polls

A friend replied to my last email reacting to my inclusion of the Gallup daily tracking poll on the presidential race. He noted that the national polls understated the Obama advantage because the real race is the electoral college and Obama is way out in front when you look at polling of individual states. Specifically, he cited the electoral college tally at http://www.electoral-vote.com/

I agree with him on the limited value of national polling numbers at this stage in the campaign – which is why I have generally not been citing them. My point is this case was an observation based on a months-long polling trend: For a politician as well known as McCain, who has (or at least used to have) one of the best “brands” in American politics (the “straight-talking maverick” willing to reach across party lines) his base of support has been remarkably low and appears to be, if anything, a ceiling not a floor. It suggests that the “undecided” factor relates more to Obama than to McCain, which I view as a positive for Obama. Despite Obama still being somewhat of an “unknown”, he has been holding a pretty consistent lead over McCain. And I have no doubt that the images of him as an “acting president” abroad, holding substantive discussions with world leaders and drawing enthusiastic receptions everywhere he went, will help undecided voters “visualize” him in that role.

I also agree that it is the electoral college that counts and that the electoral college map at this point is much more favorable to Obama than the national horse-race numbers would seem to indicate. Indeed, for about a month now I have been taking bets that Obama will win with at least 300 electoral votes. The map at www.electoral-vote.com is an excellent one. It is similar to the map at pollster.com. Another site that has excellent analysis is http://www.fivethirtyeight.com/ (they probably had the best track record during the primaries – unlike virtually all polling, which attempts to convey an accurate snapshot of where the race stands at any given moment, the models on this site attempt to be predictive). It is hard to see how McCain puts together a winning electoral map. However, to some extent this reflects the fact that once a lead in the national polls gets into the 4 to 6 percent range, the electoral map will almost certainly reflect a blow-out (this follows from the “winner-take-all” nature of the electoral college system). And that is probably where Obama stands now. Pollster.com’s “poll-of-poll” composite currently has the race at Obama 47%/McCain 41.9%. If Obama ends up winning the general election by that kind of five-point margin, his electoral college win would be an absolute blow-out.

But I don’t want to get too distracted with polls at this point. The larger picture has me convinced that Obama will win comfortably. And it’s not just the record unpopularity of Bush and the Republican party generally, or the horrible state of the economy or all the other generic factors that favor a Democrat this year. It’s also Obama’s organization and “ground game” (which gets virtually no coverage by the corporate media, which prefers polarization, conflict and the kind of trivia that lends itself to opposing pundits shouting at each other – like the entirely fictitious story currently dominating the nation’s cables that Obama “snubbed” wounded troops in Germany).

The Obama campaign is attempting to register 10 million new voters (none of whom would be picked up in the pollsters’ “likely voter” models, of course). His supporters are vastly more enthusiastic than McCain’s. Given Obama’s huge leads among young voters, African-Americans and Hispanic voters, even very modest increases in turn-out among those groups could result in a blow-out. For example, Bush garnered 40% of the Hispanic vote. McCain will probably get only 30% or so – that difference alone could swing a few states, especially if Hispanic turn-out is strong (and I have confidence in the Obama campaign’s ability to turn out the vote – among other things, they are spending $20 million just on targeting Hispanic voters). In 2000 and especially in 2004, Bush relied to a considerable extent on evangelicals as his ground troops. But that group has been changing somewhat (as old folks are replaced by a younger generation), focusing less on abortion and gay rights and more on social issues and the environment. As a group, they really don’t like or trust McCain (and vice versa), and Obama speaks their language. Obama is not going to win among that group, but he can peel away a few percentage points (a few points here, a few points there, next thing you know you have a blow-out). It is enough for Obama that evangelicals are simply less enthusiastic in their support for McCain than they were for Bush.

I also think McCain is a much weaker candidate than the corporate media has been willing to acknowledge. Part of that is that they have had a long love affair with McCain and have assumed that the rest of the country shares it. (The corporate media tends to be cynical and they love a fellow cynic like McCain.) Part of it is that McCain has never really been tested in the way he is being tested now. (It’s easy to be a free-wheeling gadfly – a party of one. Harder to lead a party and have your every utterance scrutinized.) But McCain is not particularly smart and he is a horrible, undisciplined manager. He has absolutely no purpose to his campaign – no overarching themes or narrative to his campaign (other than the fact that he was a POW 30 years ago). His shallowness and lack of discipline is becoming apparent to even the most smitten observers. He can also be pretty nasty and has a mean temper – and Americans tend to vote for the candidate that they like best. McCain’s campaign has already gone 100% negative (reflecting the fact that key Rove operatives are now running his campaign). Clearly, the McCain campaign is a cornered dog – and its going to get really, really nasty.

As a young, smart African-American, Obama is hard for a lot of voters to relate to. But he is relentlessly positive, even-tempered and gracious, and I expect his “likability” factor to increase as voters get to know him better and contrast him with McCain.

So, take the polls with more than a grain of salt at this point in the campaign.

And check out this piece.

The Myth of a Toss-up Election

By Alan Abramowitz, Thomas Mann and Larry Sabato

"Too close to call." "Within the margin of error." "A statistical dead heat." If you've been following news coverage of the 2008 presidential election, you're probably
familiar with these phrases. Media commentary on the presidential horserace,
reflecting the results of a series of new national polls, has strained to make a case for a hotly contested election that is essentially up for grabs.

Signs of Barack Obama's weaknesses allegedly abound. The huge generic Democratic Party advantage is not reflected in the McCain-Obama pairings in national polls. Why, according to the constant refrain, hasn't Obama put this election away? A large
number of Clinton supporters in the primaries refuse to commit to Obama. White
working class and senior voters tilt decidedly to McCain. Racial resentment limits Obama's support among these two critical voting blocs. Enthusiasm among young voters and African-Americans, two groups strongly attracted to Obama, is waning. McCain is widely seen as better prepared to step up to the responsibilities of commander-in-chief. Blah, blah, blah.

While no election outcome is guaranteed and McCain's prospects could improve over the next three and a half months, virtually all of the evidence that we have reviewed - historical patterns, structural features of this election cycle, and national
and state polls conducted over the last several months - points to a comfortable
Obama/Democratic party victory in November. Trumpeting this race as a toss-up,
almost certain to produce another nail-biter finish, distorts the evidence and does a disservice to readers and viewers who rely upon such punditry.

Consider the following.

Except for a few days when the Gallup and Rasmussen tracking polls showed a tie, Barack Obama has led John McCain in every national poll in the past two months. Obama's average margin has consistently been in the 4-6 point range during this time. By contrast, the polls in 2000 and 2004 showed much more variation over time.

State polling data have also consistently given Obama the advantage. According to realclearpolitics.com, Obama is currently leading in 26 states and the District of Columbia with a total of 322 electoral votes; McCain is currently leading in 24 states with a total of 216 electoral votes. Obama is leading in every state carried by John Kerry in 2004 along with seven states carried by George Bush: Iowa, New Mexico, Ohio, Indiana, Virginia, Nevada and Colorado.

Obama is leading in 11 of the 12 swing states that were decided by a margin of five points or less in 2004 including five of the six that were carried by George Bush. And while Obama has a comfortable lead in every state that John Kerry won by a margin of more than five points in 2004, McCain is in a difficult battle in a number of states that Bush carried by a margin of more than five points including such solidly red states as Indiana, Montana, North Dakota, Virginia, and North Carolina.

And remember these June and July polls may well understate Obama's eventual margin. Ronald Reagan did not capitalize on the huge structural advantage Republicans enjoyed in 1980 until after the party conventions and presidential debate. It took a while and a sufficient level of comfort with the challenger for anti-Carter votes to translate into support for Reagan. If Obama's performance over the last eighteen months is any guide, a similar pattern is likely to unfold in 2008.

Aside from the horserace results, there is evidence of a growing Democratic Party advantage in the electorate. A recent analysis by Rhodes Cook of voter registration data in 29 states and the District of Columbia that permit registration by party
shows that since November of 2004, Democratic registration has increased by
almost 700 thousand while Republican registration has declined by almost one
million.

Democrats now enjoy a substantial lead over Republicans in voter identification. According to the Gallup Poll, the two parties have gone from near parity four years ago to a 12 point Democratic advantage in the first half of 2008. And polling data continue to show that Democrats are more satisfied with their party's nominee than Republicans voters and more highly motivated to vote. While Republicans normally benefit from higher turnout among their supporters, that may not be the case this year.

In order to defeat Barack Obama, John McCain will have to convince a lot of disgruntled Republicans to turn out and vote for him. But mobilizing the Republican base, a strategy employed successfully by Karl Rove in 2002 and 2004, won't be enough for McCain to win in 2008. He'll also have to convince a majority of independents and a substantial number of Democrats to vote for him. That's a task that proved too difficult even for Rove in the 2006 midterm election and it may be even more difficult in 2008. That's because since 2006 the political environment has gone from bad to worse for Republicans.

It is no exaggeration to say that the political environment this year is one of the worst for a party in the White House in the past sixty years. You have to go all the way back to 1952 to find an election involving the combination of an unpopular president, an unpopular war, and an economy teetering on the brink of recession. 1952 was also the last time the party in power wasn't represented by either the incumbent president or the incumbent vice-president. But the fact that Democrat Harry Truman wasn't on the ballot didn't stop Republican Dwight Eisenhower from inflicting a crushing defeat on Truman's would-be successor, Adlai Stevenson.

Barack Obama is not a national hero like Dwight Eisenhower, and George Bush is certainly no Harry Truman. But if history is any guide, and absent a dramatic change in election fundamentals or an utter collapse of the Obama candidacy, John McCain is likely to suffer the same fate as Adlai Stevenson.

Abramowitz is a professor of political science at Emory University. Mann is a senior fellow at The Brookings Institution. Sabato is a professor of politics at the University of Virginia and director of its Center for Politics.