I have long argued that we should give more weight in redistricting reform to the importance of competitive districts. You often hear that "the big sort" makes it difficult to create competitive districts. In an article with @ProfNickStephan, "Districting and Competition," we make use of simulated districting to show that we could have twice as many competitive districts without any sacrifice in traditional districting criteria. That would make the House more responsive to shifts in voter preferences and reduce the extremism associated with safe seats. Link in reply