In Wisconsin’s 2012 state assembly elections, Republican candidates received 48.6% of the statewide vote but won 60.6% of the seats[s]. By 2018, Democrats captured 54% of the popular vote yet Republicans retained a 63-seat majority[s]. This is not a quirk of geography or an accident of demographics. This is gerrymandering mathematics at work: the systematic geometric manipulation of electoral boundaries to predetermine election outcomes.
The tools exist to detect and quantify this manipulation with scientific precision. Federal courts have acknowledged that partisan gerrymandering is “incompatible with democratic principles.”[s] And yet, in 2019, the Supreme Court declared itself powerless to act. The mathematics is clear. The evidence is overwhelming. The judiciary has simply chosen to look away.
Gerrymandering Mathematics: The Numbers That Expose the Fraud
The core of gerrymandering mathematics lies in a concept called “wasted votes.” When one party’s supporters are strategically scattered across districts where they lose by narrow margins (crackingA gerrymandering tactic that splits opposition voters across multiple districts, diluting their influence so they form losing minorities in each.) or concentrated into districts where they win by huge margins (packingA gerrymandering tactic that concentrates opposition voters into one district, wasting their votes on landslide wins that don't translate into additional seats.), their votes become inefficient[s]. The efficiency gapA statistical measure of partisan gerrymandering calculated by comparing wasted votes between parties. A gap above 7% suggests the map entrenches one party's advantage., developed in 2014 by University of Chicago law professor Nicholas Stephanopoulos and political scientist Eric McGhee, measures this disparity[s].
The calculation is straightforward: subtract one party’s wasted votes from the other’s, then divide by total votes cast. An efficiency gap above 7% indicates a map so skewed that it likely locks in majority control for a decade[s]. Wisconsin’s 2012 map registered an efficiency gap of 13%. By 2018, it had grown to 15%[s].
North Carolina’s 2012-2014 congressional plan produced an efficiency gap of 20.3% in favor of Republicans[s]. Despite Republicans receiving just 53% of the statewide vote, they secured 10 of 13 congressional seats[s]. State Representative David Lewis said the quiet part aloud during redistrictingThe process of redrawing electoral district boundaries, typically after each census. When done to favor one party, it becomes gerrymandering.: “I propose that we draw the maps to give a partisan advantage to 10 Republicans and three Democrats, because I do not believe it’s possible to draw a map with 11 Republicans and two Democrats.”[s]
The Scale of the Problem
In Texas, Democrats hold only 13 of 38 House seats (34%) despite consistently receiving 46-48% of the statewide vote[s]. A fair map would give them 18 seats. Florida transformed a 16-11 Republican edge into a staggering 20-8 advantage through aggressive redistricting[s].
North Carolina, with an evenly divided electorate, elected 7 Democrats and 7 Republicans under a court-drawn map in 2022. After the state supreme court’s conservative majority dismantled anti-gerrymandering protections, the new map could produce 11 Republicans and just 3 Democrats[s].
These numbers represent millions of voters whose representation has been mathematically stolen. Gerrymandering mathematics proves this theft with precision. When your vote is worth less because of where you live, the fundamental promise of democracy collapses.
The Federal Courts’ Surrender
In Rucho v. Common Cause (2019), the Supreme Court ruled 5-4 that federal courts cannot review partisan gerrymandering claims. Chief Justice Roberts wrote that while such practices may be “incompatible with democratic principles,” they present “political questions beyond the reach of the federal courts.”[s]
Justice Elena Kagan’s dissent was scathing: “Of all times to abandon the Court’s duty to declare the law, this was not the one. The practices challenged in these cases imperil our system of government. Part of the Court’s role in that system is to defend its foundations. None is more important than free and fair elections.”[s]
The ruling gave state legislators a green light. A statistical analysis found that some of the worst offenders from 2010, including Florida, Georgia, Illinois, Indiana, North Carolina, Texas, and Wisconsin, scored even worse on partisan gerrymandering measures after 2020[s].
What Needs to Change
Gerrymandering mathematics provides the diagnostic tools. The efficiency gap, along with other metrics, can identify manipulation with scientific rigor. State courts have stepped up in Pennsylvania, Ohio, New York, and other states to enforce limits on partisan gerrymandering where federal courts will not.
Independent redistricting commissions, as adopted in Michigan, Virginia, and Colorado, remove the fox-guarding-the-henhouse problem. Congressional legislation like the Freedom to Vote Act would establish federal standards. Some progress has occurred: overall gerrymandering effects dropped from 23 extra Republican seats in 2012 to just 3 in recent analyses[s].
But “better than 2012” is a low bar when individual states like Texas, Florida, and North Carolina have doubled down on manipulation. Until federal courts recognize that mathematical proof of electoral theft demands judicial remedy, or until Congress acts, the survival of representative democracy depends on state-by-state battles that most voters don’t even know are being fought.
Gerrymandering Mathematics: Computational Detection Methods
The emergence of gerrymandering mathematics as a rigorous field traces to a fundamental challenge: how do you prove a map is unfair when every redistrictingThe process of redrawing electoral district boundaries, typically after each census. When done to favor one party, it becomes gerrymandering. involves inherent tradeoffs? The answer lies in ensemble analysis, comparing a proposed map against thousands or millions of algorithmically generated alternatives that satisfy the same legal constraints.
The efficiency gapA statistical measure of partisan gerrymandering calculated by comparing wasted votes between parties. A gap above 7% suggests the map entrenches one party's advantage. metric, developed by Nicholas Stephanopoulos and Eric McGhee in 2014, provides one quantitative standard[s]. It measures wasted votes: all ballots cast for losing candidates, plus all ballots for winners beyond the 50%+1 threshold needed to win. When one party systematically wastes fewer votes through strategic district drawing, the efficiency gap captures this asymmetry. A gap exceeding 7% suggests entrenchment: the advantaged party will likely retain control regardless of statewide vote shifts[s].
But the efficiency gap alone cannot distinguish intentional manipulation from natural geographic clustering. This is where Markov chain Monte Carlo (MCMC) methods enter gerrymandering mathematics. Researchers including Kosuke Imai at Harvard and Jonathan Mattingly at Duke apply MCMC to generate massive ensembles of neutral redistricting plans[s].
The Combinatorial Explosion Problem
The mathematical challenge is staggering. A simple 4×4 grid divided into four contiguous districts of four squares each has 117 valid configurations. A 6×6 grid: 451,206 possibilities. A 9×9 grid with nine districts: over 700 trillion configurations[s]. Real states are vastly more complex: North Carolina has over 2,500 precincts, Pennsylvania over 9,000, and official maps use census blocks, of which Alabama alone has 185,976[s].
Finding an optimal redistricting falls into the class of NP-hard problems, computationally intractable for exact solutions[s]. MCMC sidesteps this by sampling from the distribution of valid maps. The ReCom algorithm, developed by the Metric Geometry and Gerrymandering Group (MGGG) under mathematician Moon Duchin, merges adjacent districts, generates random spanning trees across the combined precincts, and splits them in a statistically principled way that produces representative samples quickly[s].
Duchin’s group has provided expert testimony in redistricting cases across Pennsylvania, Alabama, Virginia, and elsewhere[s]. In 2022, when a federal panel threw out Alabama’s congressional maps, Duchin drew replacement maps placing Mobile and Montgomery together to create a second Black-majority district[s].
Quantifying the Manipulation
The outlier test asks a simple question: what fraction of neutral maps are less extreme than the proposed plan?[s] When researchers applied their algorithm to Maryland’s 2011 map, 99.79% of 250 million generated alternatives showed less Democratic advantage[s]. The official map was a statistical outlier in the tail of a massive distribution.
Wisconsin’s 2012 map registered an efficiency gap of 13% in the Gill v. Whitford case, the first federal ruling to strike down redistricting for partisan bias[s]. The state’s own mapmakers projected that Republicans could “expect to win 59 Assembly seats, with 38 safe Republican seats” under the plan[s].
North Carolina’s efficiency gap hit 20.3% for Republican candidates under its 2012-2014 plan[s]. Texas currently hands Republicans 25 House seats, with 21 of them in districts Donald Trump carried by 15 or more points in 2020, creating an electoral firewall: Democrats hold 13 of 38 seats despite 46-48% statewide support[s].
The Rucho Doctrine and Its Consequences
The Supreme Court’s 2019 ruling in Rucho v. Common Cause declared partisan gerrymandering a “political question” beyond federal jurisdiction[s]. Chief Justice Roberts acknowledged the practices may be “incompatible with democratic principles” but held that no judicially manageable standard existed.
This reasoning collapses under scrutiny. Eric Lander, later Biden’s science advisor, argued in an amicus brief that the outlier test provides “a straightforward, quantitative mathematical question to which there is a right answer”[s]. Lower courts had successfully applied these metrics. As Michael Li, senior counsel at the Brennan Center, put it: “The five justices on the Supreme Court are the only ones who seemed to have trouble seeing how the math and models worked.”[s]
Post-Rucho, the worst gerrymandering states have intensified manipulation. Legislators now cite partisan intent as a defense against racial gerrymandering claims, arguing that targeting Democrats (who happen to be disproportionately Black) falls outside judicial review[s].
The Path Forward
State courts have partially filled the void. Pennsylvania, Ohio, New York, and North Carolina (before its supreme court flip) have struck down partisan gerrymanders under state constitutional provisions. Independent commissions in Michigan, Virginia, and Colorado remove legislators from the process entirely. Overall gerrymandering effects have declined from 23 net Republican seats in 2012 to roughly 3 in recent cycles[s].
Yet this progress is fragile. A single state supreme court election can reverse protections, as North Carolina demonstrated. Federal legislation establishing uniform standards remains blocked. Meanwhile, the computational tools that made sophisticated gerrymandering possible, costing $500,000 to $1 million in the 1990s and now widely accessible[s], continue to enable manipulation at unprecedented precision.
The mathematics of gerrymandering has given us the ability to see exactly what is being done to American democracy. The failure is not computational but political: the refusal of those in power to let transparent, quantitative standards constrain their manipulation of the electoral process. Every decade, mathematicians offer more precise tools. Every decade, politicians find new ways to dismiss them. The question is whether democracy can survive this asymmetry, or whether gerrymandering mathematics will remain a diagnosis without a cure.
