In "Alabama Votes; I Remount My Hobby Horse," I noted how on the congressional generic ballot question Democrats currently enjoy an 11-point advantage over Republicans. I suggested, further, that an 11-point victory by Democratic candidates for US House in the 2018 midterms might--but might not--be enough to deliver control of the House to Democrats. This is because gerrymandering, such as may be seen in the map of Alabama's congressional districts, requires Democrats to win the national vote in a near landslide in order to win a slim majority of the 435 House elections.
Now a new CNN poll, released yesterday, places the Democratic advantage in the generic ballot at 18 points. It's still almost a year until people will vote, but a result such as this brings into view a possibility I haven't before mentioned.
To oversimplify just a little, the essential strategy behind Republican gerrymandering is to create, wherever possible, congressional districts that are either about 55% Republican or 60% or more Democratic. If voters from sea to shining sea are about evenly divided, the roughly even national vote will then result in Republican control of the US House--because there will be more districts that are 55% Republican than are 60-plus% Democratic. Simple math. You want to keep the Republican districts at 55% so that you can withstand an election in which there is a pretty strong shift toward the Democrats. The recent US Senate election in Alabama is illustrative, especially when compared to the presidential result of 2016, in which Hillary Clinton received 35% of the statewide vote and carried one of the state's seven congressional districts. Then, last week, Doug Jones, the Democratic US Senate candidate, won the special election with 50% of the statewide vote--but still, like Clinton, carried only one congressional district.
Regarding representation in the US House, Alabama is for Democrats an impregnable fortress erected on the foundation of a gerrymandered map. The question of what it would take for Democrats to win four congressional seats, a majority within the state, would seem to be of only theoretical interest, since it's very hard to imagine any Democrat in Alabama doing better than Jones did. Still, along this line, it's worth noting that the 15-point swing toward Jones, compared to Clinton's share of the Alabama vote, had the effect of making the result quite close in several of the districts that Jones lost (which is why so much of the above map is light pink). Indeed, his Republican opponent got no more than 51% of the vote in four of those districts that Jones lost. If Jones could have done even better--say, 2.5% better evenly distributed across the state--he would have carried not one but five of Alabama's seven congressional districts.
Here's the thing: the strategy on which Republicans have relied to give themselves a seemingly shockproof advantage has the admittedly remote downside of creating a tipping point at one outlying end of imaginable election outcomes. If that tipping point is crossed, they are wiped out. You can think about it like this (again, from an altitude in order to illustrate the general truth about the lay of the land, electorally): If Democrats win 50% of the national vote for US House, they will not win anywhere near half the seats; and if they win 52% of the national vote, they will hardly win any additional seats over what 50% gives them; and if they win 53% of the national vote, they will win a negligible number of more seats; and, at 54%, they again win a few more seats though they likely remain in the minority; but, at some point, as their share of the national vote inches upward, one extra percent will suddenly give them fifty or more additional seats. I'm not Nate Silver, and so don't know where this tipping point is, but we now have a reputable poll showing a 56-38 Democratic advantage in the congressional generic ballot, and the tipping point probably isn't north of that. Republicans need to hope for at least a modest reversion toward the mean, in which case the advantage they've given themselves will yield a result better than they deserve.
The conclusion of the matter is that Republicans have a decent chance of holding their majority. But if they don't, they have a good chance of being buried beneath a few tons of electoral rubble. Because of how they've organized things, the space between these divergent outcomes is narrow.
Comments