Not to make too much of this one poll, but it does have a number of attractive aspects (in addition to showing Dino Rossi in the lead of course). It has more than double the sample size of the next most recent poll, and its breakdown by party affiliation is entirely reasonable when compared to 2006 figures:

PPP poll (Oct 29-31, 2010):

Democrat: 35 percent

Republican: 32 percent

Independent: 33 percent

CNN exit polling (November 2006):

Democrat: 39 percent

Republican: 29 percent

Independent: 32 percent

These numbers seem perfectly in line with a modest enthusiasm gap favoring Republicans and a slight uptick in independents, who are breaking for the GOP by considerable margins this year.

Overall, the poll has Rossi leading Sen. Patty Murray (D) 50 to 48 percent among likely voters. More significant, however, is the fact that among voters who said they had already sent in their ballots — 76 percent of those polled — Rossi leads 52 to 47 percent.

Among the remaining 24 percent who said that haven’t sent their ballots in yet (nearly all voting in Washington state is done by mail), Murray leads 51 to 45 percent, with 5 percent undecided. So assuming all 24 percent end up turning in a ballot, with the undecideds staying undecided, you get (roughly):

Rossi: 50.3 percent

Murray: 47.9 percent

Split the undecideds between the candidates and you get:

Rossi: 50.9 percent

Murray: 48.5 percent

Give all the undecideds to Murray and the Republican still wins:

Rossi: 50.3 percent

Murray: 49.1 percent

This is what PPP means when they say that “Murray’s ability to keep the race close is predicated on the 24 percent of respondents who have not yet done so returning their ballots” and “if Democrats in the state vote at a higher rate than response to our poll suggested over the weekend that would likely be enough to put Murray over the top.”

That’s a big IF, given the political climate even in blue state Washington, which does, contrary to popular belief, appear to have a significant enthusiasm gap favoring the GOP candidate. Conventional wisdom suggests that if anything Democratic turnout will be lower than expected.