(but it doesn't look good for Russ, here in WI):
After having lost ground in the Senate forecast for three consecutive weeks, Democrats have demonstrated improved polling in several key matchups over the past week, dimming Republican hopes for taking over the chamber.
The FiveThirtyEight model now gives Republicans an 18 percent chance of claiming control of the Senate after the Nov. 2 elections — down from 24 percent last week. The projected composition of the Senate has also changed slightly: more than 100,000 simulation runs of the forecast model show the Democrats finishing with an average of 52 senators, up from 51.5 last week, and Republicans with 47.9 senators, down from 48.4 last week.
It would be dubious to assert that Democrats have some “momentum” at the national level — pollsters have released surveys in dozens of matchups for the House this week, for instance, and they continue to be broadly in line with large (potentially very large) Democratic losses in that chamber.
Control of the Senate, however, will boil down to a relatively small number of races — possibly not more than a half-dozen. And in several of those races, Democrats have made small, but important, gains:
In Nevada, two of three polls released this week show the Democrat, Harry Reid, with a slight lead over the Republican, Sharron Angle. The model still regards Ms. Angle as an incremental favorite — but it now sees the race as closer to 50-50 after having assigned her a two-in-three chance of winning last week.
In West Virginia, two polls, from Rasmussen Reports and Public Policy Polling, show Gov. Joe Manchin III, the Democratic Senate nominee, improving his position slightly: he may be benefiting from sharply differentiating himself from President Obama, while simultaneously hitting his Republican opponent, John Raese, over a series of minor gaffes. West Virginia was also surveyed for the first time this week by a traditional pollster, CNN/Opinion Research, which showed the race in a tie. Like Nevada, West Virginia is a state that had moved toward Republicans in last week’s forecast, but it is now closer to being a tossup; the model makes Mr. Manchin a 57-to-43 favorite.
In Washington, the polling has been contradictory. But on balance, the two polls showing the Democrat, Patty Murray, with meaningful leads of 8 and 15 percentage points have moved the forecast back toward her. She is now given a 78 percent chance of winning, up from 71 percent last week. Ordinarily — given that some other polls of the state show a tie — this forecast might seem a bit on the aggressive side. But ballots are being sent out right now in Washington, which votes almost entirely by mail — and if Dino Rossi, the Republican, indeed trails Ms. Murray by anything more than a couple of percentage points, it might be hard for him to make up the gap.
In Illinois, where the polling has been close for months, a Rasmussen Reports poll now shows the Democrat, Alexi Giannoulias, with a 1-point lead; he had trailed by 4 points in their previous survey of the race. Other nonpartisan surveys, like one from Southern Illinois University, also show the race within a point or two in one or another direction. Particularly given the high number of undecided voters in the state, the forecast is likely to remain a tossup unless one candidate makes a fairly definitive move. The forecast now has Mr. Giannoulias with a 53 percent chance of winning, and the Republican, Mark Kirk, with a 47 percent chance.
One high-profile race, Wisconsin, does show a Republican trend this week. Nonpartisan polls of the state are unusually consistent, giving the Republican, Ron Johnson, a roughly 8-point lead over the Democratic incumbent, Russ Feingold. The polls also show that there are few undecided voters remaining in the state — and most of those surveys show Mr. Johnson with more than 50 percent of the vote. Although internal polling by Mr. Feingold’s campaign shows the race closer, candidates simply do not come back very often when the public polling tells a story like this one, and Mr. Feingold’s chances of retaining his seat are down to just 6 percent in the model.
Wisconsin, however, has looked for several weeks to be a probable G.O.P. pickup — so while it may get them their 46th or 47th seat in the Senate, it would probably not represent their 51st.
And it is the 51st seat that poses the problem for Republicans. If Republicans were to win each of the races where they are currently favored, as well as the essentially tied races in Illinois, Nevada and West Virginia, they would emerge with control of exactly 50 Senate seats. But there is something of a gap between these states, and their next-best opportunities are in Washington and California — states where the consensus of polling suggests that Democrats would probably win if the election were held today.
And apart from Washington and California, Republicans are running out of opportunities. A poll using Rasmussen Reports methodology suggested some tightening in Connecticut, but with a wide assortment of polling last week having shown the Democrat, Richard Blumenthal, with a double-digit lead and over 50 percent of the vote, modest tightening would not be enough to give the race to the Republican, Linda McMahon.
Meanwhile, their chances of pulling off an upset in Delaware — already slim — are now close to nil.
http://fivethirtyeight.blogs.nytimes.com/2010/10/13/democratic-polling-improves-in-key-senate-races-lengthening-g-o-p-takeover-odds/
Printer-friendly format
Email this thread to a friend
Bookmark this thread
(1000+ posts)
Send PM |
Profile |
Ignore
