Gerrymeander: A New ‘Redistricting Game’

This is a guest post from James Prince, a solo game developer in Chicago, who built a free successor to The ReDistricting Game after the original stopped running with Flash. He runs Subastra, a small game design studio.

This post was drafted with Claude, an AI assistant, working from the author’s notes and direction, and was reviewed and edited by the author, with some additional edits by the ALPS team.

The game that went dark

In 2013 Amanda wrote here about The ReDistricting Game, USC’s Flash simulation in which students redrew a fictional state’s districts. It was a good tool. Then Adobe retired Flash at the end of 2020, the game stopped running, and its domain no longer resolves.

The links are still on course pages, teaching libraries and parent guides, pointing at nothing. That would matter less if redistricting no longer mattered. This isn’t the case: by the count of Loyola Law School’s All About Redistricting, five states redrew their congressional maps mid-decade in 2025 and 2026 without a court ordering them to.

I’m a solo game developer in Chicago, and I built a replacement called Gerrymeandering. This post covers what it does, why it’s built the way it is, and how I’d use it in one class period. The game can be found at: gerrymeandering.subastra.dev or on the App Store for Iphone users.

What students do

Students pick or are assigned a state to redistrict. In this case, we are assigned Iowa.

The big change from the original game is that the state is real. Students draw on actual Census TIGER/Line geography, with 2024 presidential vote returns shown as dots across the map.

  1. They’re given a state and a side, and told to win as many seats for that side as they can.
  2. They cut the state into contiguous districts with equal populations. The game won’t accept a map that breaks either rule.
In this example, students draw districts in Iowa. Two districts are drawn for Blue, one around Des Moines and one along the Cedar Rapids-Iowa City-Davenport cooridor.
  1. They submit. The districts the 119th Congress actually uses then draw themselves over the student’s map. Same state, same voters, two maps.
Here we can see the overllay of the actual districts, in the yellow outlines, superimposed over the districts drawn by the student.
  1. They’re scored against the enacted plan on three measures: seats won for their side, Polsby-Popper compactness, and the efficiency gap.
In the results panel, we can see that Blue has 43% of the vote, wins 2 of the 4 seats on this map, but 0 out of 4 on the ‘real’ map.

All 50 states are available, plus 40 tutorial levels on fictional states that build up gradually. A daily mode gives everyone in the world the same state and the same side each day. That makes it easy to set as homework: everyone plays on identical ground, and the comparisons the next morning are fair.

Read on for more on how to use this in your classroom.

Teaching it backwards

Redistricting is usually taught definitions first. Here is packing, here is cracking, here is the grid of fifty squares, here is the efficiency gap formula. Then students apply it. The game runs that order in reverse.

The task is simply to win seats. On a real state, the only way to do that is to put as many of the other side’s voters as you can into a few districts they win by huge margins, and spread the rest thinly across districts they narrowly lose. Students complete the task, and by the time you name packing and cracking, they have already done both.

The efficiency gap works the same way. It counts wasted votes: every vote cast for a losing candidate, plus every vote for a winner beyond the bare majority needed. The gap is the difference between the two sides’ wasted votes, as a share of all votes cast.

Efficiency gap = (wasted votes for side A − wasted votes for side B) ÷ total votes cast

In a lecture that’s a formula to memorize. Here it’s a number on the student’s own results screen, produced by a map they drew.

Compactness is the counterweight. The map that wins the most seats is usually not the most compact one, and Polsby-Popper puts a number on that trade-off. Students feel the pull between the two scores before anyone explains why courts and commissions argue about shape.

Then the overlay lands. Some students will find the legislature’s map looks a lot like the one they drew to win. Some will find the enacted plan beat them on seats. Either result is the discussion.

One class period, 50 minutes

So far the game has been tested with playtesters, not in a classroom, so treat this as a starting point rather than a proven plan. It assumes no reading beforehand and no vocabulary up front.

  1. Set up, 5 minutes. Put students in pairs. Give every pair the same state, ideally your own, and split the room so half play each side. Tell them only the goal: win as many seats as you can.
  2. Draw, 15 minutes. Leave them alone. Pairs argue, and that’s useful.
  3. Submit, 5 minutes. Each pair writes down three numbers: seats won, compactness, efficiency gap. Then they look at the enacted map over theirs.
  4. Name it, 10 minutes. Put two maps on the projector, one from each side. Point to a district that’s 80 percent one party and ask what it’s for. That’s packing. Find a city split four ways. That’s cracking. Then show the efficiency gap as the number they already wrote down.
  5. Discuss, 10 minutes. Use the questions below.
  6. Exit ticket, 5 minutes. One sentence: what would a rule have to say to stop what you just did?

The exit ticket is the point of the lesson. I’d expect it to be harder to write than students think, and that difficulty is a good way into why this has been so hard for courts.

Four discussion questions

Use these to debrief the exercise. These work for a section of thirty or a lecture of three hundred.

  1. Is either map fair? Your map and the enacted one turned the same votes into different seat counts. What would “fair” have to mean before you could answer?
  2. Which number should a court care about? Your best-scoring map and your most compact map were probably different maps. In Rucho v. Common Cause (2019), the Supreme Court held that federal courts can’t decide partisan gerrymandering claims. So if not courts, who enforces any of these numbers?
  3. What did you need to know? You drew a winning map using nothing but last year’s presidential vote. Real mapmakers have far better data. Should they be allowed to use it?
  4. Can you tell who drew it? Michigan’s map is drawn by an independent citizens commission; most states’ maps are drawn by legislators. Draw one of each. Can you tell which is which from the overlay alone?

What it doesn’t do

  • The overlay is the 119th Congress map. Several states, including North Carolina and Texas, adopted new maps for 2026. In those states the overlay is the previous plan. That makes a good discussion, but say so up front.
  • It uses presidential votes as a stand-in. People don’t vote the same way all the way down the ballot. Real mapmakers model that; the game doesn’t.
  • It leaves out most of the law. There’s no Voting Rights Act requirement, no communities of interest, no protection for incumbents. Students are optimizing a simpler problem than a real legislature faces.
  • It asks students to gerrymander. That’s deliberate, and some instructors will find it uncomfortable. What keeps it nonpartisan is that every state can be played from either side and both are scored the same way. The debrief ends up being about what the geometry allows, not about which party to blame.

Using it

Gerrymeandering is free, with no purchases. There’s no account and no login, and it collects no student data, so there’s nothing for an institution to review or sign. It runs in any browser, Chromebooks included, and there’s a free iPhone version, which shows ads. One state takes about ten minutes.

It’s a one-person project. If you use it in a class, I’d like to hear what worked and what didn’t, especially what didn’t. You can reach me at hello@subastra.dev.

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.