More on the shortest split-line algorithm with pretty pictures

I’m a sucker for pictures. They communicate ideas quite succinctly.
Here I show some examples of congressional districts as they are drawn currently and as the shortest split-line algorithm would draw them.

If the images are cropped please click through on them or use your browsers features to view the full image.

Let’s pick on Massachusetts yet again:

Current congressional districts:

From split-line algorithm:

And here is Texas:


From shortest split-line algorithm:

And here is Maryland.



PS: A side-by-side comparison of these images would be much more feasible if:

  • all the current district maps were available in a variety of standard sizes as images, not pdfs
  • all of the algorithm generated maps were available a variety of standard sizes

Hopefully the folks who developed the split-line algorithm will provide this.

Some examples of spectacularly gerrymandered districts


5 responses to “More on the shortest split-line algorithm with pretty pictures

  1. Hey, would it be possible to get a look at what California would look like? Heavens knows we’re gerrymandered enough

  2. I hope this project will also allow the projection of changes in the number of districts when an apportionment is increased or decreased.

  3. Nick:

    Here is CA:

    D. Frank Robinson:

    The algorithm certainly allows for redistricting to occur based on an arbitrary number of districts.

    One issue with the algorithm is that as the number of disricts change, the districts will potentially change radically. Some may view this as a problem. In particular incumbents whose district changes to hold a very different population will likely be less than thrilled. (especially if there is a residency requirement that dictates which district someone can run for based on where they live.)

  4. Good. Make the incumbents work. No one should have a lock on an election unless they truely have done something to earn the vote.

  5. Lord. This thing has been up 6 years and I’ve not had a chance to reply. Oh well. The gist of what I wanted to say was:

    a) yes, states should get whole numbers of seats
    b) hell, no, shortest-split-line is not the best method. Look what it does to (I think) Colorado. Otherwise, look how it splits cities in the US without just cause. It looks neat, but it’s the equivalent of taking a sledgehammer to crack a walnut.

    Look instead at the UK model. They have a non-partisan redistricting commission which divides the counties according to various rules. Happily, there is no distinction between the rules, and the commission are allowed to overrule rule (a) in favour of (b), or in another case overrule (b) in favour of (a), or any other permutation, so long as the result isn’t a gerrymander. And like all these things, it’s like an elephant, it can’t be written into law: you know a gerrymander when you see one, but you can’t describe one accurately into a volume of law.

    The split-line algorithm isn’t the answer. But a fair independent redistricting commission, with no prospect of legislative override, is.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s