What Would the Nation Without Gerrymandering Look Like?

Vox: “What would a world without gerrymandering look like? Check out the map below, in which each colored district has a roughly equal population, for one possible glimpse. (Note that this map draws districts that cross state borders as well, which is impossible under our current system, but would end the overrepresentation of some small states.)”

“The map was created by the Center for Range Voting, which was founded by math PhD Warren Smith and engineer Jan Kok to float innovative election reform proposals. To make it, they used what they call the shortest splitline algorithm. Basically, they used the shortest possible line to cut a state into two halves with roughly equal populations. Then they did so again, and again, and again, until they had the proper number of overall districts.”

FavoriteLoadingSave to Favorites
  • stpaultim

    Well, that’s cute and pretty and all, but how about doing the same thing within state boundaries. You know, one that might actually be constitutional?

    • Calbengoshi

      There is a link in the Vox article that enables you to see the same sort of division into districts for each state.

  • Hayaka

    Really. What is the point of crossing state boundaries? You might as well throw some Canadian provinces in there if you’re going to ignore a fundamental reality of American politics.

  • Calbengoshi

    The problem with basing districts along straight lines, as shown by the separate state map accessible in the underlying article, is that it doesn’t take into account pre-existing intra-state boundaries, such as city limits and county lines. While it may make sense solely from a statistical perspective to have the residents of a city split between two different members of Congress, unless that city has a population large enough to warrant having more than one congressional district it makes no political sense to split up that municipality into more than one congressional district.

  • Detroitdic

    Fair representation at each level of government is a goal worth pursuing and does’nt have to be fraught with difficulty and angst. It’s hard to say where the process got off track for the dems but the GOP has had a headlock on redistricting for decades in Michigan. Spell it with an E for Engler.

  • Rhodent

    This sounds great in theory, but in practice it doesn’t work very well.

    If you go to http://rangevoting.org/SplitLR.html you can see maps where they do it on a state level (which, as others in this thread have mentioned, is the only way this could possibly be done without first enacting a Constitutional Amendment that has zero chance of getting ratified). But when you look at the states, you see the same two problems happening over.

    First, the largest cities are split among multiple districts, diluting their power. When the issue is that a city is too large to fit in a single district, that’s not necessarily an issue, but it happens when that’s not the case. For example, North Carolina has thirteen districts, meaning that each district should have roughly 7.7% of the population. Raleigh, the second largest city, is 4.3% of the population, but gets split among four districts. Another extreme example is Denver, which is split among five districts, and all but one of those five extend all the way to the edge of the state (one of them includes roughly one quarter of the area of the state).

    Second (and related to the above) is that just about every state with two districts finds its largest city split into those two districts. The best example of how absurd this can be is probably Hawaii. In the real world, Hawaii’s 1st district is Honolulu and its suburbs and the 2nd is the rest of the state. In this system, Honolulu gets split in two. Similarly, each of New Mexico’s three districts includes roughly one third of Albuquerque, and each of Utah’s three districts includes part of Salt Lake City.

    But the area where this system really falls apart is that after it splits up to many cities, it has a tendency to join portions of farflung cities together. A few of the more extreme examples are as follows:

    1. In New York, half of Rochester gets put in a district with part of Buffalo, and the other half gets put in a district with part of Syracuse.

    2. In Virginia, Richmond gets split into three districts. One of those three reaches up to the D.C. suburbs in the northern part of the state, a second reaches to the Tidewater area in the southeast portion of the state, and the third reaches all the way to Danville in the southcentral part of the state.

    3. Ohio has one district that includes part of Dayton and part of Cincinnati, and another district that includes another part of Dayton and another part of Cincinnati.

    4. In Maryland, there’s a district which is basically the Eastern Shore. So far, so good. But there are some islands in the Chesapeake which are just off the Eastern Shore, culturally identical to the Eastern Shore, which somehow manage to be in the same district as downtown Baltimore.

    I agree that something needs to be done about gerrymandering, and there are certainly states where this system would be an improvement over the current setup. But all in all, I think there are much better methods than this.

  • Warren Smith

    Rhodent: ” you see the same two problems happening over.
    First, the largest cities are split among multiple districts…
    Second… it has a tendency to join portions of farflung cities together.”

    Those phenomena can indeed happen, and sometimes do (and sometimes do not). Why is this bad? That, Rhodent does not say, he simply seems to think it is bad, for no stated reason.

    My belief is that (and I have good reason to believe most US citizens examining our maps would agree) that the splitline districting map is superior to the current, officially drawn (i.e. gerrymandered), districting maps, in almost all US states. The flaws Rhodent points out (if indeed it is even correct to consider them to be flaws) are of minor importance compared to the massive intentional biases in the official maps today which cause the US congress re-election rate and election year-ahead predictability rates to be in the 95-100% range. US legislators are more likely to die in office than be defeated in a re-election bid. That is not democracy, it is a sham. Meanwhile, the splitline scheme has no intentional biases and would yield genuine democracy.

    • Warren Smith

      More generally, Rhodent without actually saying so explicitly, implies that districts should be made of “culturally similar” people (whatever that means). Why they should be that way, rather than being, say, diverse, Rhodent does not say. But
      if we go his way then there is only one way to do that: have districters figure out the “cultural type” of the people, then use that info to draw the maps. As soon as they do that, then they are being biased. Do you WANT districters to be intentionally biased? For example, what they often do right now is, they first figure out who is the “republican” cultural type. Another possible “cultural type” might be “enjoys pornography” or “pizza lover.” What should we do about those types? Ignore them? Take them into account? How? Rhodent does not say. And I suggest that if he/she did say, then he/she would be opening one hell of a can of worms.

      Leave that can closed. Just use splitlining. It is very simple,
      it is far cheaper, simpler and faster than the methods used today, it is 100% unbiased, and it will start the USA toward actual democracy for the first time in your life.

Read previous post:
Does the SCOTUS Redistricting Verdict Really Help Democrats?

Nate Cohn: "When the Supreme Court ruled Monday to allow independent redistricting commissions in Arizona, the decision was greeted with...