Today, I am excited to announce a new feature on Open Left that will be updated every weekday between now and the 2010 midterm elections: the National House Ballot monitor. It is simply a small button that will appear over Quick Hits, providing my latest forecast of the national House vote. The button will link to a table, like the one below, that includes all of the polls I used to produce the forecast.
The current forecast shows Democrats with a 5.3% advantage in the National House Ballot. For the sake of comparison, Democrats won the 2008 House national vote by 8.9% and the 2006 House national vote by 7.9%. Republicans won by 2.6% in 2004, and by 4.6% in 2002.
In terms of predicting seat totals, the National House ballot is, at best, a crude and inexact measure. As the 2010 elections approach, a far more detailed, seat by seat forecast will take prominence over the national House Ballot monitor. However, it is still a useful stat to see where the current national political environment stands.
The forecast is based on a new and improved electoral forecasting methodology that I will be using in 2010. My preliminary research--which I hope to complete this weekend and release next week--indicates that this new method is about 7-8% more accurate in terms of mean error, and 15-16% more accurate in terms of median error, in predicting the final margins of 2008 statewide elections than Pollster.com and fivethirtyeight.com. Further, the method actually performed slightly better in 2006 than in 2008, showing that its strong performance in 2008 was not a fluke. I have yet to examine how this method would have fared in 2004, which is why I still consider the research preliminary.
The methodology is pretty straightforward, and I explain it in the extended entry. For now, here are the polls used in creating the current forecast:
National House Ballot, 2010 (Polls can be found here and here)
Until the final six weeks of the election, take the simple mean of virtually all polls where the majority of interviews were conducted over the last 30 days.
During the final six weeks of an election, use the simple mean from the last 15 days.
Do not include Zogby Interactive polls and Columbus Dispatch polls, due to their horrendous past performance and questionable methodologies.
Do not include Strategic Vision polls, as it is starting to seem likely those are not real polls.
Include campaign-funded polls. Further, if there is more than one poll from a single organization, include all of them.
The basic idea is to cram as many polls with sound methodologies into the averages as possible, and weight them evenly to include more overall data in the sample. Because voter preferences don't really change that much in high-profile elections, I thought this method might produce a more accurate result through logic of regression to the mean. It seems to work pretty well, as my research has shown so far.
This is different from my 2006 and 2008 methodology in that it includes polls from 15 days out from an election, instead of only 8. Further, campaign funded polls, and multiple polls from a single polling firm, are now included. All of these changes were made to include more polls in the averages, since my previous methodology was about 10-20% less accurate than Pollster.com and fivethrityeight.com. Since they had already raised the bar so high, and since they will probably improve their methodologies for 2010 even more, it was time for Open Left to step it up. I believe we have, as I intend to show in detail on Monday.
As an anti-spam measure, there is a 24-hour waiting period after registering before new users can comment. blog advertising is good for you
blog advertising is good for you
