>>There's a reason no one else quotes him.
Rasmussen blew the 2000 election for whatever reason. He claims that his polling company made adjustments to its methodology. I am willing to forgot the 2000 results for that reason.
However, I don't trust his polls because the only time he agrees with the other polls is when the others show the race close. Does his poll ever show anyone with a lead beyond the MOE? I have a hard time believing that the electorate has been that evenly divided for that long. Although the election might wind up close (but I don't think so), the candidates are going to have periodic surges in popularity. His polls never seem to reflect any current events that historically would have affected the election's horse-race numbers.
In addition, as several people have pointed out, there seems to be inconsistencies between his national polls and his state polls.
I'm afraid that Scott Rasmussen might have to readjust his methodology again after the 2004 election.
I've been keeping a rolling average of all poll results (by all pollsters) since May. I have yet to show either candidate ahead outside the margin of error.