Why It’s Too Early to Predict What May Happen in the House

by Stuart Rothenberg July 24, 2006 · 11:03 AM EDT

The Democratic Congressional polls keep rolling in, and almost all of them look very, very good for the party’s House hopefuls.

For example, a Cooper & Secrest survey for Democrat Joe Donnelly showed him leading Republican incumbent Rep. Chris Chocola by 10 points, 48 percent to 38 percent, in Indiana’s 2nd district. Two years ago, Chocola defeated Donnelly by 10 points.

Even more surprising is a Momentum Analysis poll for Victoria Wulsin (D) that showed her tied with GOP Rep. Jean Schmidt at 44 percent in Ohio’s 2nd district. And recently, a Global Strategy Group poll for Patrick Murphy (D) showed him trailing incumbent Rep. Michael Fitzpatrick (R) by only 6 points, 44 percent to 38 percent, in Pennsylvania’s 8th district.

Assuming, for the purpose of argument, that these surveys are accurate, why shouldn’t they lead me (or anyone else) to go out on an early limb and predict a Democratic takeover of the House?

After all, if Donnelly is going to give Chocola a 10-point drubbing and Wulsin — who finished a distant second to Iraq war veteran Paul Hackett in the 2005 special Democratic primary but won her party’s nomination for 2006 — is running even with a Republican incumbent in a very Republican Ohio district, isn’t that compelling evidence of a Democratic tsunami?

Well, I’m not buying just yet. Ask me again in a couple of months.

Democrats have a number of surprisingly good polls that suggest a wide range of Democratic challengers, from top-tier hopefuls to second- and third-tier long shots, have a serious chance of winning this fall.

But that’s the problem. Rather than reflecting the appeal of Democratic candidates, those surveys primarily reflect the national political landscape.

The president is in a deep, deep hole, and Congress is even less popular than President Bush. The GOP is divided over a number of issues. And Americans tell pollsters they want change and that they prefer a Democratic Congress.

Given that, voters are likely to have a less-than-positive view of Republican incumbents, such as Chocola and Schmidt, and their dissatisfaction with the status quo is likely to make them tell pollsters that they prefer Democratic candidates for Congress.

Donnelly’s campaign manager explained her candidate’s strength — and his improved numbers from a November 2005 Cooper & Secrest survey that showed him holding a 6-point lead (46 percent to 40 percent) over Chocola — by saying, “It’s clear that Joe’s message of change is getting out to the voters of the 2nd district.”

There are two problems with that interpretation. First, while MoveOn.org and Chocola were up with TV buys between the two polls, it’s hard to believe that a lot of people were focusing on the race. Moreover, Donnelly’s campaign was not on the air.

Second, the ballot test in the July 10-13 Cooper & Secrest poll strongly mirrors the generic ballot in the district. Donnelly leads Chocola by 10 points in the ballot test (48 percent to 38 percent) while a generic Democratic candidate leads a generic Republican by 10 points as well (46 percent to 36 percent). The Democrats’ generic ballot advantage grew from 1 point in November 2005 to 10 points earlier this month, which also helps explain Donnelly’s improved standing in the July survey.

The Momentum Analysis poll also appears to be shaped by general voter attitudes about Republicans. While Wulsin drew 44 percent against Schmidt in the poll, only 13 percent of those surveyed had a favorable opinion of the Democratic challenger. In other words, she largely was unknown and, therefore, irrelevant. To most people in the district, only her party affiliation and the fact that she is not an incumbent mattered.

As I’ve argued over the past six months or so, the midterms already have been nationalized and that’s why so many Democratic candidates are doing so well in ballot tests against GOP incumbents. Unknown Democrats, in many cases, are benefiting from their anonymity.

But the GOP strategy for holding the House is based on localizing the midterms, and that means making Democratic challengers unacceptable alternatives.

Nobody knows yet whether the Republican attacks successfully will discredit their Democratic challengers, or whether voters are so dissatisfied with current political leaders that they will vote for Democrats no matter what Republicans say about them. And we won’t know until September and October, when the strongest Republican attacks will hit and voters start to give serious consideration to their choices. It’s only prudent to wait to see whether Republicans can change the dynamic in enough districts to hold the House.

My point definitely is not that the national dynamics don’t matter. Of course they do. The lay of the land is a huge problem for Republicans, and it makes it more difficult for them to make their opponents unacceptable to voters.

Nor am I saying that Chocola, Schmidt and Fitzpatrick are without their own problems. They have specific issues to deal with in their races, and those problems may make it more difficult for them to shift the focus of their contests to their opponents’ shortcomings.

I have been quite outspoken in arguing that the Democrats have a very good chance of winning the House in November, even with the structural problems they have because of the paucity of competitive districts. National and district polling confirms that conclusion, as does the scarcity of upbeat GOP district polls.

But anyone who buys the early Democratic polls at face value is making a mistake. Many of these surveys measure the landscape, not the combatants, and we won’t know how well Democratic candidates will do until much later in the year, possibly mid-October — after Republicans have spent some of their sizable war chests on demonizing their opponents.