Decision 2014: Forecasting the November 2014 midterm election results

Will the 2014 midterm elections be a Republican blowout year, a mildly Republican year, or a year where Democrats “break even”? That is the central question of this election cycle, and you will get substantially different answers, depending on which political observers you talk to.

With less than four months to go the November elections, we are in fact in a “quiet period” in the 2014 election cycle. Congressional filing has concluded in all but two states: Delaware will conclude its filing for Congressional/statewide races in two days, while Louisiana’s qualifying period is August 20-22. And while 31 states have already held their primaries, the next primary elections are not until August 5, although Alabama and North Carolina are holding its runoffs on July 15 (Louisiana’s primary is on November 4).

Given this “quiet period”, this is a good time to evaluate what we think will happen in November based on what has happened so far in the primaries. In the opinion of the author, partisan participation in recently concluded primaries is a good gauge of future (i.e., November) partisan enthusiasm. Particularly since we have a good benchmark (the 2010 primary election cycle) to compare the 2014 primary results against.

When we last did this analysis in 2010, we noted that 55% of primary voters participated in Republican statewide primaries. This, combined with Scott Brown’s upset win in Massachusetts, turned out to be good forecasts for the November 2010 GOP landslide. Furthermore, if you were to go back to the 2008 primaries, what was ultimately a strong Democratic year had its genesis in the Presidential primaries: the factional fight between the constituencies supporting Barack Obama (young people, minorities, well-educated voters) and Hillary Clinton (Jewish voters, older voters, white ethnic voters, blue collar workers) created a surge in new Democratic voter registration and Democratic primary participation.

Now that more than half of the states have held their primaries, we would like to revisit this issue. How can we best assess party enthusiasm from  primary results? Below is our methodology;

  • Find those states which in 2010 (and for one state, 2012) had contested statewide primaries for Senator or Governor for both the Democratic and Republican primaries,
  • Only include those states that similarly had contested statewide primaries for both parties this year, and
  • Only include those states (i.e., the 31 states) that have already held its primaries.14 states meet all three criteria, and it is these 14 states that will be the scope of our analysis.


There are two conclusions which we can draw from the data: (1) Republican enthusiasm (percentage-wise) is stronger than it was in 2010, and (2) Overall turnout volume is lower than in 2010, although Democratic turnout volume has deceased far more than Republican turnout. More specifically, here are the details of what we have found:

  • 2010 primary turnout: 10.6 million
  • 2010 party composition of primary turnout: 55-45% Republican
  • 2014 primary turnout: 9.0 million
  • 2014 party composition of primary turnout: 63-37% Republican
  • Change in turnout: 15% decrease (29% for Democrats and 4% for Republicans)
  • States where Republican primary participation % increased: 13
  • States where Republican primary participation % was flat: 1 (Idaho)
  • States where Republican primary participation % decreased: 0
  • 2012 Presidential preference: 53-45% Romney over Obama
2014 to 2010 primary turnout comparison

2014 to 2010 primary turnout comparison











As additional confirmation of what we have seen in terms of partisan enthusiasm, there was a 15th state (Pennsylvania), which had contested primaries in 2010 and 2014, but there was no Republican contest at the top of the ballot. While we could not include this state in our analysis for that reason, it’s also worth noting that Democratic turnout between 2010 and 2014 did decrease 18%.

While the 14 states used in our analysis are several percentage points more Republican than the nation as a whole, it’s important to remember that we are comparing “apples to apples”: 2014 versus 2010 turnout in states that had contested statewide primaries for both parties. Later in this election cycle, we will revisit this analysis as the remaining states conduct their primaries in August and September.


While the November election results will depend on the quality of the candidate each party selects, as well as the tone of the campaign waged, at this point, we are seeing more Republican than Democratic enthusiasm, and we believe that in a lower turnout midterm election year, that “enthusiasm gap” will make a difference in several critical House, Senate, and gubernatorial races.