Ten years ago, in 1994, I wrote a column in U.S. News arguing that there was a serious possibility that the Republicans could win control of the House of Representatives. It was the first article in the national press foreseeing a possible Republican victory, and it appeared on the newsstands in the second week of July; we take the Republicans' 1994 victory for granted now, but only a few months before it seemed unthinkable to everyone but Newt Gingrich and his acolytes. How did I spot the trend? By noting that five Democratic incumbents were trailing their challengers in polls.
advertisement
Browse through an archive of columns by Michael Barone.
This year, the analysis of most reporters and of professional election trackers like Charlie Cook and Stuart Rothenberg has been that Democrats have no chance of regaining their majority in the House, no matter what happens in the general election. In that I have generally concurred, relying on district-by-district analysis. Yet by another measure, the generic vote, the Democrats would seem to have a chance. The generic vote—which party's candidate will you support for House of Representatives?—has been a poor indicator of the final vote over the past 10 years. Republicans have trailed in the generic vote consistently, except for the last weeks of the 1994 campaign and for a few weeks in other elections. Yet the Republicans have won a majority of seats and at least a plurality of popular votes in the past five House elections in a row.
Still, the Democrats have had a sustained and often a wide lead in the generic vote for most of this year. In the August 26 Rasmussen poll, which represents the results of the past three days, Democrats led 45 to 38 percent. During August their lead has averaged 44 to 39 percent; in July it averaged 44 to 38 percent.
So I decided to look at all available poll results in House races and see if I could find any with Republican incumbents behind. What I came up with is this:
Colorado's Seventh District. A Democratic pollster puts Democrat Dave Thomas ahead of incumbent Republican Bob Beauprez by 48 to 45 percent. That's striking, but there are reasons to discount it as indicative of a national trend. First, a Republican poll showed Beauprez ahead 39 to 32 percent. The undecideds are suspiciously high here and suspiciously low in the Democratic poll. Second, Thomas is Jefferson County district attorney, a well-known local official. His name identification might be better, or more in-depth, than freshman Congressman Beauprez's. Third, this was the closest district in the country in 2002. Beauprez won in 2002 by 121 votes. Al Gore carried the area covered by the district by 50 to 49 percent in 2000. It was a new district created for 2002 and designed to be politically marginal. A Beauprez loss would hurt Republicans but would not signal a disaster.
Connecticut's Second District. A recent Democratic poll here had two-term incumbent Republican Rob Simmons ahead by only 41 to 39 percent. That's not trailing, but it's not encouraging for an incumbent. This was a district carried by Gore-Lieberman over Bush-Cheney by 54 to 40 percent. Simmons's weak showing, if this poll is accurate, may be a sign of trouble for other northeastern Republicans in districts solidly carried by Gore. But most of them have won multiple times and have mastered the art of holding a district inclined toward the other party.
