House Race Summary for Oct. 6:
The trend on the so-called generic ballot question, which pollsters use to try to forecast the combined national vote for the U.S. House of Representatives, has remained essentially flat since Labor Day. As our chart shows below, preference for "the Democratic candidate in your district" has led the "Republican candidate" by an average margin of roughly 12 percentage points.

We continue to track the results as reported among all registered voters to keep the trend line comparable, but a few pollsters have also started reporting results among the smaller subgroups of voters they consider most likely to vote. The first was a USA Today/Gallup poll released two weeks ago that showed a big difference between all registered voters (who favored the Democrats by a nine-point margin) and likely voters (who were evenly split).

Since then, two polls each by CNN/ORC and AP/Ipsos have shown Democrats leading among likely voters by margins of 10-to-14 percentage points. Collectively, these four polls showed Republicans doing only about a point better among likely voters, on the margin, than among all registered voters.

Political junkies everywhere are now wondering whether the recent Foley and Woodward revelations will have an impact on voters' preferences. So far, at least, the handful of new surveys conducted in the last week show a small downturn in the Bush job approval rating, but no clear trend on the generic congressional ballot. Why not? One reason may be that opinions toward Bush are changing mostly among reliably Republican voters. Another may be that the dynamics of individual campaigns have favored Democratic candidates over the last month, a phenomenon we have certainly seen on our U.S. Senate Scorecard.

House Race Summary for Sept. 19:
The poll creating the big buzz today is the latest from USAToday/Gallup, which offers some good news for Republicans hoping to retain control of the House of Representatives. When respondents were asked whether they planned to vote for the Democratic or Republican candidate in their district, likely voters were evenly split at 48 percent each. That's good news for the GOP because polls have consistently shown a double-digit advantage for Democrats. But the findings are confusing. When registered voters were asked the same question, they preferred the Democrats by a nine-point margin (51 percent to 42 percent).

So, is it close or not? First, some perspective. Each poll inherently comes with a lot of random variation, so it's not wise to put much stock in any one result. We plot a five-poll rolling average trend line to try to smooth out some of the random noise.

Ideally, pollsters and pundits prefer to watch likely voters because, well, they're more likely to vote than those who are simply registered. But identifying the likely electorate is much more difficult when an election is still months away, because respondents are less able to honestly assess whether they're really going to vote. (Getting a large enough sample of likely voters also costs more money, so media pollsters usually wait until closer to the election.)

The problem is that once pollsters start screening for likely voters, their methodologies vary widely. This produces the scattershot results we've seen recently. An AP-IPSOS poll conducted last week showed likely voters preferring the Democrats by a 14-point margin (53 percent to 39 percent). Other surveys conducted over the last two weeks by Zogby, Harris, and Fox show results that were more encouraging for Republicans but not the even split that Gallup shows. Further, the Gallup poll's likely voter model has been criticized for producing volatile results, especially when used a month or more before the election.

So, the bottom line is that this new poll might be the start of a trend or just aberrant noise. For now, we're sticking with the more conservative estimate of our five-poll model that shows the average Democratic lead is closer among likely voters (six points) than among registered voters (nine points).