(Dark Blue (236): Obama +7.6% or more
Lean Blue (28): Obama +2.6%-+7.5%
White / Toss-up (91): Obama +2.5% to McCain +2.5%
Lean Red (87): McCain +2.6%-+7.5%
Dark Red (96): McCain +7.6% or more)
New polls from Colorado, Iowa, Michigan, Missouri, Oregon, Virginia and Washington since the last front-page update. Because both of recent polling and because I have tightened the categories a bit, Missouri moves from "Toss-up" to "Lean McCain." The categories are now in the final form they will take between now and Election Day. In the coming months, future changes will include increasing the number of polls in the averages from 4 to 5 and, come mid-October, removing all pre-October data from swing state polls.
Overall, at either the state or national levels, I don't see any evidence that the campaign has changed in about two months. That isn't to say that it won't change in the future. Rather, I would argue that the appearance of daily change brought on by the Gallup and Rasmussen tracking polls is a mirage. A more comprehensive look at polling indicated that this is where the campaign has been ever since Obama clinched the nod.
* Maine four electoral votes are awarded as follows: two for the statewide winner, and one for the winner of each congressional district. ME-01 is about 3.5% more Democratic than the state as a whole, while ME-02 is about 3.5% less Democratic than the state as a whole.
** Nebraska's five electoral votes are awarded as follows: two for the statewide winner, and one for the winner of each congressional district. NE-01 is about 6.0% more Democratic than the state as a whole, NE-02 is about 11.0% more Democratic than the state as a whole, and NE-03 is about 18.0% less Democratic than the state as a whole.
Methodology I will update at least once every day between now and November 4th. The methodology is simple and straightforward.
For each state, take the last four polls conducted for the state, and average them.
If more than four polls were conducted in the state over the previous thirty days, all polls conducted during that time period are included in the averages.
If polling dates overlap, and make it difficult to determine which polls were the four most recent, include all of the overlapping polls.
No polling firm discrimination whatsoever. Polls are never excluded because the organization has a bad or partisan reputation. Also, if a polling firm has conducted more than one of the most recent four polls, all of the polls from that organization are included.
As we move closer to the election and more data becomes available, both the time frames for polls included in the averages will decrease and the definition of a "solid" lead will eventually drop to 7.0%.