Reverse Gerrymandering Part 1

With the Supreme Court’s recent decision that non-partisan redistricting efforts are indeed constitutional, the call to adopt non-partisan redistricting commissions in other states has grown. Two years ago we assessed how non-partisan redistricting efforts were being oversold as a way to promote closer elections and more moderate candidates. While we certainly support non-partisan redistricting over the current model, we need to understand what the likely outcomes are of these efforts. As we discussed in the previous memo, people throughout the country have been self-sorting so that Democrats typically live near other Democrats and Republicans typically live near other Republicans. This has made it so creating compact districts actually results in heavily partisan districts that end up electing even more partisan candidates.

All of this is not to say that we are opposed to independent commissions drawing the lines. Independent commissions prevent party leaders and representatives from picking their voters to protect their jobs, and they give voters more confidence in the legitimacy of their representatives and the districts. We support independent redistricting commissions and the benefits of their theoretical goals, but in most cases, commissions’ intended aims do not match up to the end result. If the goal is to create competitive districts, non-partisan commissions will not solve the problem.

In most articles about redistricting, authors focus on ensuring either geographically compact districts orrepresentative ones, where the percentage of red and blue districts reflect the red and blue breakdown of the state. This approach, in which making maps that have 40% Democratic districts because 40% of the state votes Democratic, is one way of looking at gerrymandering, but it does not solve the problem of gridlock in Washington and state capitals. Instead, this would create more partisanship by making safe but “representative districts,” essentially closing the door to moderate candidates. Instead of the representative or geographical approach, we looked to create maps that would result in the highest number of competitive districts humanly possible with the intended outcome to be more moderate candidates.

With that in mind, we want to show just how bizarre some districts would have to look in order to produce competitive districts throughout states. Utilizing a web tool, we have created a series of new Congressional district maps that result in competitive districts that would allow moderate candidates to win. We aimed to make as many districts as possible no worse than a 53-47 partisan split. While looking at all 50 states would be fun, for the purposes of this piece we assess fictitious districts in Wisconsin, Virginia, North Carolina, Minnesota, New Jersey, Illinois, Maryland, Missouri, and Louisiana.

As you will see, if you think some districts look ridiculous on paper now, making congressional districts competitive can be even worse. Clearly most of these districts do not meet the compact goal for districts, and would never be approved by a panel. This being said, if changing the way Washington does business, we as a country, or as states more accurately, will need to consider other options at our disposal such as top-two primaries (also known as Jungle Primaries) if we want to put a premium on results over partisanship.


Wisconsin illustrates what will become a common theme among all the states in trying to make as many competitive districts as possible. The major cities, in this case Milwaukee and Madison, need to split their concentrated votes among several districts. Milwaukee is split between the purple, dark blue, and green districts, and Madison is broken up between yellow, purple, and red.

The Wisconsin map also illustrates the other problem with trying to make all similarly competitive districts that some districts simply cannot end up competitive. In Wisconsin’s case, the dark blue, green, purple, yellow, red, and blue green districts are almost all 50-49 districts, but the light purple district will be a romp for Republicans every election, and gray also favors Republicans.

The Wisconsin map would clearly never survive a legal challenge if the districts were drawn this way. The red and blue-green districts certainly ignore the idea of being compact, but these districts would make it much harder to be an extremist candidate.


Virginia doesn’t have as many egregious districts as Wisconsin, but nearly half of the eleven districts have tentacles reaching into Northern Virginia to split the heavily Democratic area across as many districts as possible. The blue, light purple, pink, blue-green, and yellow-green districts all stretch out to balance out the more rural areas in the rest of the district with plenty of liberal Democrats. Only the light blue district in the southwest portion of the state is unable to achieve a 53-47 party split.

North Carolina

North Carolina was lambasted for its most recent redistricting effort as the state went from being gerrymandered in favor of Democrats to being gerrymandered in favor of Republicans. The blue and purple districts in our map pay homage to the state’s proud history of gerrymandering. All of the resulting districts are at worst 54-46 and most of the districts are much closer. While these districts may not be visually appealing, but they are certainly more competitive than the state’s recent efforts.

The light purple district in the eastern part of the state, would win the prize for least compact district, and clearly would never pass legal muster, but this is what it would take to make this portion of the state competitive.


Our Minnesota map is the most radical departure from the state’s actual maps as all eight of the districts enter the Minneapolis, St. Paul metro area, and fan out to cover the state. However, even after breaking up the state’s biggest reserve of Democrats into all eight districts, each of the districts ends up slightly favoring the Democrats, but all are competitive except for the blue district which also includes the state’s Iron Range Democrats.

The liberal states

 While most of the states in this analysis are swing states, we wanted to look at some of the states that have reputations for gerrymandering their districts. MarylandNew Jersey, and Illinois received a lot of criticism for their most recent maps favoring Democrats. We wanted to put these states to the test to see how different the maps would have to look to be balanced. However, the problem is that these states are so liberal in their voting patterns that there simply aren’t enough Republican voters to make all of the Districts even, no matter how bizarrely the districts are drawn.

In New Jersey, the state leans so heavily Democratic that we had to expand the definition of competitive to 55-45, and the light blue and light purple districts still exceed that threshold. The New York suburbs had to be split among six districts, and they were still too stuffed full of Democrats. Only the two districts that stretch to the other side of the state, the yellow and gray districts manage to get to 52-48 in favor of the Democrats. Northern New Jersey is simply too Democratic to produce even districts.

The Maryland map is fairly boring compared to most of the other maps because the state voted so liberally in recent elections that the closest to competitive is the blue district containing all of the conservative western portion of the state. This district would still be 55-45 in favor of Democrats and it only gets worse for Republicans in the other districts.

In honor of the Illinois gerrymander, our Illinois map is also one of the more blatant gerrymanders. The blue and green districts stretch from Chicago in a thin band down to the Southern portion of the state. Meanwhile, the orange district wraps around both of those districts. Once again, Chicago had to be split up as many ways as possible but it still wasn’t possible to spread out all of the concentrated Democratic votes in the north evenly with the rest of the state.

Conservative States

On the Republican side, both Missouri and Louisiana are prime examples of conservative states with heavy gerrymandering, but unlike the heavily liberal states, both were a bit easier to make balanced districts. The liberal states tended to have massively concentrated pools of Democrats which were nearly impossible to break up, whereas the conservative states still have the liberal pockets but they are surrounded by areas ranging from conservative to very conservative. This pattern allowed for creative map drawing to reach balanced districts.

In Louisiana, the six new districts are virtually identical in partisan splits, just a hair closer than 55-45 in favor of Republicans. Again, half of the districts share in parts of New Orleans, and virtually every other major city is broken into multiple districts.

For Missouri, the districts ended up being more competitive than Louisiana and the liberal states we looked at before. With the exception of the blue district, which heavily favors Democrats, every other district is a 53-47 split or better. Amazingly, each of the eight districts ends up including either a portion of the St. Louis or Kansas City metro areas, all taking a sliver of the metro area and then filling up with as much of the rest of the state as they can.


Clearly this is not the direction congressional maps are going to head in. The requirement for compact districts that are at least somewhat similar would result in most of these maps being laughed out of the room. No one could propose the North Carolina, Minnesota or Missouri maps with a straight face.

It most likely wouldn’t even be a worthwhile end result. Some of the states, such as Maryland or Louisiana, are so one-sidedly partisan that adopting these maps would simply ensure that the other party got no representation in Congress. To make matters worse, the districts would still be one-sided enough that the winning candidates could be just as extreme as the ones we hoped to move past.

However this is a useful thought experiment to show that even with the best of intentions to create maps that favor moderate candidates, most states simply can’t get there. Non-partisan redistricting may eliminate some of the most outlandish gerrymandering, but people’s own choices about where to live have already guaranteed that nearly every district is either solidly Democratic or solidly Republican. If people are sick of the partisanship in DC, they need to get comfortable with it because it is not bound to change anytime soon.

For a downloadable pdf of the whitepaper, click here.