Horserace Talk
Comparing the Democrats’ 2006 Challenge to Republicans’ 1994 Challenge

When evaluating Democrats’ potential for seat gains, it is instructive to contrast the political landscape for Republicans in 1994 to the landscape for Democrats in 2006. Reflecting the general tilt in congressional district partisanship toward Republicans (a tilt that is primarily due to how concentrated each party's base of support is) and how many Democrats in 1994 were able to represent Republican-leaning districts, there are several stark differences. Even if Democrats in 2006 match the two-party vote totals of Republicans in 1994 (when the Republican advantage was about 54% to 46%), they have significantly fewer opportunities to translate that advantage into a comparable gain in seats. To experience anything comparable to the Republicans’ electoral success in 1994, they almost certainly will need a two-party win more along the lines of 57% to 43%.

Consider the following analysis of open seats and vulnerable incumbents, in which the discussion assumes a 54%-46% Democratic year. Note that the parties are sometimes referred to “R” for Republican and “D” for Democrat.

Open seats

1994:      52 open seats
  • Republicans leaving 21 seats and Democrats leaving 31 seats
  • 37 seats in R-leaning districts (Republicans win 35, including 20 pick-ups from Democrats)
  • 8 seats in districts with Democratic partisanship of 50% to 57% (Republicans win 4, including 2 pick-ups from Democrats   
  • 7 seats in districts with Democratic partisanship of at least 57% (Republicans win zero)                 

2006:     30 open seats
  • R's leaving 19 seats, D's leaving 11 seats
  • 11 seats in Democratic turf (Democrats already hold 9 of 11 of them)
  • 11 in districts with Republican partisanships of 50% to 57% turf (R’s hold 8 of the 11)
  • 8 in districts with Republican partisanships of at least 57%+ (R’s hold all 8)


  • In 1994, Republicans picked up 20 open seats previously held by Democrats that were in Republican-leaning districts. But in 2006, Republicans only have to defend two open seats in a similar category - e.g., Democrats at most can only pick up 2 open seats in districts that are Democratic-leaning.
  • In 1994, Republicans picked up 2 seats in districts that were less than 57% Democratic partisanship, held onto 2 open seats here and failed to win four. In 2006, Republicans have to defend 8 such seats. In a 54%-46% Democratic year, Democrats on average would win 5.
  • In 1994, R's did not win any open seats in 57%-plus Democratic districts. Democrats have to take on 8 Republicans in such seats in 2006. Democrats on average would win 1 or 2 seats.

Bottom line:

The terrain for open seats was much more fertile for a 54%-46% partisan result helping Republicans in 1994 than Democrats in 2006. Based on the law of averages, Democrats would seem fortunate to gain 8 or 9 open seats in 2006, while in 1994, Republicans gained 22 open seats (and also lost 4 they had held).

Vulnerable incumbents

In 1994, Republicans defeated 34 Democratic incumbents. Of these incumbents, 27 were in districts that leaned Republican, including fully 11 representing districts that were at least 55% Republican.

In the current Congress, Republicans only have three incumbents who represent districts that FairVote measures as at least 55% Democratic (as opposed to 149 Democrats who represent these heavily Democratic districts). There are only 17 additional Republican incumbents who represent Democratic district in the 50%-54% range of Democratic partisanship. Even if the Democrats were able to take over all of these 20 districts – a highly unlikely result -- that would still be far short of the 27 defeated Democratic incumbents and 20 open seat gains in such districts in 1994. (Note that in addition, for Democrats to have a chance at long-term control of the House, one must hold onto your districts, not just win them once. Of the 7 Republican candidates who defeated Democratic incumbents in Democratic-leaning districts, four them lost in 1996. Of the 27 remaining Republicans’ victorious challengers, only two were defeated that year.)

Bottom line:

Democratic challengers have to do far better in Republican-leaning districts in 2006 than Republican challengers needed to do in Democratic-leaning districts in 1994. The basic math of a significant Republican advantage in districts (after 2004, 238 districts were Republican leaning and only 197 Democratic-leaning) is a big boost for Republicans’ ability to win elections and govern in a manner that pleases their base.

Competitiveness trends:

1992 (last pre-alignment election)                           
  • 84 seats won by less than 10%            
  • 52 seats won by less than 5%                 
  • 24 incumbents lost                             
  • Average victory margin: 30.5%               

2004 (with partisan alignment all the more solidified)
  • 23 seats won by less than 10%
  • 10 seats won by less than 5%
  • 7 incumbents lost (only 3 not directly affected by Texas gerrymander in 2003)
  • Average victory margin: 40.5%

Bottom line:

There is a much smaller universe of competitive seats, meaning more incumbents have greater cushions of support than incumbents did after the 1992 election. Note that this is much less due to gerrymandering than it is to hardening partisanship in the wake of the partisan realignment that accelerated after the Cold War, with Republicans solidifying congressional representation of much of the South, rural and exurban America. The biggest decreases in the competitiveness trends above in fact took place after the 1996 election, with a jump in the average victory margin in 1998 to 43% (from 30%) and a plunge in the number of close races from 80 won by less than 10% in 1996 to 43 in 1998. No more than six challengers have defeated incumbents in any House election since 1996, and more than 90% of races in each election have been won by at least 10%.