One of the most important findings from my voter file analysis of the 2018 House elections is that even with very moderate Democratic nominees, GOP turnout was HUGE. That's weird, bc the theory is, extreme nominee juice opposition party turnout. Yet, even in the Blue Dog 
districts, GOP turnout surged. And it should be noted, these districts universally underperformed turnout among Ds compared to regular D candidates. Suggests these mod Ds paid a price for which they earned no benefit. Also suggests turnout of opposition party (at for Rs in
polarized era) may be unaffected by Dem candidate ideology- its just going to be high bc of R campaign stimulus & strategy, and across the board, D turnout will be lower bc of D campaign techniques. As for whether its better to be moderate or extreme, there was only 1 "extremist"
in the 2018 data set (outside those top 45 or so races, we're not really talking about competitive races anymore, so noting all the extremists lost really needs that caveat that they were, aside from Eastman in NE-2, running in long shot races w no party investment. And its not
likely we'll ever really get to test the performance of "extremists" bc I don't envision a world where Ds start nominating Sanders people in competitive races. BUT we might get to a point where Ds candidates start running as liberal Ds rather than Blue Dog (I would separate the
line at calling oneself a fiscal conservative). I'd argue Katie Hill, Katie Porter, Jennifer Wexton (to name a couple) just ran as liberal Ds. Certainly not extremists, but not as Blue Dogs. As Ds begin to recognize the role that partisan composition (demographic composition)
plays in the outcomes of close contests, I won't be surprised to see an continued increase of variability in competitive races with more Ds running as liberals. That would allow us to potentially examine electoral performance differences. But in terms of truly competitive
electoral environments (not the stretched definitions that have been used in these studies) its so rare to see an actual extremist prevail, you can't model it. BTW, Eastman, the one that did in 2018, interestingly, despite having the DCCC abandon her race performed exactly on par
w the party's 2016 nominee, who was a Blue Dog incumbent heavily supported by the D-trip. LOTS of spending. Nearly identical vote share. Its almost like the candidate (and their positions) don't matter. Hmmm, who told you that??
If you haven't read it yet, I strongly encourage you to read this voter file analysis to prep for my forecast update. If you're on a phone, don't miss the district graphs newrepublic.com/article/156402…
