Who'd control Congress today? Midweek midterm elections update
Bottom Line: We're up to our 31st midterm election update. And now that our primaries are behind us it’s getting down to crunch time with under five weeks until Election Day. First, here's what history tells us about midterm elections...
Since the advent of the current two-party system (39 midterm elections) we've averaged the President's party losing 4 Senate seats and 30 seats in the House. If that happens this year Democrats would retake control of both chambers of Congress. Democrats only need to flip two Senate seats to retake control and they need 24 seats in the House. History is on the side of the Democrats reclaiming control going into this cycle.
There are only three times that the incumbent President's party has gained seats (1934 during FDR's first term, 1998 during Bill Clinton's second term and 2002 during George W. Bush's first term) thus only 3 out of 39 midterm elections have resulted in the President's party gaining seats. Here's another way of looking at it... History suggests there's a 92% chance Democrats will gain Congressional seats this year. The question becomes how many. That's where it's helpful to look at the history of generic ballot polls and outcomes. These are the past four cycles:
The first number is the average generic ballot polling on Election Day and the second is the actual result:
- 2014: GOP +2.4 - GOP +5.7 = GOP +3.3%
- 2010: GOP +9.4 - GOP +6.8 = GOP -2.6%
- 2006: DEM +11.5 - DEM +7.9 = DEM -3.6%
- 2002: GOP +1.7 - GOP +4.6 = GOP +2.9%
Polls average being off by about 3% - however history has shown that the party with a generic ballot advantage has always performed the best in the midterm elections. This is analogous to a "home field advantage" and represents about a third of the picture when attempting to determine the likely outcome of elections. As of today, the generic ballot says (average of accredited polling over the past week) ...
Current: DEM: +7%
So here we sit one week later. The Kavanaugh fiasco having happened and we’re exactly where we were last week. With under five weeks to go the outcome of who’ll control what Congressionally is still very much a tossup. As of now we’re pacing a divided Congress.
If the current landscape holds on Election Day, we’re looking at a range of around 27 to 32 pickups for Democrats in the House which would be enough to gain control. There wouldn’t likely be any change in the Senate. It is worth noting that in most simulations of House scenarios I’ve found that the party which wins two of the three South Florida toss-ups (18, 26 and 27) will control the House. As of now we’re seeing Brian Mast lead in 18 but for the first time since polling has emerged for the general, Carlos Curbelo in 26 is polling behind by one point. The 27th between Donna Shalala and Maria Elvira Salazar shaping up to be closer than pundits thought originally (coming into the cycle this was considered the easiest pickup in the House for Democrats).
The individual characteristics of these races project a more neutral election landscape in which the candidates and the issues matter most. In a wave style election we wouldn’t be having this conversation.
Democrats have led on the generic ballot by an average of 3% to 12% in 2018. Anything in the 8%+ range would likely produce a "wave" type of election. Anything in the 4% or under range would likely result in Republicans retaining complete control. Individual candidates and issues in specific races can heavily influence outcomes however Democrats appear to be gaining momentum as we’re down to the final two months before the midterms.
Until next time...