President Trump’s low approval ratings and the Republican Congress’ lack of legislative accomplishments to date would appear to create a favorable situation for Democratic candidates to gain the 24 seats they need to take control of the House of Representatives in the November 2018 midterm elections.
However, even if Democratic candidates in the aggregate receive more votes for the House than Republican candidates across all 435 districts, the Republicans may still be able to hold on to their House majority.  Over the past decade, Republican legislatures in states across the country have successfully gerrymandered House district lines in such a way that Democratic gains in vote share will not necessarily translate into Democratic gains in House seats.
After President Obama’s election in 2008, Republican operatives in Washington decided that gaining state legislative seats in the 2010 elections would be a high priority for the party, especially in states that also had gubernatorial elections that year.  If Republicans could control both houses of the legislature and the governor’s office in states with multiple House districts, they could control the process of redrawing congressional district lines that would be required in all states with more than one House district after the 2010 census.
Aided by the Tea Party-inspired wave that helped Republicans in the 2010 midterms, GOP candidates picked up almost 700 seats in state legislatures all across the country.  When redistricting began, Republicans controlled the process in key states such as North Carolina, Ohio, Pennsylvania, Michigan, Wisconsin, and Texas.  By the time all the maps were drawn, Republicans had complete control over drawing 193 seats, only 25 fewer than the number needed for a House majority.
Some of these seats were formerly competitive ones that became safe for Republicans.  Others were formerly held by Democrats who could no longer win in their new districts.  A common Republican redistricting tactic was to pack Democratic voters into a relatively small number of House districts, which Democratic candidates would win with 70 to 80 percent of the vote, while taking those voters out of competitive districts that then became much safer for the GOP.
The best estimates for the 2018 cycle are that, in states with Republican-gerrymandered districts, Democratic candidates would need to win 53 to 56 percent of the aggregate House vote in the state in order to win a majority of House districts in the state.  In other words, GOP-drawn district lines give the party an advantage of 6 to 12 points in House elections in those states.
There are two possible remedies to this sort of gerrymandering.  The first is to take the task of drawing district lines out of the hands of legislatures and governors, and give it to an independent bipartisan commission.  Since politicians are most unlikely to relinquish the power to determine who their voters will be, this approach is only possible in states such as California and Arizona where voters could establish a redistricting commission through referendums. 
The second remedy is to seek relief in the federal courts.  In November 2016, a federal court in Wisconsin ruled that state’s legislative district map unconstitutional, after recounting the extensive history of Republican attempts to manipulate the line-drawing process to advantage the GOP.  In 2012, the first election the new districts were in effect, Republican candidates won 60 of the 99 state assembly seats with only 48.6% of the statewide vote.
The Supreme Court will hear the Wisconsin Republicans’ appeal of this decision in early October.  Many Court-watchers expect a 5-4 decision, with Justice Anthony Kennedy determining the outcome of the case.  If the Court rules against the Wisconsin map, the redistricting cycle after the 2020 census might see less blatant gerrymandering than after the 2010 census.
Eric L. Davis is professor emeritus of political science at Middlebury College.