Since the end of the conventions, I've been giving weekly updates on the competitive states for both the presidential election and U.S. Senate every Saturday on Overnight News Digest. I determined the competitive states based on the percentage chances to win at Nate Silver's FiveThirtyEight blog at the New York Times. I counted those states that showed the two major party candidates having probabilities to win between 20% and 80% inclusive as competitive that week.
Here's what I wrote about the state of the race in last night's Overnight News Digest: Science Saturday (Sandy's aftermath edition) on Daily Kos.
As of November 3rd, the presidential swing states are Colorado, Florida, North Carolina, and Virginia, while the states with competitive races for the U.S. Senate are Arizona, Indiana, Montana, Nevada, and Wisconsin. Since my previous report last week, Iowa, Nevada, New Hampshire, and Ohio have moved out of the competitive presidential state category with an 80+% chance of an Obama victory, while North Carolina has moved in with a 20+% possiblity of an Obama win. Also, Wisconsin has been added to the competitive Senate category because Tammy Baldwin now has less than an 80% chance of being elected.
Since I wrote that, the state of the presidential contest has already changed, as
Nate Silver calculated Obama's chances of winning New Hampshire dropping from just over 80% to just under, 79.4%. That's the bad news. The good news is that his chances of winning Colorado increased to over 60% (68%) and Virginia his odds went up to over 70% (71%). A week ago, he had less than 60% odds of winning Colorado and just over a 50% probablity of winning the Old Dominion. Things are moving fast, and I wouldn't discount
the effect of Sandy on the polling.
For details about Virginia and Iowa, follow over the jump.
Speaking of Virginia, here is an article about the latest polling in the state, courtesy of the University of Virginia.
Obama and Kaine Ahead in Central Virginia, U.Va. Poll Finds
November 1, 2012
With less than a week remaining before Election Day, the University of Virginia’s Jefferson Area Community Survey shows Democrats Barack Obama and Tim Kaine holding substantial leads among registered voters in the greater Charlottesville area.
In a survey conducted Oct. 2 through 31, 49 percent of registered voters said they planned to vote for Obama for president, compared to 33 percent for the Republican challenger, former Massachusetts Gov. Mitt Romney. Another 8 percent were undecided, and most of the remaining 10 percent declined to state a preference. Fewer than 1 percent of voters supported other candidates and 1 percent of registered voters said that they had decided not to vote.
Similarly, 51 percent of respondents said they favored Kaine in the U.S. Senate race, compared to 32 percent for Republican George Allen and 9 percent undecided. Most of the remaining 8 percent declined to state a preference.
Nearly all of that was pre-Sandy, so maybe the effects of the Frankstorm just cemented an already existing trend. As for Kaine, Nate now gives him an 85.0% chance of winning Virginia. I got what I was hoping for when
I wished that Kaine would leave Allen in the dust at their last debate.
What about Iowa, which Nate Silver thinks is even more likely to go Obama's way?
University of Iowa: Tight race
Hawkeye Poll: Presidential race close in Iowa, likely determined by ability to turn out supporters
By: University Communication and Marketing
2012.10.31
With the presidential election less than one week away, it’s still a close race in Iowa, a key swing state. Mitt Romney has a slight edge over Barack Obama among likely voters in the state, according to a University of Iowa Hawkeye Poll released today. The Hawkeye Poll is a teaching, research, and service project of the Department of Political Science in the College of Liberal Arts and Sciences (CLAS).
Obama has a slight lead in Iowa among all respondents, with 42.7 percent of the vote to 41.0 percent for Romney, with 10.5 percent undecided and 5.8 percent preferring a third party candidate. Romney leads among likely voters, though, with 45.2 percent of the vote compared to 44.4 percent for Obama, with 6 percent undecided and 4.3 percent preferring a third party candidate. The margin of error for the survey of 320 Iowans is 5.6 percent.
“Our results show Romney making advances and perhaps taking the lead in Iowa, and that the race continues to be close and within the margin of error,” says Frederick Boehmke, associate professor of political science in the UI CLAS and faculty adviser of the Hawkeye Poll. “It appears that the final result will be determined by each campaign’s ability to turn out supporters and to capture the votes of those last few undecided voters.”
Nate is using the registered voter numbers from this poll, not the likely voter numbers. I'd be worried, except that there have been lots of polls since then that confirm a small but solid Obama lead in the state (3.3% with an 83% of winning).
Finally, it wouldn't be a state of the contest post without a mention of the Iowa Electronic Market (IEM).
University of Iowa: Traders see status quo
Iowa Electronic Markets sees Obama winning popular vote, GOP House and Democratic Senate
By: Tom Snee<
2012.10.30
Prices on the Iowa Electronic Markets’ (IEM) political prediction markets have been largely steady the last week, unmoved by the rapidly approaching Election Day or Hurricane Sandy thrashing the northeast.<
As of 9 a.m. Tuesday, a contract for President Barack Obama was trading for 63.8 cents on the IEM’s Winner Take All market, which means traders believe he has a 63.8 percent probability of winning the popular vote in next week’s election. He has been trading in the 60 to 65 cent range for most of the last two weeks.
Romney, meanwhile, was trading for 38.9 cents, which means traders believe he has a 38.9 percent probability of winning the popular vote. His current price also reflects his recent trading trend.
That was Tuesday the 30th. The next day, prices started moving. Right now, the
current quotes show Obama's likelihood of winning at 70-72% and getting about 51.5% of the two-party vote share. Here's the graph showing the recent price swing.
Looks like the IEM is getting back in line with Nate, whose current forecast is for an 80.5% chance of Obama winning the popular vote (this is what the IEM is forecasting, not his chances of winning the Electoral College, which Nate has at 85.1%) and earning 50.6% of the popular vote.
Speaking of Nate, I'm amazed at the meta fight over his predictions. I never thought this election would turn into a referendum on Nate Silver!