I make it a tradition to add up all the House votes to see which party did best in the midterm elections. Using the numbers on the Fox News website from about 24 hours after the polls closed, I come up with:
The problem with this, however, is that the Democrats had 30 seats where their House candidates were unopposed, so Fox didn't list a vote total, versus only 4 where the Republican candidates had a walkover. So, it was actually significantly worse for the Republicans. So, I decided to get the average number of votes for all parties for the 401 contested districts (178,481), and add that to the party totals for each walkover they had.
Now, that may exaggerate the Democrats' advantage because uncontested districts are likely to have lower turnout in general than competitive districts. So, if I assume that turnout would have been 80% of average in the 34 uncontested districts and the dominant party would have won 80-20, then I get:
Care to share your opinion on what is the best way to handle this methodological problem?