Today's Polls, 7/18

Although it is more apparent in some polling than in others, there now unambiguously seems to have been some movement toward John McCain.

In Maine, Barack Obama now leads by 8 points according to the latest numbers from Rasmussen. While Maine is unlikely to be competitive, Obama had led by 22 points in Rasmussen's June poll of the state. Polling movement like this in Maine, and what Rasmussen had shown earlier in the week in Kansas, is probably caused by statistical noise to a certain degree. Nevertheless, it is of a large enough magnitude that it almost certianly isn't caused entirely by statistical noise, especially considering that Obama's lead in both the major national tracking polls is down to a single point. McCain's electoral position has brightened a bit since the July 4 holiday.

In Alaska, a Research 2000 poll for Daily Kos has McCain expanding his lead to 10 points; he had led by 7 in Research 2000's only previous poll of the state in May. While it would certainly be worth Obama's while to visit Alaska, the state now looks like a somewhat less attractive target than Montana or either of the Dakotas.

In Virginia, Rasmussen has John McCain with a one-point lead (the candidates are tied if leaners are not included). In June, Rasmussen had shown Obama with a one-point lead -- however, McCain had led in all of its previous polling of the state. We have characterized Virginia for some time as the closest of the toss-up states. (With the entire Washington establishment within driving distance of Richmond, one can only imagine the chaos that might ensue if there were a recount there).

Lastly, in New Jersey, Strategic Vision has Obama ahead by 9. Based on some research that I have done, Strategic Vision is one of only two polling agencies so far this cycle to show a statistically significant bias in their numbers; their polls tend to be tilted about 3 or 4 points toward the Republican candidate. (The other polling firm in this category is Zogby Interactive, whose numbers tend to be tilted by 3 or 4 points toward the Democratic candidate). So, if Strategic Vision is showing Obama with a 9-point lead in New Jersey, the state is probably pretty darn safe for him.

There's More...

Contract Post

Today's Polls, 7/9

If Memorial Day marks the start of political spring training, and Labor Day the start of the playoffs, the July 4th holiday might be the beginning of the regular season. John McCain comes out of the break with one of his stronger polling days in some time.

There are two polls out in Missouri, both of which give McCain the lead. Rasmussen places his lead at 5 points, and Public Policy Polling puts it 3. This marks a reversal from a period where Obama had been polling fairly strongly in Missouri, having held the lead in Rasmussen's last poll conducted about a month ago.

There might, however, be a relatively simple explanation for this one: McCain has been saturating the state with advertising. Should Obama fight back? Well, he already is, having tripled his staff in the state. So, we might have the makings of an old-school air-versus-ground battle. But I also think that Missouri could be a little bit of a trap state for Obama. Our model, which now accounts for the relationships between the states in more sophisticated ways (more on this shortly), found only 43 simulation runs out of 10,000 when Obama won Missouri while losing Ohio. Missouri is essentially Ohio with 15 percent more evangelicals -- I don't think that's outweighed by its sharing a border with Illinois, but I can understand if the Obama people feel differently.

Another interesting state is New Jersey, where Rasmussen shows John McCain closing to within 3 points (5 if leaners are not included). Rasmussen has generally had New Jersey much closer than other pollsters, and as we noted this weekend, it is an exceptionally expensive state to compete in. Nevertheless, it has different demographics from virtually any other swing state -- McCain's one window into the Mid-Atlantic region. If he runs somewhat to the left, emphasizing his fiscal conservatism to wealthy New Jersey suburbanites while deemphasizing the guns and gays stuff (Scott Rasmussen also likes the offshore drilling issue for him here), the state could be competitive.

Lightning Round: In a now somewhat outdated Pan Atlantic SMS poll of Maine, Barack Obama has a 14-point lead (counting leaners). The poll also has great numbers for Susan Collins. And in Alabama, a Capital Survey poll has Obama within 13 points after having trailed by 24 last month.

Lastly, the Zogby Interactive results are now incorporated into the averages. Although we discount the value of these polls heavily, there are nevertheless 34 of them, and they provide enough help to Obama numbers in states like Arizona and North Carolina to mitigate McCain's gains in Missouri and New Jersey. Obama, however, has been polling below his trendline for the past couple of days, including mediocre results in the Gallup and Rasmussen trackers. If that continues through the end of the week, we may be able to credit McCain with a little momentum.

There's More...

Contract Post

Afternoon Polling Update, 6/18

So many polls, so little time! This is everything that's been released today, including the results from our morning edition.



Let's begin in Ohio, where Rasmussen's latest data contradicts the notion that Barack Obama is surging in the state. He still trails John McCain by one point in their polling; McCain had also led by one point in their May survey. There are several factors to consider here. Firstly, this poll postdates the Quinnipiac poll by a few days. I'm not yet convinced that Obama's bounce is receding, in part because there haven't really been any intervening news events to give some momentum back to John McCain, but that's always something to keep in mind. Secondly, some of the pollsters (like SurveyUSA and PPP) that had shown Obama leading in Ohio by fairly large margins tend to have a more fluid/less grounded conception of party identification than does Rasmussen. In Ohio, they're finding a huge shift in party identification, with as much as 50 percent of the state identifying as Democrat, and running with those numbers as is. Since SurveyUSA and PPP identify a lot more Democrats in their sample, and since much of Obama's bounce appears to be in the form of bringing Democrats back home to their party, it is not surprising if they are showing more movement toward him.

Unfortunately, there are no definitive answers about how one should measure party idenitifaction, and we won't really know who got it right until November. At a gut-feel level, it's hard not to imagine Ohio being somewhat close. At the same time, Obama's problems in the state had stemmed from his poor performance in Southern Ohio, which is part of Appalachia. And this is the region in which his bounce appears to be most profound: his polling has improved by 12 points in Kentucky and 15 in Arkansas, and he took some huge steps forward in Southern Ohio in Quinnipiac's regional breakdown.

Other Rasmussen polling does show a bounce for Obama. In Maine, he's ahead by 22 (up from 13 a month ago), and in Alaska, he trails McCain by just 4 points (down from 9 a month ago). There is certainly some novelty value in the notion of a Democrat competing in Alaska. But it's a state that the Obama campaign ought to be taking reasonably seriously: Alaska is the the youngest state in the country in a year where we have the largest-ever age gap between the two nominees. Indeed, it's probably time for Obama to visit Alaska. I don't have any numbers on this, but I would guess that candidate visits make more difference in smaller states, and particularly those that are out of the way geographically. If Obama visits Alaska, it will create a ton of earned media, and McCain will probably have to follow him to defend the state.

The notorious A.R.G. (American Research Group) is out with their first general election polling of the year; they have Obama leading by 5 points in Florida, and 12 in New Hampshire. You probably know that I don't have the highest opinion of ARG, but in their defense, their general election polling has tended to be pretty decent -- it's their primary polling that has been a mess. I do not, incidentally, find the New Hampshire result implausible, precisely because New Hampshire has some history of overreacting to current trends. With few New Hampshirites fitting into hard-and-fast demographic categories, and many of them identifying as independents, there is probably a higher fraction of swing voters there than any other state in the country. So as the country swings, New Hampshire swings twice over.

Finally, two results that tend to confirm our current impressions in a couple of states. In Virginia, Public Policy Polling has Obama ahead by 2 -- this is their first poll of the Commonwealth -- and in Wisconsin, SurveyUSA has Obama ahead by 9, up from 6 last month.

There's More...

Contract Post

Today's Polls, 5/16

You know it's a different kind of election season when you're having an easier time getting polling data for Alaska than for Michigan:



The headline today is that for the first time since we started tracking the state polls in early March, we are now rating Hillary Clinton as a favorite against John McCain. One of the big reasons why is that she is starting to consolidate her positions in blue states like Maine and Washington that previously looked like they might be somewhat vulnerable. What she needs now are some better numbers in Michigan and Wisconsin; if she gets those, her win percentage will go way up.

Barack Obama is also polling quite strongly in blue states recently. However, he appears to be backtracking somewhat in red states like Kansas. This is undoubtedly fairly typical for this point in the election cycle, when support begins to revert toward being more partisan. But the reason why we don't yet show Obama getting a surge in his electoral math even as he has improved his standing in national polls is because we haven't gotten a lot of data from key purple states recently, and what data we have gotten -- as in Wisconsin and New Hampshire -- has not been especially good for him.

There should be a lot of polling coming out next week, notably including SurveyUSA's monthly refresh of data in 15 or 16 states, so we should have a better idea by then.

There's More...

Contract Post

Today's Polls, 4/4

Rasmussen has a poll out in my favorite state in the country. Obama has a 10-point over advantage, and Hillary a 5-point advantage, over John McCain in the state of Maine.

Maine is a state I'd been curious about as being a potential swing state in a Hillary-McCain matchup. Maine has a substantial anti-establishment streak, which probably works better for McCain or Obama than for Clinton, and meanwhile, Hillary's fundraising has been slack there. On the other hand, this is a state that has no trouble electing female politicians, and the polling numbers give Hillary no real reason to worry as of yet.

Also, you'll notice that the weights of the polls have changed; this is a response to the good discussion we had in this thread about the relationship between sample sizes and the pollster reliability ratings. I've done a lot of thinking about this and have come up with a solution that I think is pretty good, but am going to ask your indulgence as it's a busy time and it will probably be a couple of days before I get the chance to explain everything.

There's More...

Contract Post