigger question is when the political split is more like 80/20 case should the 20 be guaranteed representation?

Yes they should. They should get around 20% of the representatives.

[clmckinley]

So then without making other more fundamental changes we would have to gerrymander. In fact a number of strange looking districts are gerrymandered with creating a majority minority district as the stated goal.

I believe we need structural changes to really solve this problem but absent that, I think having a computer algorithm that has the following priorities in the following order:

1) Compact (ie smallest circumference)

2) Least number of axis points and smallest difference in the sides.(ie a square is the goal)

3) Least number of wasted votes (ie use the proposal from the supreme court case)

4) Least change from a previous district (maybe not for this first run)

[matt_holden]

The focus on compactness (or other geometric properties) as a goal is misguided.

The districts don't exist to look pretty or have nice geometric properties. They exist to translate the will of the voters into a representative government.

538 compared maps gerrymandered for various goals [1]. Prioritizing compactness has the effect of a) skewing political power toward rural areas away from cities, and b) reducing the expected number of non-white House members (for similar reasons).

The problem with compactness is it doesn't take into account the political geography of where people live. Increasingly, Democratic majorities are concentrated in cities, while Republican majorities are outside of them. Drawing compact districts has the effect of packing Democrats in fewer (urban) districts, resulting in more wasted votes. This gave the Republican party about a 30 seat advantage in the 538 map, compared to a more proportional one.

The Senate and Electoral College already unfairly skew political power toward rural voters. Compact maps would do the same in the House.

The efficiency gap (your #3) is a much better measure of partisan fairness to focus on than compactness.

[1] https://projects.fivethirtyeight.com/redistricting-maps/#alg...

[runarberg]

Why do we need a large number of small districts? It seems like a relic from an obsolete philosophy that people from a geographically separate areas need a different representation. But that is not how reality works. People want representation based on their class status, not geography. We have local governments to deal with local issues and it is silly to also divide federal representation based on geography.

Alternatively I propose the state being a single district, with representatives directly partitioned based on the proportion of vote they get in the state. So if a state gets 10 representatives in congress, and a party gets 10% of the votes in that state. That party will get 1 representative in congress.

---

Edit: And before this starts being called unrealistic, I believe all we would need is a constitutional amendment (even just in state constitutions) that states that every vote should count the same.

[lsaferite]

My understanding of Gerrymandering is to specifically have the outcome of 9-0 though. The natural distributions would make a natural outcome of 9-0 highly unlikely without rigging the district lines.

[endlessbeing]

Actually, I think gerrymandering aims to have the result be more like 7-2 consistently, where 7 is the party with less actual representation. Gerrymandering aims to put all of the opponent's voters in the smallest number of districts possible, while spreading out your own base.

The result is that the opponent wins fewer districts with over 90% of the vote where you win more districts with 55% or 60% of the vote.

[c3534l]

Non-representative outcomes is a symptom of Gerrymandering, ont the goal of it. The reason for gerrymandering a district varies, but overall it tends to preserve the status quo. Those in power can use it to disenfranchise a group of people, or they can use it to create safe districts for party leaders. Gerrymandering should be thought of as a tool. The illustrations popular online showing how gerrymandering can be used to produce paradoxical representation have done a lot to raise awarenss, but it's also made people confused about what it is. Gerrymandering is a tool, not a goal.

[lozenge]

Gerrymandering uses cracking and packing. Packing is giving your opponent 2 very safe districts (eg urban centre), and cracking is making it impossible for them to win the other 7 (eg making them 30% suburb, 70% rural).

[Rebelgecko]

I think gerrymandering is typically willing to sacrifice a few districts to make the majority of districts less competitive

[on_and_off]

Interesting how everything is only Democrats vs Republicans in the USA.

It probably does not help the political climate that for most people it is just a binary choice.

[tptacek]

That's a consequence of first-past-the-post voting (Duverger's law), not gerrymandering.

[on_and_off]

I should have used quotes, I was answering specifically about the 55/45 part of gp text :)

[souterrain]

I want at-large representatives for the entire state, based on general election polling rank. This is unlikely, of course, because it threatens the two-party hegemon.