How Off Were The Final RCP Averages in 2004?

Answer: a few points, generally.

Sources: RCP Final by state, Actual Voting, double-checked here. I’m only using states close enough that polls would be considered reliable. 

State	Bush/Kerry RCP Margin	Bush/Kerry Actual Margin	Difference
Florida	Bush +0.6%	Bush +5.01%	4.41%
Ohio	Bush +2.1%	Bush +2.11%	0.01%
Pennsylvania	Kerry +0.9%	Kerry +2.5%	1.6%
Wisconsin	Bush +0.9%	Kerry +0.38%	1.28%
Iowa	Bush +0.3%	Bush +0.67%	0.37%
Minnesota	Kerry +3.2%	Kerry +3.48%	0.28%
Michigan	Kerry +3.5%	Kerry +3.42%	0.08%
Missouri	Bush +4.2%	Bush +7.2%	3%
New Mexico	Bush +1.4%	Bush +0.79%	0.61%
Nevada	Bush +6.3%	Bush +2.59%	3.71%
Colorado	Bush +5.2%	Bush +4.67%	0.53%
New Hampshire	Kerry +1.0%	Kerry +1.37%	0.37%
Maine	Kerry +9.5%	Kerry +8.99%	0.51%
West Virginia	Bush +8.5%	Bush +13%	4.5%
Oregon	Kerry 4.8%	Kerry +4.16%	0.64%
New Jersey	Kerry 7.0%	Kerry +6.68%	0.32%
Arkansas	Bush 6.5%	Bush +9.76%	3.26%
Hawaii	Bush 0.9%	Kerry +8.75%	9.65%
Average			1.95%
Average w/out Hawaii			1.50%

The Hawaii polls were not really believed at the time either, so I included a separate average without them. As you can see, the numbers were generally quite close, with only a handful of states exceeding the excessive margin.

The point for this year is that most states will likely be very close to what is predicted, but a few may differ. And the few that differ are likely to be the states polled the least, like North Dakota.

The best that McCain partisans can hope for is that there are very, very large breaks against Obama. I think that is really unlikely to occur.

Sphere: Related Content

2008Central.net Presidential Election Podcast (11/02/08)

In this podcast, we discuss projections for election day and reflect on the campaigns of Barack Obama and John McCain in order to glean insight into how they would serve as president.

[Subscribe to 2008Central.net's Presidential Election Podcast]

Sphere: Related Content

Homogenizing The Electorate: Look At The Data, Not an Imaginary Story

 

Data is the lifeblood of not only analysis, but also of conscience. The more information you have, the better informed your opinion can be, presumably. For Jennifer Donohue of the Huffington Post, facts should be disregarded in lieu of a good story, which most often insults anyone who thinks rationally about issues.

Jennifer Donahue supported Sen. Clinton in the primaries, and seems to be upset that she did not win. That’s obviously a fair opinion to have - I have close relatives who feel the same way. But she seems intent of yelling about it at odd junctures. For instance, she wrote the following one month ago,

Young voters, 18-29 years old, are registering at a higher rate than any other age group. They gave Obama the nomination, despite consistent national polling that showed Senator Clinton ahead from the start.

Now the same polling shows a dead heat, after a month of showing McCain ahead. The only way to understand current polling data is to discard it.

Young voters surely helped Obama, but Clinton did not hold leads throughout the process: the Gallup tracking poll was very much split. In the end, the accepted opinion from even inside the campaign is that the failure to contest caucus states cost Obama.

But it’s not just that. The article is about cell phone voters. There’s been some very good stuff done by Pollster and Nate Silver on cell phone voters. But 500 words about how the youth will all send a protest vote on the economy is both useless and pointless, and it seems gratuitous to drag Clinton’s defeat as a jumping off point.

Finally and most importantly, her conclusion is absurd. Disregarding poll numbers is a trap. They should not be disregarded, they should be adjusted, or considered within the margin of error. Polls are not an all or nothing affair - either being accurate or being useless. Quite the opposite. Each poll has a use, but factors including the sample size, question wording, target audience, etc. must be considered. Again, I point to Pollster and Silver, and a number of others. (Tom Jensen’s PPP blog is very informative.)

She continued the trend of marginalizing the data today:

There is still a gender split in New Hampshire. Husbands are shaking their heads, annoyed by what they feel is the hijacking of McCain’s message by the far right. But sleeping next to them are their wives, who shout Sarah and “drill, baby, drill” at the same rallies they drag their husbands to.

These women are not Democrats. They are Republicans and “Undeclared” voters. Four out of five Democrats in New Hampshire have lined up behind Obama, regardless of gender. If McCain were to win New Hampshire, much of the credit would have to go to Palin for energizing Republicans and independent voters.

She’s apparently a big fan of identity politics, which I loathe on face. Moreover, the data just isn’t there to support her strange assertions.

From the UNH poll today, which Jonathan Martin calls the best around. Here are some statistics, aside from the topline numbers showing Obama up 54-39.

Favorability ratings:

Candidate	Favorable	Unfavorable
Barack Obama	61%	29%
Joe Biden	54%	27%
John McCain	48%	42%
Sarah Palin	39%	48%

The poll also broke it down a bit demographically (download the report from the page linked above):

6. Obama and Biden get their highest favorability ratings from Democrats, young voters, those with higher levels of education and proponents of legal abortion.
7. McCain and Palin get their highest favorability ratings from Republicans, older voters, religious voters, and abortion opponents.

To hear Donahue tell it, everyone pails in comparison to the popular Palin. In reality, she’s consistently the candidate with the worst favorability ratings. All that is true is that she has excited the base. That has uses - more excitement will have more turnout. And it would be really useful in a 2004 type election that was only about the base. But that’s not this year. Women are smarter than to just side with Palin because of “drill, baby, drill.” In fact, I’m offended a bit at how Donahue portrays these women voters, who are apparently smarter than she thinks they are.

Specially regarding Obama’s favorability - since Donahue makes it the crux of her argument, here’s some breakdowns:

Demographic	Favorable	Unfavorable
Men	58%	32%
Women	64%	26%
Republicans	22%	59%
Democrats	91%	6%
Independents	65%	22%

Donahue says Palin is the powerhouse in the state. Her numbers:

Demographic	Favorable	Unfavorable
Men	44%	45%
Women	35%	51%
Republicans	77%	13%
Democrats	11%	77%
Independents	36%	43%

Barack Obama is more popular than Sarah Palin by far. Moreover, the lynchpin of her argument is that women are flattered by Palin but men are not. That is also incorrect - she’s consistently more unpopular among women than men is virtually every poll I have looked at.

What Jennifer Donahue does is not journalism. It’s homogenizing the electorate into one story that may or may not be true (in this case, it’s most certainly not true) for the sole purpose of telling a story to fit her agenda. The agenda in this case seems to be questions Obama’s appeal among independents or to promote implicitly the conception of identity politics. Unfortunately for Ms. Donahue, women in real life have not been drawn to candidates solely by identity politics, because they valued principles more. Now, when principles and idealism overlap, THEN you see the results of identity politics (I’m sure GOP women will turn out at a high rate this year, just like the African American vote).

But identity politics is not a component to swing voters. It’s a sweetner at the end.

Frankly, when I think of the GOP base, I don’t think of New Hampshire. It has gone to the GOP when the GOP seems more moderate. When it’s a base election (as in 2004) recently, it succumbs to it’s demographics and tends to be more Democrat. Point being, there’s very man reasons to think John McCain could compete there, under different circumstances. There’s very few reasons to think Sarah Palin in an asset in the state.

Why did she speak to bigger crowds? Because she appeals to the GOP base that was even to the right of McCain in the primary. That’s the reason. But you can’t get a long form essay out of that, so instead we get this hypothetical story about women being turned on like Rich Lowry by Palin. Shameless.

Sphere: Related Content

More on Tracking Polls

Nate Silver evaluates the growing list of tracking polls here. Interesting stuff.

Are Republicans Costing McCain The Election?

Answer: Not really, but they’re not helping.

Nate Silver dug up this fascinating data; I’m attaching his analysis as well because it’s so on point:

Support within own party:
Pollster             DEMS       GOP
Rasmussen            86         87
IBD-TIPP             88         83
Research 2000        87         89
ABC/Post             91         84
Zogby                87         84
Battleground         89         85

AVERAGE              88.0       85.3

2004 Exit Poll       89         93
2000 Exit Poll       86         91

Among Democrats, Barack Obama is now winning 88 percent support, comparable to John Kerry in 2004 or Al Gore in 2000. And there are a couple of points’ worth of undecideds left in there, so it’s possible that Obama could scrape up against the 90 percent number on election day.

By contrast, John McCain is winning the support of just 85.3 percent of Republicans, well down from Bush’s 93 percent in 2004 and 91 percent in 2000. There are some undecideds in there as well, so his numbers should improve some, but McCain is likely to underperform Bush by several points.

This is really the key theme of the whole post-Lehman Obama surge. Between his more populist talking points on the economy, the backlash to McCain’s attacks, and — I’m guessing here — a deep level of antipathy among Democrats toward Sarah Palin (Battleground has her favorability ratings at 12/78 among Dems), Obama has really brought the Democratic base home. By contrast, Obama’s support among independents varies quite significantly from poll to poll, ranging from essentially even in the Rasmussen tracker to a +15 in Zogby.

 

These numbers are just stunning. There’s two different ways to look at this for Republicans: The way that Silver portrays, that the party is running too far to the right in the current economic conditions to hold the party together. The converse lesson, and the one I suspect will become conventional wisdom in most conservative circles, is that the party did not run far enough to the right, or hard enough: if only had McCain taken a principled stand against the bailout  on conservative  grounds, maybe he would have had a chance. (Look for Newt Gingrich to especially drive that ppoint hard in Iowa over the next 36-40 months.) The best evidence of this is House Republicans, who are widely known to be wildly against the bailout. 

I think Silver is right and that hypothetical argument is wrong, though. Palin’s approval ratings are in the tank. Had McCain made a choice like Lieberman, his base would have yelled loudly at him but ultiately still voted for him, given the alternative. Obama has truly earned the wrath of Republicans. Had McCain picked a candidate that brought credibility among moderate Republicans and indepdents, he still might be in this. Picking Palin at the end of the day was about a base election. If McCain could run a moderate campaign based on personality and Palin could rally the base, that sort of campaign may have been successful. But the financial crisis actually forced the election to be about issues, and it collapsed the carefully constructed House of Cards. Furthermore, Sarah Palin never had a chance of winning all the most dead end Hillary Clinton supporters. Silver makes the accurate point that it gave a lightning rod for Democrats to rally against. If anything else, it also forced Hillary Clinton out on the trail even more for Obama, which has really helped Obama consolidate support. (And erased any suggestion that she would cost Obama the campaign.)

In short, I think the leftward drift of independents would have probably cost McCain the election no matter what. But the financial crisis plus Sarah Palin will probably produce a landslide.

Brett Farve throws a lot of touchdown passes by being reckless. But he also throws a lot of interceptions.

Sphere: Related Content

Tracking Polls: What They Can and Cannot Do

One of my biggest peeves this election has been people using tracking polls to give analysis on a day to day basis. 

Aaron Zelinsky of the Presidential Debate Blog:

 The latest Zogby poll shows Obama widening his lead to over five percent, beginning to answersCleve’s question: Colin Powell might be trumping Joe the Plumber.

I’ve met Aaron before (or at least I have seen him debate); he’s a very smart guy. And this problem is not limited to Aaron; it’s endemic to the blogosphere where data + information generally equals analysis, even if it doesn’t hold up to intellectual rigor. But he and others apparently do not understand how these polls work. First, he ignores every other tracking poll, which show conflicting results for the days he’s looking at. If anything, Rasmussen and Gallup are generally more reliable and have a longer history. Secondly, he ignores the fundamental problems of the Zogby tracking poll. Third, he ignores the structure of a tracking poll: what it does is take a three day rolling sample, and every day drops the oldest day for the newest day. Thus, a drop from one day to the next reflects not necessarilly a gain on a particular day, but also potentially a bad day of polling being dropped out.

Moreover, there has to be some external context for using all polls. It’s pretty obvious, for instance, that the CBS poll last week that had Obama up 53-39 was an outlier, and the race is much closer. Obama’s own people admit that there’s no chance of winning the popular vote by more than 5%. It’s not really possible to get a firm grasp of the polling numbers on any given day - just a range. And last week, Obama’s lead was probably 6-8%. This week, it’s probably 5-7%. The catalyst for that is not Joe the Plumber or a line by McCain or anything - It’s the race naturally tightening, as everyone predicted it would. (Literally, everyone.)

I like tracking polls. I think they’re good at showing the state of the race at this point. But they’re terribly ill-suited to showresponses by the electorate to any one singular event, unless the response is overwhelming. (I’m sure a tracking poll of presidential approval rating could definitely show change after 9/11 for instance). 

The tracking polls cannot show the response to a phrase of John McCain, or even a singular debate in today’s climate. (Again, barring a historically horrendous performance by one or the other candidate, which definitely did not happen.) The fact of the matter is that at this point in the race, both candidates are solidifying some previous weak and undecided support. Do not use tracking polls to assign that gain to random events. 

Related at 2008 Central:

• Why Is Zogby’s Tracking Poll Closer Than All Others?
• Gallup Tracking Poll and Likely Voter Models
• On Polls, Bitterness, and Clinging…

Sphere: Related Content

Why Is Zogby’s Tracking Poll Closer Than All Others?

Recently, conservatives looking for evidence of a McCain bounce have cited the Zogby tracking poll, including Drudge and Hugh Hewitt.

Nate Silver explains the difference in methodology:

The national tracking polls are actually in pretty good agreement with one another, with IBD/TIPP, Research 2000, Gallup, and Hotline all settling in the 5-7 point range. Zogby is the outlier at Obama +2.7, and that’s because Zogby has the odd practice of fixing his poll’s party identification weights based on what they were in the last presidential election. In Zogby’s world, then, it’s still 2004, when there were roughly as many Republicans as Democrats. Although Zogby’s trendlines may be worth looking at, his topline numbers are basically unusable.

Party ID is one of the most controversial things a poll can do. But fixing it only on the last presidential election is a questionable - at best - practice. It’s widely known the number of Democrats has increased the past four years and the number of Republicans has decreased.

Sphere: Related Content

Electoral Projections and Racism

Racism is obviously a terrible thing, but there’s something about racists who give a low priority to racism that’s somewhat amusing. There’s a number of examples, but to wit:

So a canvasser goes to a woman’s door in Washington, Pennsylvania. Knocks. Woman answers. Knocker asks who she’s planning to vote for. She isn’t sure, has to ask her husband who she’s voting for. Husband is off in another room watching some game. Canvasser hears him yell back, “We’re votin’ for the n***er!”

Woman turns back to canvasser, and says brightly and matter of factly: “We’re voting for the n***er.”

In an economy like this, I’d put aside stuff like the Bradley effect. That doesn’t mean Obama will definitely win, but most of the test races with the Bradley effect or the reverse Bradley effect did not have the importance of this election.

That said, the preliminary evidence is that the race has tightened ever so slightly this week, about to where I thought it would. As such, we’re not running a new Electoral Projection - the one last week still holds.

Sphere: Related Content

Gallup Tracking Poll and Likely Voter Models

One of the big questions regarding polls is projecting who is a likely voter and who is not. Gallup is now using two separate models. Time’s Jay Carney:

In an effort to measure the impact of Obama’s candidacy on the electorate, especially among young and minority voters, Gallup has begun publishing presidential tracking poll results based on two different “likely voter” models. The first is the traditional model, which designates likely voters based on how they answer questions related to both their current intention to vote as well as their past voting behavior. In today’s tracking poll, Obama registers a seven point lead over McCain using this model. The second model, Likely Voters II, designates likely voters based solely on their current voting intention, and in this way attempts to account for the intensity of interest, and mass of new voter registrations, among groups that typically do not vote in high numbers, including the young and racial minorities. Using that model, Obama has a 10-point lead. One of the big unknowns in this cycle has long been what the electorate will look like on Election Day. Gallup deserves credit for trying to apply a model that accounts for the electorate’s likely new complexion. It’s also understandable why they’re hedging their bets by going with two models, because we simply can’t be sure who will vote.

Looking at the models, I’d be more inclined to find the truth, as usual, somewhere in the middle. But publishing both models gives poll watchers the best information possible. So kudos to Gallup for providing more information.

Here’s the complete tracking poll. As usual, take th emargin of error seriously. One point or two points is utterly meaningless in a national survey.

 

Sphere: Related Content

2008 Central Election Projection

[LOOKING FOR ELECTION RESULTS? Then Check Out Our ELECTION NIGHT LIVE BLOG]

[This projection is outdated.  For the most recent projection published Oct. 24, go here.]

This is a projection of what will happen in November, and not a prediction of what would happen if the election were held today, tomorrow, yesterday, or any day in between. You’ll notice only one state changed its color from last week - Florida. A few other states have changed their categories though.

McCain Safe: 154 Electoral Votes
These states are not being contested by Obama now, and show no signs of being contestable by election day. Any one of these states flipping would cause absolute shock and consternation to the McCain campaign. There’s no surprises on this list. No changes here, but I am considering moving Georgia to McCain Lean. First, the Senate race there is tightening. Second, Nate Silver gives a compelling analysis of how black voter registration may signficantly be closing the gap in way polls are not meant to indicate. However, it’s intrsuctive to note that Georgia will in no way be a swing state, and McCain’s base there is very strong. Also, Obama is not competing in the state. So it stays here.

Alabama (9)

Alaska (3)

Arizona (10)

Arkansas (6)

Georgia (15)

Idaho (4)

Kansas (6)

Kentucky (8)

Louisiana (9)

Mississippi (6)

Nebraska - West Congr. Dists.  and Overall (4)

North Dakota (3)

Oklahoma (7)

South Carolina (8)

South Dakota (3)

Tennessee (11)

Texas (34)

Utah (5)

Wyoming (3)

Obama Safe: 206 Electoral Votes
These states are not being contested by McCain now, and show no signs of being contestable by election day. Any one of these states flipping would cause absolute shock and consternation to the Obama campaign. No changes here.

California (55)

Connecticut (7)

Delaware (3)

District of Columbia (3)

Hawaii (4)

Illinois (21)

Maine - Southern Congr. Dist. and Overall (3)

Maryland (10)

Massachusetts (12)

Michigan (17)

New Jersey (15)

New York (31)

Oregon (7)

Rhode Island (4)

Vermont (3)

Washington (11)

McCain Lean: 9 Electoral Votes

These states are being contested by the Obama campaign, but are unlikely to flip by election day. McCain continues to poll between a 5 and 13 point lead in Montana. I have not read about Obama pulling out, but there’s almost no chance Obama wins the state. West Virginia, on the other hand, is again confounding expectations. A single ARG poll in West Virginia gave Obama an eight point lead; Nate Silver is skeptical - as the result is wholly out of line with other polls of the state recently. While Obama is on the air in the state, and presumably has a ground game there, he has not been going to the state to campaign. West Virginia will not decide the election, but it could move into tossup territory if more polls follow ARG’s lead.

Montana (3)

Nebraska - Omaha Congr. Dist. (1)

West Virginia (5)

Obama Lean: 54 Electoral Votes

These states are being contested by the McCain campaign, but are unlikely to flip by election day. I’ve already addressed why Minnesota is here. The newest state here is Pennsylvania. Obama has consistently polled well in Pennsylvania for the past few weeks, without McCain having a lead in any single credible poll. Moreover, most recent polls have given Obama a double digit lead. That’s a big uphill battle for McCain, not just a “tossup.”

McCain is still competing hard in the upper midwest - Minnesota, Wisconsin, and Iowa, with diminishing results. Rumors from field workers in Wisconsin have McCain on the brink of pulling out, even though his campaign held two events there yesterday. Moreover, he’s still going to Iowa, a move that no one seems to understand. He’s also back in Minnesota this afternoon. The McCain camp seems intent on stealing one of these three states, but spending resources and time among all three strikes me as being the leas efficient way possible. I’d recommend that McCain pick one and spend a lot of time in that state (either Wisconsin or Minnesota) and forget the others.

Iowa (7)

Maine - Northern Cong. Dist. (1)

Minnesota (10)

New Mexico (5)

Pennsylvania (21)

Wisconsin (10)

Tossups: 113 electoral votes

I’m making a few assumptions in this part that not all projections are making.

First, that the race is going to tighten down the stretch. If that doesn’t happen, all of the following states are liable to go to Obama, as well as perhaps some of the above.

Second, I am assuming that A McCain bounce over the last month wins him a state that right now Obama would be favored in. If a bounce appears unlikely, future updates will be more aggressive. I find it hard to believe that McCain gaining a point or two nationally would not indicate something moving for him in Ohio, for instance.

Third, I am assuming states whose populations are static are more unlikely to flip from Republican to Democrat or vice versa than states who have seen significant population growth during the past four years.

Lastly, I’m going to err on the side of being conservative (the non-political use of the word). If I’m wrong, I’m most concerned with still calling the right winner in the election, and just being conservative regarding the margin of victory.

Going by the above numbers, McCain needs 107 electoral votes from these tossup states. Obama needs only 10 electoral votes.

McCain tossups: 57 electoral votes.

Indiana (11)

Missouri (11)

North Carolina (15)

Ohio (20)

I still think Missouri is a tease for Democrats. And that is underlined by Obama not spending a lot of time there lately. Obama is at the nadir of his polling so far, and the state is a dead tossup, with recent polls going anywhere from slight lean to Obama to slight lean for Obama. Looking at Missouri results both in the McCaskill Senate race in 2006 and in the primary race, it seems the best case for Obama is a race that is undecided and depends on the late returns from St. Louis to overturn McCain’s edge out in what Chuck Todd incessantly refers to as “Missoura.” (Not to hate on Chuck Todd, I’m just sick of hearing about Missoura.) It’s certainly conceivable that Obama wins the state, but if he does, the announcement will come long after he’s won enough states to confirm that he won the election. We will probably not be staying up late waiting for Missouri results like we did in 2000 with Florida. In that vein, Missouri is interesting to watch, and absolutely crucial for McCain (Why is he not spending more time there? Two or three well timed appearances there could lockdown the state.), Missouri is just a bonus for the Obama campaign, that they’ll probably spend enough energy on to be competitive.

The other three states are in a similar boat. Obama has slight leads in Ohio and North Carolina right now, and Indiana is a dead heat or slightly leaning to McCain. If McCain does not close the gap on Obama, he’ll lose at least two of these three, if not all of them. But if he closes the gap at all, these would all probably lean to McCain at the very least. Obama’s massive edges in advertising have to be helping him.

Obama tossups: 58 electoral votes

Colorado (9)

Florida (27)

Nevada (5)

New Hampshire (4)

Virginia (13)

The big change is Florida. I said last week that I considered strongly giving Florida to Obama, and this week I did. The McCain campaign is showing absolutely no momentum within the state, with more rumors of dissension and anger in the GOP camp emerging by the day.

Here’s Pollster.com’s chart of Florida Polls:

It’s becoming increasingly obvious that the economy is driving Obama’s numbers; meanwhile, the McCain camp is talking about anything but the economy.

Obama is also outspending McCain five to one in the state, and has sent his top field generals to the state. Tom Jensen of PPP looked at white voters in the state; Obama is up strongly among Hispanics and black voters.

New Hampshire may be on the verge of becoming Obama lean in this column. After a couple weeks of either candidate leading by a single percentage point or so, McCain has not polled well there at all in the past week, and has only one lead among the most recent results of every poll. If there’s not movement towards McCain over the next week in NH polls, it will probably be reclassified as Obama lean.

I’m still not sure how competitive the McCain camp thinks Nevada is. And I am not the only one wondering that.

Obama has gotten very good poll numbers in Virginia this past week, with three polls giving him at least an 8 point lead. Other polls show the race as closer. McCain will need both significant turnout in Virginia and some good fortune between now and the election will win it. Most importantly, he’s going to have to actually compete in the state. When he starts doing that, we’ll see what happens to the poll numbers. This state will probably be a tossup to the end in this column

Another state I will likely leave as tossup to the end is Colorado. The two most recent polls in the state both give Obama a 6 point lead, but that might be Obama’s ceiling in the state. Obama is outspending McCain in the state, and one blogger found bad news for McCain in the heart of GOP country there.

Conclusion:

The election is not likely going to be a deadlock or 269-269 tie anymore, and is increasingly likely to be a solid Obama win. If McCain does not find a message that resonates with voters, it could be a landslide. I’d recommend some sort of economic message first; he can return to the attack ads later. McCain 220, Obama 318.

Sphere: Related Content

Obama Up Big in Minnesota: Validates My Projection

The new Minneapolis Star Tribune poll has Obama up on John McCain 55 to 37. A SurveyUSA poll this week had McCain up one point. Nate Silver weighed similar differences among the Senate polls of the state, and found that both were using too extreme a sample in each direction.

Conclusion? Obama is up, just not by 18 points. Minnesota will break Democrat unless something even bigger than the economic crisis happens over the last month that is uniformly bad for Obama and good for McCain. And if such an event happens, Obama will lose all over, not just in Minnesota.

To look at more of my projections, check out our first of many electoral college projections.

Sphere: Related Content

Social Issues: Not That Important Anymore?

Looking over yesterday’s Washington Post poll, a few numbers stuck out. First, of all the issues surveyed, only one issue was below 40% in terms of importance to voters, that is social issues at 39%. *

Moreover, Obama is trusted more to handle social issues by a wide margin: 56% to 32%.

The poll found that a strong majority of people felt that Obama shared their values, 56% for to 39% against. By contrast, 47% found that McCain shared their values to 49% that McCain did not share their values. (Obviously, this included a number of conservatives who are doing the “holding their nose to vote for McCain” routine.)

These numbers beg a question: is the era of social issues being a strong influence in elections over? Or is it just on pause because of economic problems and two wars overseas? I’d say the economic problems have a stronger influence than foreign policy problems, but after the high voltage campaigning in 2004 over social issues with little payoff for either conservatives or liberals, people seem to have just shrugged it off and moved on. Moreover, it’s clear that the next generation of voters is a lot more liberal/libertarian on these social issues than the previous one.

I think it’s possible that we’ve reached a tipping point where a lot of social issues mean a lot less. The one exception I would note is abortion: I doubt that will ever really go away as an issue; it will just fade in an out as a top issue.

(The only other issue one under 50% was VP choices; clearly the people understand how to ignore that, even if the media does not. And the third least important issue is gun control, right at 50%. A strong amount of people know that Heller has really settled that issue in most significant ways. Just noting that the wisdom of the people often surpasses that of the media.)

Sphere: Related Content

Massive Bounce for Obama in Ohio

March 19, PPP Ohio Poll: McCain 49, Obama 41

June 16, PPP Ohio Poll: Obama 50, McCain 39.

The crosstabs seem about the same in both. This is just a massive bounce, in three short months. I have a feeling it might be Obama’s high water mark in Ohio though.

Sphere: Related Content

Not-So-Shocking Development: Obama-McCain Demographics Differ From Obama-Clinton Ones

First Read today (my emphasis):

There was a growing consensus during the Democratic primary season that Obama was going to struggle with Latino voters — due to the exit polls, his race, and McCain’s immigration stance. In fact, in that now-famous conference call in which Hillary Clinton indicated that she would be open to serving as Obama’s running mate, that response was spurred by concern by New York Rep. Nydia Velasquez (D) that Obama was going to have trouble with Latinos. But it looks like that CW — at least right now — was wrong. In addition to our recent NBC/WSJ poll, which showed Hispanics breaking for Obama 62%-28%, a new survey of 800 Latino voters from 21 states finds that 60% of them plan to vote for Obama versus 23% for McCain. That is down considerably from the 40%-plus Bush received in 2004. It’s no longer fair to say that Obama has a problem with Latino voters; McCain does.

This should not be a surprise. We’ve talked about this in our podcast before, and I alluded to it a few months ago. The key point is that the Obama and Clinton race revealed sharp preferences within the party, but not necessarilly sharp problems.

For instance, Hillary Clinton did not have a general election problem with black voters. (Bill on the other hand…). Obama did not have a general election problem with Hispanics.

On the other hand, Clinton did have a problem in the northwest. It’s almost certain the Washington, Oregon, Minnesota, and Iowa would be hurt comparably to Obama if Clinton was the nominee. On ther other hand, Obama has real problems in Appalachia. Arkansas would have been at least a swing state if Clinton was nominated; now it is a long shot at best.

That said, there’s an infinite number of ways to divide the electorate, and many show preferences. Women understandably favored Clinton generally in the primary; Now, they’re favoring Obama, because the alternative is McCain. This is not rocket science.

If you can’t decide between a red truck and a blue truck, you don’t arbitrarily decide to buy a white station wagon. It’s only the people who feel strongly about the color red and the color blue that matter. And despite a lengthy, lengthy, lengthy primary, the research into that was somehow so scarce that the best analyst in the news is shocked that Obama is doing well with Hispanics?

Come on now.

Sphere: Related Content

Surprising Alaska Senate Poll Numbers

A lot of people thought Sen. Ted Stevens would be vulnerable in this election, both because of corruption charges and a sharp spike in his craziness in recent years. Still, seeing a new poll having Anchorage mayor Mark Begich beating Stevens by 7 points (51-44) is pretty shocking.

Given that Democrats are also favored in the House race, this should also be seen as a positive sign for Barrack Obama, and perhaps a sign that McCain and Governor Sarah Palin have work to do.

Sphere: Related Content

On Polls, Bitterness, and Clinging…

TNR’s Isaac Chotiner:

Yes, yes, we should not get carried away with overnight tracking polls, but today Obama is up 11 points in Gallup (his biggest lead ever) and 9 points in Rasmussen. Since Saturday–which was the first night that included any post-bittergate data–Obama has moved up 10 points in Rasmussen and 4 points in Gallup.

There are two possible conclusions: The first is that polls are meaningless. The second is that the Democratic electorate is completely reactionary. I vote for option 2! Tell Hillary to leave the race…and her polls go up. Attack Barack over Reverend Wright…and his polls go up. Perhaps Clinton’s only hope is a monumental scandal…about her!

I’m voting for option 1. These tracking polls really are showing nothing.

If anything, it’s a sign that scandals take time to permeate, especially ones driven by the media, without supporting video. (And especially when the only audio sounds like whales copulating.)

Brief thoughts on this scandal: It’s a typical Obama scandal. The words themselves are confusing and maybe meaningless, but there are very positive and negative ways to spin them. And you can’t spin all one way without looking like an idiot. (For instance, there’s definitely some bitterness about the economy, and no matter what Obama says, he probably did not mean religion in a positive way).

I’m taking it light, because there’s just not that much happening right now. Luckily, once Pennsylvania votes in a couple weeks, this long break will be over and the campaign will probably be over within a month.

Sphere: Related Content

UPDATED: General Election And Democratic Primary Polls

Our polls section has been updated to include the latest national Democratic Primary polls and we’ve also added polls for general election match-ups (McCain vs. Obama and McCain vs. Clinton).

Polling data will expanded in the coming weeks.

Sphere: Related Content

Russert: ‘In A Survey, 11 People Referred To Hillary As A Bitch, So We Have To Talk About It’

This morning on Meet The Press, Tim Russert reached a new degree in sexism and shoddy analysis with a tortured segment on a recent Pew Survey.

Russert’s introduction to the segment:

MR. RUSSERT: [...]

Then Pew said, “Give us a word that you think best describes these candidates.” The first one was McCain. Old was the first word people threw out. Honest, experienced, patriot, conservative, hero, liberal.

Then they asked, OK, Obama. Inexperienced, charismatic, intelligent, change, inspirational, young, new.

Clinton. Experienced, strong, untrustworthy, intelligent, smart, determined, rhymes with rich. That’s what the poll said. They used a different word. But it’s a family poll, and this is a family program. What does that tell you, James?

I’ll spare you the agonizing discussion that followed, but Read more

Sphere: Related Content

Next Page »