Hands-Off Mapping¶

A Primer Series Article of the Congressional Elections Modernization Act (CEMA)¶

Prepared by Albert Ramos for The American Policy Architecture Institute

Every ten years, Americans watch the same fight play out. The census is taken. Politicians, commissions, and consultants start drawing congressional maps. Lawsuits follow. Courts step in. Maps are redrawn -- sometimes more than once. Elections are held under boundaries that may be thrown out next year. The process eats up years, hundreds of millions of dollars, and whatever public trust is left in the fairness of the system. Then the next census arrives, and the whole cycle starts over.

The Congressional Elections Modernization Act (CEMA) proposes to end this cycle by taking human judgment out of congressional map-drawing entirely. Under the Act's Hands-Off Mapping framework, all congressional boundaries are drawn by a math formula that takes in census population data and a seat count and produces one -- and only one -- map. The formula follows a fixed set of rules: same numbers in, same map out, every time. No mapmaker, no commission, no judgment calls. Anyone with the same census data can run the formula themselves and verify they get the same result.

Many reformers have pushed hard for independent commissions, and the goal is a good one -- get the map-drawing out of partisan hands. Hands-Off Mapping wants the same thing and takes a further step. Even a fair-minded commission still makes judgment calls, and judgment calls are the opening that manipulation and lawsuits slip through. Remove the judgment, and the opening closes. The formula is overseen and verified by the Electoral Science Office, a national elections body the Act creates to certify the formula, publish its code, and let anyone check the results. And it works because the Act pairs it with two other changes: a percentage-based voting method in larger, multi-seat districts, and a bigger House to make those districts possible.

What Gerrymandering Actually Costs¶

The case for Hands-Off Mapping starts with a simple observation: every person who draws a district boundary is making a choice. Even boards and panels set up specifically to draw fair maps have to make judgment calls -- which communities should be kept together, how compact a district should be, whether the map should favor competitive races. These are real questions with no objectively right answers. That fact is exactly what makes the current system so fragile, in two concrete ways.

The first is manipulation. Modern mapping software lets line-drawers predict election outcomes block by block, and the temptation to exploit that capability belongs to both parties. In the 2020 redistricting cycle alone, more than 225 lawsuits were filed challenging congressional or state legislative maps across at least 30 states. Individual states routinely spend millions of dollars per cycle drawing and defending maps that courts may strike down anyway -- and those are only the costs that can be tracked.

Gerrymandering

The practice of drawing district lines to favor one group over others. The term dates to 1812, combining the salamander-like shape of a district drawn under Massachusetts Governor Elbridge Gerry with his surname: Gerry + salamander = gerrymander.

The second is competition -- or the loss of it. Under the maps drawn after the 2020 census, the number of genuinely competitive congressional districts fell to the lowest level in at least fifty-two years. Heading into 2026, about eighteen of the 435 House seats -- roughly 4 percent of the chamber -- were rated as true toss-ups. Seats decided in advance don't just insulate incumbents from accountability. They push representatives toward their party's most committed voters and punish those who try to work across the aisle. The polarization that voters say they hate is in part a structural product of the maps they live under.

Safe Seat

A district where one party's victory is so certain that the general election is effectively decided before it happens. The competition that would normally hold a representative accountable never materializes, because the outcome was decided when the lines were drawn.

Hands-Off Mapping eliminates both problems by eliminating the judgment. The formula has no political preferences. It can't be lobbied, pressured, or negotiated with. It produces the same map from the same numbers every time. There is nothing left to manipulate and nothing left to sue over.

What Researchers Have Found -- and Why It Changes¶

Researchers at major universities have built and tested a range of methods for generating district maps without human input. Some use simple geometric rules. Others generate thousands of possible maps to serve as baselines for evaluating whether a human-drawn map is fair. The tools work at scale and have been used in court cases across the country. The most important finding is straightforward: formula-based map-drawing is technically feasible when the rules are clearly defined.

But the research community has not recommended handing the entire process over to a formula. Researchers consistently present their tools as measuring sticks -- ways to test whether a human-drawn map is an outlier -- rather than as replacements for human judgment. The reason: a formula that sees only population numbers and geography cannot identify communities that share common interests. Under a winner-take-all system, a community split across two single-member districts loses all representation. The researchers weren't wrong. That risk is real -- within the system they were evaluating.

Two other findings explain the caution. First, finding the mathematically "best" map is a problem so complex that no formula can fully solve it -- there are simply too many possible ways to divide a state into districts. Any sophisticated formula designed to chase an optimal result has to make judgment calls about what to prioritize, quietly reintroducing the human discretion the formula was supposed to eliminate. Simple methods sidestep this by accepting narrow goals and producing a single, verifiable result. Second, a sophisticated formula with adjustable settings can be quietly tuned to favor one party while appearing neutral -- a concern the Act addresses through transparency requirements covered below.

These objections are genuine. Their objections apply to formula-based mapping within a winner-take-all system. What changes when you change that system is the subject of the next section.

Why These Changes Work Together¶

The case for Hands-Off Mapping is not a case for formulas on their own -- and not a case for adopting a new technology simply because it exists. A formula dropped into the current winner-take-all system, without the surrounding changes, could produce outcomes worse than what it replaced. Several of the Act's changes bear directly on the map-drawing problem, and together they resolve the researchers' objections not by answering them within the existing framework, but by changing the framework.

House Expansion matters because the current House has 435 members for roughly 330 million people -- a ratio frozen since 1929. The Act expands the House to approximately 720 seats. A larger House creates enough seats to build multi-seat districts with real depth -- five, six, or seven representatives per district instead of one. The more seats a district elects, the smaller the share of the vote a cohesive group needs to win one of them. At five seats, a group needs roughly twenty percent. At seven, roughly fourteen percent.

STAR Voting ensures that seats in those larger districts are allocated to match how people actually voted. A community making up twenty percent of a five-seat district elects roughly one representative -- not because a mapmaker drew them a favorable district, but because the voting math distributes seats in proportion to voter support. Competitive races are built in. Fair representation for every cohesive group follows from the arithmetic rather than depending on a mapmaker's good intentions. But someone still has to draw the district boundaries -- and under human-drawn maps, that step remains an opening for manipulation.

Hands-Off Mapping closes that opening. The formula has no political preferences and no adjustable settings. It cannot be lobbied or negotiated with. And because percentage-based voting makes boundary placement irrelevant to representational outcomes, the formula's inability to see communities is not a flaw to work around -- it is what guarantees that no one can embed preferences into the process.

Each one covers the others' weaknesses. Hands-Off Mapping removes the human discretion that makes boundary-drawing a target for manipulation -- but a formula alone cannot guarantee fair representation under winner-take-all rules. Percentage-based voting ensures that fair representation flows from voter preferences rather than boundary placement -- but someone still has to draw the lines. House expansion makes districts large enough for percentage-based voting to deliver proportional representation reliably -- but without Hands-Off Mapping, those larger districts remain vulnerable to the same manipulation the current system enables. Together, they produce a map-drawing process that cannot be rigged, cannot be meaningfully challenged in court, and delivers fair representation for every cohesive group regardless of where the lines fall.

Addressing Concerns¶

"A formula can't see neighborhoods and communities"¶

This is the concern heard most often. Real communities -- neighborhoods, cultural enclaves, areas that share infrastructure or environmental concerns -- have a stake in who represents them. A formula that processes only population numbers and state geography cannot identify those communities. If it splits one across two districts, that community loses its voice.

The factual premise is correct. The formula cannot see neighborhoods. But the conclusion -- that this causes real harm -- depends on a voting system the Act replaces. Recall that under a percentage-based voting method in larger, multi-seat districts, a community doesn't need to be neatly packaged inside a single district to earn representation. A group making up twenty percent of a district's voters earns roughly one seat in five -- not because anyone drew a line on their behalf, but because the math works in their favor. The formula's blindness to community geography stops being a defect once fair representation no longer depends on where the lines fall.

It is also worth noting that "community of interest" is harder to define in practice than it sounds in theory. Every state defines it differently; some definitions are broad enough to include political affiliation. Redistricting panels typically identify communities through public testimony and self-identification -- participatory, but not consistent from one state to the next. Even data-driven approaches still require someone to decide what to prioritize when communities overlap, which brings human judgment back into the process and reopens the door to litigation. A formula that sidesteps this definitional tangle entirely, while percentage-based voting protects community representation through the math, is not ignoring communities. It is protecting them without having to define them.

"What about fair representation for minority communities?"¶

The natural worry is that a map-drawing process with no awareness of community makeup will weaken the communities most in need of protection. It is a reasonable concern, and it deserves a direct answer.

Under the Act, the protection does not have to come from the boundary. A bigger House lowers the share of the vote a community needs to win a seat. Percentage-based voting turns that share into actual representation. And a formula nobody controls means no one can crack a community apart or pack it into a single district. A community earns representation by how its members vote, not by whether a mapmaker chose to draw a line on their behalf.

There is also a deeper point about the kind of protection the Act provides. The protections most people associate with fair representation for minority communities are legal protections -- rights enforced by courts. Legal protections can be narrowed. Courts change, and a protection won one decade can be weakened the next. The Supreme Court has recently done exactly that to a major protection against diluting minority votes. Structural protection works differently. When representation comes from the math -- smaller thresholds, seats allocated by vote share, lines drawn by formula -- it doesn't depend on winning a lawsuit. There is no legal test for a court to narrow, because the guarantee is built into how the seats are filled.

"Can't someone just rig the formula?"¶

This risk is real, and the research community was right to flag it. A complicated formula with adjustable settings can be quietly tuned to favor one party while appearing neutral on the surface. But that concern applies to formulas that are hard to follow and easy to manipulate behind the scenes -- not to a straightforward method whose entire logic can be explained in a paragraph and checked by anyone.

The Act's default formula divides a state's population using straight lines, accepting only census data and state boundaries as inputs. There are no adjustable settings. There is nothing to tune. The formula produces one map, and anyone can verify it. If the formula is ever replaced with something more sophisticated, that change must clear a high bar: a structured public approval process overseen by the Electoral Science Office, a public comment period, congressional review with the power to block the change, published source code, independent verification by two independent organizations, and the right of any member of the public to run the formula and check the results. The concern about rigged formulas applies to black boxes. CEMA doesn't build black boxes -- it builds transparent ones.

Keeping the Formula Honest¶

A formula that can be tampered with is no better than a mapmaker who can be lobbied. The Act builds layered protections designed to hold up in a hostile political environment, not just a cooperative one.

The formula runs directly from census data, with no official required to authorize or approve the output. Officials who certify and maintain it are protected from politically motivated removal during active map-drawing periods; recess appointments are prohibited; any executive attempt to influence a map-drawing determination must be documented and reported to an independent oversight body. A guaranteed budget prevents the office from being quietly starved into ineffectiveness. The full source code is published and independently verified by two independent organizations, giving any member of the public the means to download the census data and confirm the map themselves. No other entity may draw federal congressional maps, attach conditions to the oversight office's funding, or fold the office into another agency where it could be brought under political control.

Capturing the appointments process doesn't help if existing officials can't be removed during redistricting. Cutting the budget doesn't stop the formula from running. Pressuring officials doesn't help if those contacts must be reported. There is no partisan back channel into the process -- just census data, a published formula, and a map anyone can verify.

What This Does and Does Not Reach¶

The Act governs congressional elections. Its authority comes from the Constitution's grant to Congress over the manner of electing Representatives, and it reaches no further than that. The formula draws congressional districts; it does not touch the maps a state draws for its own legislature, and the presidential election is a separate constitutional question -- one addressed in a forthcoming paper in this series.

The same machinery the Act builds for congressional map-drawing -- the certified formula, the published source code, the independent verification, the public replication tools -- is available to any state that wants to adopt it for its own legislative maps. The Act extends the offer; it does not compel acceptance. What a state cannot do, under the Act, is carry a manipulated map into the People's House. The composition of Congress is a national concern, and the Constitution commits the manner of its election to national resolution. The Act exercises that authority to take congressional map-drawing out of contention entirely -- not by trusting any state to draw fair lines, and not by trusting any court to police unfair ones, but by removing the discretion that makes the question contestable in the first place.

Conclusion¶

The case for Hands-Off Mapping is not that formulas are perfect mapmakers. It is that human judgment in map-drawing is the vulnerability -- the opening that enables manipulation, guarantees litigation, and systematically destroys electoral competition.

Hands-Off Mapping closes that opening. But the full case depends on what surrounds it -- the House Expansion that creates districts large enough for fair representation, and the STAR Voting that delivers it at the ballot box. The other papers in this series address what this one does not: the Electoral Science Office that provides the institutional backbone the formula requires, and Full-Field Elections, which ensures that the widest possible field of candidates reaches every voter -- so that the open map and the open ballot arrive together.

No human touches a boundary. No judgment call invites a lawsuit. No line drawn in a back room determines who wins before anyone votes.

Other articles in the CEMA Primer Series:

Works Cited¶

Altman, Micah. 1997. "Is Automation the Answer: The Computational Complexity of Automated Redistricting." Rutgers Computer & Technology Law Journal 23(1): 81-142.

Brennan Center for Justice. 2010. "Communities of Interest." Redistricting series monograph.

Brennan Center for Justice. 2022. "Gerrymandering Competitive Districts to Near Extinction."

Brennan Center for Justice. 2025. "Redistricting Litigation Roundup."

Chen, Sandra J., Samuel S.-H. Wang, Bernard Grofman, Richard F. Ober Jr., Kyle T. Barnes, and Jonathan R. Cervas. 2022. "Turning Communities of Interest into a Rigorous Standard for Fair Districting." Stanford Journal of Civil Rights and Civil Liberties 18(1): 101-189.

Cook Political Report. "2026 House Race Ratings." https://www.cookpolitical.com/ratings/house-race-ratings.

Levin, Harry A. 2023. "The Rise of the Hal-mander: Is Gerrymandering by Algorithm the Next Frontier of Partisan Gerrymandering?" Georgetown Law Journal 111(4): 891.

Levin, Harry A., and Sorelle A. Friedler. 2019. "Automated Congressional Redistricting." ACM Journal of Experimental Algorithmics 24: 1.10:1-1.10:24. DOI: 10.1145/3316513.

Levitt, Justin. All About Redistricting. Loyola Law School. https://redistricting.lls.edu.

Louisiana v. Callais, 608 U.S. ___ (2026).

Makse, Todd. 2012. "Defining Communities of Interest in Redistricting Through Initiative Voting." Election Law Journal 11(4): 503-517. DOI: 10.1089/elj.2011.0144.

Najt, Elle (Lorenzo), Daryl DeFord, and Justin Solomon. 2019. "Complexity and Geometry of Sampling Connected Graph Partitions." arXiv:1908.08881. https://arxiv.org/abs/1908.08881. [Peer-reviewed companion: Najt, DeFord, and Solomon, "Empirical Sampling of Connected Graph Partitions for Redistricting," Physical Review E 104(6), 064130 (2021). DOI: 10.1103/PhysRevE.104.064130.]

Revision history available in the raw file.

📥 Download this document (opens on GitHub -- click the ⬇ download button)