Going by the measures used to predict US presidential elections, Obama should easily beat McCain - so why is this likely to be an unpredictable contest? asks Denis Staunton, Washington Correspondent.
WHEN BARACK OBAMA appeared before 18,000 people in St Paul, Minnesota, this week to announce he had clinched the Democratic presidential nomination, he was met with such a deafening, joyful roar that it was minutes before he could begin his remarks. When it came, his speech was exquisitely written and flawlessly delivered, making almost as eloquent an impact on millions of television viewers as on the rapturous supporters before him.
A few hours earlier, John McCain appeared in Louisiana before a much smaller crowd, standing before a green backdrop that gave him a sickly pallor as he struggled with the autocue, punctuating his sentences with an unlikely, mirthless grin.
The contrast seemed to many in Washington to anticipate the nature of the forthcoming contest between a young, charismatic Democrat at the head of the most meticulously organised and lavishly funded campaign in American history and a septuagenarian Republican leading a demoralised party, strapped for cash and tethered to a politically toxic president.
By all conventional measures, this ought to be a spectacular year for the Democrats as the economy worsens, the Iraq war drags onwards and President George Bush's unpopularity breaks new records each month.
DESPITE ALL THIS, tracking polls have shown Obama and McCain effectively tied in recent weeks and some senior Democrats fear they could be heading for defeat in November at the hands of a Republican candidate with unique crossover appeal.
Political scientists traditionally use three measures of the national political climate to predict US elections: the incumbent president's approval rating in the middle of the election year, the growth rate of the economy during the second quarter, and the length of time the president's party has held the White House.
The higher the president's approval rating and the stronger the growth rate of the economy, the more likely it is that the president's party will win, but if that party has controlled the White House for two terms or longer, the urge for change makes victory less likely.
Bush's abysmal approval rating, sluggish economic growth and eight years of a Republican in the White House suggest that McCain should lose by one of the biggest landslides since the second World War.
On the three biggest issues for voters - the economy, the Iraq war and healthcare reform - the Democrats are closer to the mainstream of public opinion than the Republicans. But as Bill Galston, a senior fellow at the Brookings Institution, points out, US presidential elections are not only about policies but about people.
"The vote for president is the most personal act that an American voter commits, and it inevitably revolves around the personality and the character and the leadership style of the two contending candidates," he says. "On a purely personal dimension, Senator McCain has an important advantage, namely that many, probably most Americans, think they know him pretty well. They think they know who he is, what kind of human being he is. Senator Obama, by contrast, is much less known. For most Americans, he's still virtually a blank slate."
Galston, who was a White House policy adviser under Bill Clinton, argues that a unique and unpredictable feature of this year's election is the stark generational contrast between one candidate a generation older than the current president and another a generation younger.
"John McCain came to maturity in the 1950s and his discourse and demeanour reflect that fact. And it's going to be interesting to see how many Americans are going to be able to connect with a vocabulary and an outlook that are rooted in a very different era," he says.
"I think people are always going to respect John McCain and the sort of martial virtues that he invokes and to some extent instantiates, but whether that outlook is going to be seen as directly relevant to current concerns . . . remains to be seen. I'm a bit sceptical, to tell you the truth. If campaigns are always a battle between the past and the future, I think you'd have to say: clear advantage to Obama on that one."
FOR ALL THE enthusiasm surrounding Obama's presidential bid, he struggled through the second half of the primary season, wheezing over the finish line after losing most of the final contests.
His poor showing among older voters, Catholics and poorer, less educated white voters has raised fears about his viability in some of the big, traditional battleground states of Ohio, Michigan and Pennsylvania, and some Democrats have already written off his chance of winning states such as Kentucky and West Virginia.
Up to 20 per cent of voters in some states have told pollsters that Obama's race is a factor in how they view him, while others perceive him as an exotic, metropolitan elitist.
Harwood McClerking, a political scientist at Ohio State University who studies the role of race in US politics, says studies of stereotyping suggest that Obama's race makes him more vulnerable to negative perceptions. "It makes it easier for opponents to paint him with these broader brushes and these more negative ideas. I think they cling to him in a more adhesive fashion because he is an Other, he's so different," he says.
McClerking believes McCain would be "repulsed" by the idea of using racial animus against Obama but he predicts that independent groups will use coded racism to raise doubts about the Democrat, seizing on issues such as Obama's admission that he used drugs as a young man.
"The problem is that, while America is prepared to look past these things for some politicians, drug use or any other kind of criminal behaviour is part of the cultural stereotype about blacks," McClerking says.
"I'm expecting his opponents to take advantage of the American cultural idea of what black means. Now, they're not going to do it openly. They're not going to say: don't vote for Barack Obama because he's black. Because there are tremendously strong norms in America against using that type of discourse. You simply cannot say that and win votes. But I do expect them to use these other ideas about blackness and to hint at them with the hope that these things will stick as an undercurrent so that people are more likely to remember the negative charges and to use them in the voting booth."
Obama's candidacy has energised millions of young voters across the US and raised expectations about his presidency that could be difficult to fulfil. It is African-Americans, however, who have invested the most hope in Obama's victory in November, an investment McClerking fears has its own perils.
"He's considered the most viable black candidate ever and many people believe he can win. They see him winning as a sign of racial progress, just like the Civil Rights Act of 1964, the Voting Rights Act of 1965 and various other things that were seen as signs of success and change in American norms on race. A black president would be seen as a sign that blacks have truly got over the hump and that this is a more open society," he says.
"The flipside is that, if his victory is a near-ultimate sign that blacks have made it, his defeat is a sign that blacks, as yet, have not made it. There's a tremendous amount of hope and anticipation and political interest is higher among African-Americans now than it's been for years, and I think if he were to go down to defeat, this would be alienating and push African-Americans away from the process."
Recent presidential elections have been won in a handful of battleground states, notably Florida and Ohio, both of which Bush won in 2004. McCain is confident that Obama's weakness among Hispanic voters, Jewish-Americans and older voters will ensure that Florida remains in Republican hands.
Polls show Ohio shifting back and forth between Obama and McCain while the Republican is currently ahead in Michigan and only a few points behind in Pennsylvania. McCain believes he will also be competitive in states such as Wisconsin and New Hampshire, which John Kerry won in 2004. "It's going to be tough to win the presidency without winning one of Florida and Ohio and, given the fact that Obama seems particularly weak in Florida, if I were creating an electoral college strategy for his campaign I would put a lot of chips on the Ohio square before the roulette wheel starts spinning," Galston says.
"There's both a defensive and an offensive point. The offensive point is that in modern times, going back more than 100 years, no Republican has been elected president without carrying the state of Ohio. So if Obama can deny Ohio to McCain, he has an excellent chance of being president. The worrisome side - I hope the Obama folks are paying attention to this - is the weakness in Michigan. That is simply not a state that Democrats can afford to lose."
OBAMA IS CONFIDENT that, even if he loses Florida and Ohio, he can win the 270 electoral college votes he needs in November by redrawing the electoral map, winning western states such as Colorado and New Mexico, and taking Republican strongholds in the south such as Virginia, Georgia, North Carolina and South Carolina. He also hopes to pick up Iowa and perhaps Indiana, both of which border his home state of Illinois.
Galston thinks Obama should win Colorado, where the governor, both houses of the state legislature and one of its senators are Democratic, and that he will pick up other western and mountain states. He thinks Obama's southern strategy is not entirely fanciful but predicts the Democrat will fall short.
"I happen to think that McCain's military background makes him a stronger candidate in Virginia, which has a lot of retired military people, particularly in the east and south. I wouldn't bet a lot of money on Obama carrying Virginia," he says.
"Nor am I persuaded that he really has that good a chance in states like North Carolina and Georgia, which the Obama campaign has also mentioned as targets.
"There will be an outpouring of African-Americans, but as I do the sums on the back of an envelope I don't think that's going to be enough to overcome the huge edge that Republicans have in those Deep South states."
Bush's victories owed much to Karl Rove's strategy of engaging local groups, often linked to churches, to canvass their neighbours on the Republican's behalf. Rove maintains that Democrats have lost numerous contests because they bussed in young volunteers from out of state to knock on doors, while Republicans sent their message through friends and neighbours.
Evangelical Christian groups and the National Rifle Association provided extensive networks for Bush, but both groups are less enthusiastic about McCain, who is viewed with suspicion by many conservatives.
OBAMA HAS BUILT a new system of local networking, using information technology to create remarkably effective ground organisations in every state. These will help with voter registration in the coming months and in turning out the vote in November.
"This isn't just a new technology. It brings with it an entirely different strategy, a different model," Galston says. "It means that rather than asking or directing people to do things from the top down, you can create incentives from the bottom up for people to do things for themselves that otherwise campaigns would have to do for them, traditionally always have.
"Not only does that build participation and enthusiasm at the grassroots level, it saves a whole lot of money."
Both parties' primaries defied every prediction, with McCain bouncing back from the near meltdown of his campaign to capture the Republican nomination and Obama coming from nowhere to defeat Hillary Clinton.
Calculations about the campaign over next five months could be upended by external events, such as a foreign-policy crisis or a terrorist attack, or by internal shocks such as a personal scandal.
Galston believes that, in the end, the election may be decided by a group of voters who have not yet made their voices heard.
"I think there's a whole bunch of people, about 10 per cent of the electorate, hanging back, trying to figure out whether they can develop enough trust and confidence in Obama to take a chance on him and if they all answer that question in the affirmative, as a similar 10 per cent did in 1980, when the question was Carter versus a somewhat riskier appearing Reagan, then it might not be that close an election."