The Effects of Mid-Decade Redistricting on Electoral Outcomes

Effect of New Maps Enacted Since the 2024 Election

Note: Utah redrawn after court order. Ohio redraw as required by statute. Louisiana map struck down as a racial gerrymander (LA v. Callias). Mississippi, Alabama, Tennessee, and South Carolina have taken steps to prepare new maps after Callias. Data: Dave’s Redistricting, VEST.

Introduction

This analysis examines the effects of mid-decade redistricting after the 2024 election. I began writing when it appeared to be the final phase of an extended period of partisan map drawing. However, the U.S. Supreme Court’s decision in Louisiana v. Callais has prolonged this process, creating the possibility of additional changes to congressional district boundaries before the 2026 election. However, the extent of any additional map changes hinges on timing, as the electoral cycle is already underway in many states, with petitioning in progress, primaries either ongoing or completed, and remaining contests approaching rapidly.

Analytical Approach

In this context, the study adopts a data-driven approach to evaluating the consequences of redistricting, rather than relying on the “horse-race” style of analysis commonly used in media coverage. That approach typically focuses on predicting which incumbents are likely to win and which party is advantaged in the next election. By contrast, this analysis abstracts away from individual candidates and current seat holdings. It does not incorporate incumbency advantages or seat-level vulnerability. Instead, it aggregates outcomes across multiple statewide elections to estimate the average effect of district boundaries on electoral results.

This framework provides a broader perspective on redistricting. Rather than focusing exclusively on the 2026 election, it allows for inferences about how new district lines may shape outcomes in future cycles. For example, reporting on California’s Proposition 50 frequently suggested that it would produce five additional Democratic seats. While this estimate is plausible, the data indicate a wider range of outcomes. Across twelve statewide elections between 2016 and 2024, the new district configurations produce Democratic gains ranging from three to nine seats. This variation arises because the redistricting process not only converts previously Republican districts but also reallocates Democratic voters into districts that were previously more competitive. While these effects may be limited in the short term, they are likely to have greater implications in subsequent elections, such as those in 2028 or 2030.

Methodological Framework

Assessing the impact of redistricting on future elections is inherently uncertain, as electoral environments evolve over time. Nevertheless, historical data provide a useful basis for inference. Elections from 2016, 2018, 2020, 2022, and 2024 offer a set of observed voting patterns that can be used to evaluate how different district boundaries translate votes into seats. By applying these vote distributions to alternative maps, it is possible to estimate the average effect of redistricting. I download and clean data for each state under both the map used in the 2024 election, and that which will be used in the 2026 election. My dataset includes all statewide elections held between 2016 and 2024 that are available in Dave’s Redistricting App (a wonderful tool available to anyone with internet access and curiousity).

Methodologically, this approach relies on the fact that voter preferences are fixed within a given election, while district boundaries vary across maps. That is, district lines change, but voting preferences are locked in. This structure enables comparisons that hold voting behavior constant, thereby isolating the causal effect of district design on electoral outcomes. This is a considerable improvement over other estimates, which rely on less defensible assumptions. Most analyses I’ve seen either take for granted the legislature’s objective (“…just a simple redrawing. We pick up five seats.” Donald Trump, July 15, 2025), or compare the number of incumbents that appear more vulnerable due to the shifting lines. While my analysis will not predict how many seats will be gained in 2026 compared to 2024, it does provide an empirically defensible estimate of which party fared better overall from drafting new lines.

Findings

The results indicate that Republican-controlled states generally realized limited gains from recent redistricting efforts. In many cases, these states had already implemented highly partisan maps, which constrained the potential for additional advantage. Where changes did occur, they often involved the removal or reconfiguration of minority-opportunity districts following changes in the legal environment surrounding Section 2 of the Voting Rights Act. In contrast, Democratic gains tend to be larger. This reflects both the larger populations of states such as California and Virginia and the fact that these states began from relatively neutral districting baselines, creating greater potential for change.

To account for uncertainty, the analysis reports a range of possible outcomes under each new map relative to the maps they replace. In most cases, the party responsible for drawing district boundaries secures some degree of electoral advantage. However, there are important exceptions. In North Carolina and Ohio, certain electoral scenarios suggest that the new maps may disadvantage the party that enacted them. This phenomenon is sometimes referred to as a “dummymander,” in which attempts to maximize partisan gain inadvertently produce less favorable outcomes under specific conditions.

For example, in Ohio’s 2018 U.S. Senate election, the Democratic vote distribution would translate into victories in 10 of 15 districts under the new map, compared to 9 under the previous configuration. Similarly, in North Carolina, several elections — including the 2016 Auditor, 2016 Secretary of State, and 2024 Governor races — would yield outcomes that are less favorable to the majority party under the revised district lines. Despite these cases, most simulated scenarios still produce either no change or a net gain for the party controlling the redistricting process.

Florida is the latest state to vote to change their maps. The governor vowed to add four additional seats. Many commentators and even Rep. Hakeem Jeffries suggested that any effort to add Republican districts would amount to a “dummymander”. The empirical data suggests otherwise. But the claim that it will add four Republican seats is also exaggerated. I calculate that Florida has added an average of 2.23 seats, and as few as 1 and as many as 4. There is no historical election in the past decade where this map would not be a net-positive for Republicans. In both Texas and Florida, these gains are accomplished by demolishing minority performing districts, which are no longer protected after Callais.

Note: Both maps show the Harris v. Trump election.

Interpretation

Overall, the results suggest that Democrats are currently better positioned in this phase of mid-decade redistricting. Although fewer Democratic-leaning states have altered their maps, those that have done so are both more populous and were previously drawn under more neutral criteria. By contrast, many Republican-led redistricting efforts occurred in states that had already been optimized for partisan advantage, limiting the magnitude of additional gains.

Under a range of scenarios, Republican redistricting efforts yield relatively small net changes, while Democratic efforts produce larger potential gains. This asymmetry raises a strategic question for Republican mapmakers. In particular, the responsive actions taken in large Democratic states — such as California and Virginia — appear to have offset, and in some cases exceeded, expected Republican gains. These responses may not have been fully anticipated, suggesting that some Republican strategies were formulated under assumptions that did not account for reciprocal institutional changes in opposing states.

Conclusion

These conclusions should be interpreted with caution given the ongoing legal and institutional uncertainty surrounding redistricting. The Supreme Court’s decision in Louisiana v. Callais has created the possibility of further map revisions before the 2026 election. However, the electoral calendar imposes practical constraints, as many states have already entered critical stages of the election process. As a result, the ultimate configuration of district maps — and their electoral consequences — remains unsettled.

All data can be found archived at here. Any errors or ommissions should be directed to me at cervas@cmu.edu.

Thanks to Hayden Goldberg for editorial feedback and improving this post. Map was a collaboration with The Redistricting Network and its founder Jay Fierman.