Ed Brayton is a journalist, commentator and speaker. He is the co-founder and president of 
So says the London Telegraph, which cites Obama insiders as saying they think he's going to win big with well over 300 electoral votes. They think this because they're convinced that they have the upper hand in the "ground game" of turning out their constituencies to the polls, especially young people and poor and minority voters, who typically vote in lower numbers than other groups.
Barack Obama's senior aides believe he is on course for a landslide election victory over John McCain and will comfortably exceed most current predictions in the race for the White House.Their optimism, which is said to be shared by the Democratic candidate himself, is based on information from private polling and on faith in the powerful political organisation he has built in the key swing states.
Here's the reason for that optimism:
But his aides are convinced that he has a strong chance of winning no fewer than nine states won by George W.Bush in the closely contested 2000 election, including former Republican strongholds like North Carolina, Virginia and even Indiana, which have not voted Democrat for a generation.David Axelrod, Obama's chief strategist, said last week that Obama had "a lot of opportunity" in states which Mr Bush won four years ago.
But in private briefings in Washington, a member of Mr Obama's inner circle of policy advisers went much further in spelling out why the campaign's working assumptions far exceed the expectations of independent observers.
"Public polling companies and the media have underestimated the scale of new Democratic voters registration in these states," the campaign official told a friend. "We're much stronger on the ground in Virginia and North Carolina than people realise. If we get out the vote this may not be close at all."
"If we get out the vote" is the key, of course. AP reports on some of the ways the Obama campaign is working to make sure those folks actually show up at the polls.
Obama's campaign plans to send staffers to Ohio State University, with about 53,000 students, to encourage early voting among the sometimes hard-to-engage young. Polls show Obama easily carrying the demographic against McCain.Obama's campaign plans to arrange concerts with John Legend, who will hold early voting rallies on Monday in Columbus, Springfield, Dayton and Cincinnati. For campuses far from election boards, Obama aides are organizing car pools to make sure students get to the polls.
On Saturday, Obama's campaign targeted the 100,000-strong college football crowd at Ohio Stadium with a flyover advertising early voting and visits from staffers to surrounding bars.
The Northeast Ohio Coalition for the Homeless believes it can round up 2,000 people from homeless shelters in the Cleveland area and get them to polling places. A donated van will shuttle voters from two shelters, including the city's largest downtown. Seventeen other shelters are providing their own transportation.
Project Vote will go door to door in minority neighborhoods and use vans to transport people to election sites in Cleveland, Cincinnati, Columbus, Dayton and Toledo.
If those constituencies, which tend to support Obama in very high numbers, do show up in higher numbers than in the past it could make a decisive difference in a state like Ohio, where the polls are showing a dead heat and the last two elections have been incredibly close. Either way, I think the Telegraph article is exactly right about the prospects for the two campaigns winning:
The senior Obama advisor said that the Democratic nominee is confident of winning all the states held by John Kerry, the Democratic candidate four years ago, a total of 252 votes.But his team believes he can also bank victories in Iowa, where he first emerged as a force in the campaign in January, and New Mexico, where Mr Kerry only lost by 20,000 votes in 2004. Those states would leave him just six votes short of outright victory.
Taking Colorado, as Mr Obama's team are very confident of doing, would put him over the top. Even winning the smaller state of Nevada, with its five electoral votes, would be enough to guarantee a 269-269 tie with Mr McCain. If that happens, the US consititution would hand the decision over to the Democrat dominated US House of Representatives, which would presumably come down in Mr Obama's favour.
Most pollsters would regard those expectations as uncontroversial. But the Obama camp is also confident of winning Ohio and Virginia, which commentators believe are "toss up" states with the two candidates chances at 50/50.
Last week Mr Obama began investing heavily in advertising in Indiana, Florida and North Carolina, which many had supposed to be a waste of time and money.
A Washington official who has discussed the electoral mathematics with one of Mr Obama's senior advisers told The Sunday Telegraph that the campaign is spending money only in states which it believes can, and indeed ought to, be won.
"Obama has many more paths to the nomination than McCain," the source said. "They think they can defend the Kerry states. Iowa is gone. That's five votes. New Mexico is in the bag. Then Obama has four or five different ways of winning. He can go Nevada or Colorado, Virginia, any of those, even Indiana.
"McCain has got to run the board, the whole Bush table. He can probably lose New Mexico and Iowa. He can't afford to lose anything else."
The official added: "The poll numbers say Florida's back in play. McCain hasn't spent a single penny there and that's Obama's calculation, that he can capitalise on that. The Republicans can't lose Florida or they're done for."
All of this echoes what I wrote in my poll update for the Michigan Messenger last week. The margin in 2004 was 286-252. I don't think Obama is going to lose any of the states that Kerry won. The gap in Michigan is widening for Obama at this point and I can't imagine Wisconsin and Minnesota going red despite the relatively close polling numbers currently.
Of the red states in 2004, Obama has a virtual lock on Iowa and New Mexico, but that would still leave it 274-264 for McCain. But that's close enough that any other state would put Obama over the top. Colorado, which is increasingly looking like it's going to go for Obama (he has an average lead of more than 5 points in the polls right now), would put him over the top. So would Nevada. And if he wins Ohio, Virginia or North Carolina, forget it - there's no way McCain could win if he gets any of those no matter what happens elsewhere. And right now, he's winning in every one of those states.
Politics
Comments
"The London Telegraph"
*raises an eyebrow*
Not heard it called that before. :)
Posted by: Andrew | October 3, 2008 9:32 AM
FiveThirtyEight.com puts it in the simplest possible terms. If Obama carries just the states where he currently leads by 6 points or more, the EC ties, the election goes to the House, and Obama wins. If he carries any other state, he wins outright.
Of course, a lot can change in a month.
Posted by: Count von Count | October 3, 2008 9:37 AM
Before getting too euphoric over the apparent positive poll numbers, we should recall "President" Tom Dewey who was anointed a month before the election by the pollsters. At the time, President Harry Truman wasn't a whole lot more popular with the voters then George Bush is now.
Posted by: SLC | October 3, 2008 9:39 AM
SLC,
Granted we should always be aware of a "Dewey Defeats Truman" moment, but polling is rather more sophisticated now, and unlikely to error outside the error bars, especially when multiple polls point the same direction.
That said, there are enough undecided voters to keep thing indeterminate still. My own interpretations suggest an Obama victory (at least that's where I'd put actual money, if I was wagering), but not a landslide.
I have a student in England right now, and she says all the media over there are reporting a certain and sweeping Obama victory. I personally think the country's too closely divided for that, and I won't believe in a massive youth vote until I see it (anyone remember Howard Dean?).
Posted by: James Hanley | October 3, 2008 9:52 AM
I don't have to go back to Dewey, I remember feeling supremely confident in the Bush-Kerry election as well, particularly looking at exit poll numbers on election day. That didn't turn out so well.
On the other hand, the Obama campaign was eerily accurate in their primary campaign prognostication before it all started (that infamous spreadsheet of caucuses and delegates).
Posted by: Jeff Hebert | October 3, 2008 10:03 AM
Yes, why were the exit polls so different from the official results in 2004?
Posted by: Tophe | October 3, 2008 10:20 AM
Re James Hanley
I guess I didn't make my self clear in my previous comment. I am not claiming that the polls, as we sit here today, are inaccurate. My claim is that any number of things could happen between now and election day that could change the outcome.
One of the great urban legends of the past 60 years is that the polls in 1948 were wrong. There is no evidence that they were wrong at the time they were taken. The problem was that Gallup stopped polling several weeks before the election under the premise that the gap between Truman and Dewey was too large to close. Unfortunately for him, Dewey believed this premise, acted accordingly, and paid the price on election day.
Gallup and other pollsters learned from the 1948 embarrassment and didn't make the same mistake in 1968. Polls taken a month before the vote showed Nixon with a 15 point lead. Gallup and others continued to poll right up to the day before the election and correctly predicted that Humphrey was closing fast. Nixon barely won (his margin of victory was basically 30,000 votes in California).
Posted by: SLC | October 3, 2008 10:31 AM
Oh no, I hope the Obama camp doesn't get overconfident.
Posted by: Adrienne | October 3, 2008 10:41 AM
Gah, I just hope Obama doesn't get overconfident.
Posted by: Adrienne | October 3, 2008 10:43 AM
Oops, sorry for the double comment!
Posted by: Adrienne | October 3, 2008 10:44 AM
Two things:
Bradley effect
Overconfidence
I look at this polling so far in about the same way I look at the forecast of a DC snowstorm. I'll believe it when I see it.
Posted by: BaldApe | October 3, 2008 10:49 AM
I look at this polling so far in about the same way I look at the forecast of a DC snowstorm.
Common, it does snow here. The fun part of the forecasting game is to predict how high the level of panic is gonna be (e.g., light rain = minor fear; thunderstorm = moderate panic; snow = the end of the world).
Posted by: Josh | October 3, 2008 11:10 AM
www.fivethirtyeight.com has the best analysis of things. They run all the polls, weight them, and do 1000's of calculations. With them, Obama is in the low low 300's.
Posted by: matt | October 3, 2008 11:18 AM
300 EVs is hardly a blowout. We're just not used to it because we haven't seen it since Clinton-Dole in '96. For him to earn 300 EVs, he could run his tie scenario and win almost any combination of two or more from OH, FL, VA, and NC. I don't think that's unreasonable, especially since Ohio is one of those states that goes Republican in the good years and Democratic in the bad. Ohio's economy is still recovering from the damage that Taft did to it, and Florida is also going to be particularly sensitive to the economic matters.
A blowout would be 375+ EVs, which given the map isn't impossible but is considerably more difficult. Difficult is in this case defined as him running every state that's within 5 points, not unheard of but exceptionally difficult.
I'm more interested in the popular vote, though. I took the over in a +4% spread for Obama, and have a steak dinner at one of the best steakhouses in the area riding on it. Reading about his GOTV operation makes me think of prime rib, for some reason.
Posted by: Kit Smith | October 3, 2008 11:33 AM
Intrade has Obama at 338 to McCain's 200. That with over 1.4 Million trades, each involves actual gains/losses. Some economists think futures markets (such as Intrade) are far more accurate than polls, and the results tend to bear them out (however there still are distortions and biases, so beware)-DJ
Posted by: DingoJack | October 3, 2008 11:51 AM
For those of you worried about Obama getting overconfident, all I can say is that it would be so utterly out of character that I would honestly wonder if he'd been replaced by a pod person.
First example that comes to mind? In they hit a fundraising record in August (no doubt assisted by the Palin-inspired $10 million in 24 hours), but rather than make a big deal out of that, they tried to keep it quiet so donors wouldn't get - you guessed it - overconfident, and decide they didn't need to donate anymore.
From the very beginning, Obama has planned for the long term - registering new voters, making sure they get to the polls, remaining calm and steady while McCain tries one wild swing after another, hoping for a knockout punch. This is true grassroots campaigning in an era where that has been degraded to mean "playing to the bigotry of your constituents while trying to look like just a regular guy, despite the fact you're a multimillionaire".
Obama isn't going to drop that ball now.
Posted by: Seraph | October 3, 2008 12:08 PM
Some economists think futures markets (such as Intrade) are far more accurate than polls, and the results tend to bear them out
Of course the "futures markets" for political candidates are highly dependent on polls for their predictions. If you had something like Intrade and didn't have polling, the market would probably perform terribly because of that missing information.
"Futures markets" do well because they're trying to predict where things will be based on where things are now. Polling only tells you where things are now. Traders are synthesizing a lot of information to make a prediction about the future while pollsters are synthesizing a lot of information to make an analysis of the present.
(As an aside - I love how these days if you dress a gambling operation up respectably and call it a "futures market" no one blinks.)
Posted by: NonyNony | October 3, 2008 12:28 PM
Hey NonyNony (hey nony no)
Firstly Polls are often completely wrong, but markets tend (with exceptions) to be more correct. If polls inform futures markets then somehow they are not very good at it.
Secondly if you think markets are just legalised gambling perhaps you should get your friends together and picket - I suggest then corner of Nassau and Wall Street, NYC. -DJ
Posted by: DingoJack | October 3, 2008 12:49 PM
And now Obama's got Sarah Silverman's exceptional plan for getting the elderly Jewish vote in Florida:
http://www.webtvhub.com/sarah-silverman-for-barack-obama-video/
Posted by: pough | October 3, 2008 12:55 PM
Just don't offer them clean underwear or ramen noodles.
Posted by: Benjamin Geiger | October 3, 2008 1:21 PM
Seriously, people, given the number of states still using the Diebold voting machines (now called "Premier Election Solutions", because, as we all know, a name change makes all the problems go away) -- given that these problems have been rather well documented -- how on earth do you manage to ignore the elephant in the living room of the rigged or hacked vote?
The Republicans will win, because they have the support of the voting machine company, and can probably fund more close-in hackers in key districts.
Posted by: Luna_the_cat | October 3, 2008 2:44 PM
DingoJack--
I wouldn't say polls are often completely wrong. They have error bars and it isn't hard to see how wrong they are more or less likely to be. That is, if one polls 1000 people and gets and answer of Obama 53% and McCain 47%, the margin of error is enough (depending on how you do the calculation and regression) so that a 50-50 result is perfectly reasonable. It just looks like a wrong answer if it hits the 49-51 and McCain wins.
It's like saying I was completely wrong if I said a stock will go up between $5-10 over the next year and it goes up by $4.90.
Markets can do well at showing what people think the future will bring. But they are actually pretty bad at predicting that future -- if they were perfectly accurate there would be no bid/ask spreads, and bonds wouldn't have any change in the yield over treasuries, and every stock would be static because we would all know the "correct" price already.
It's also worth noting that most people, when they try to pick a stock based on what's hot now they are horrifically wrong. It's the reason there are no part-time fund managers, and why day-trading is actually a poor strategy over any length of time. But if markets were really a good predictor of the future then the tech bubble wouldn't have happened (or it would have continued forever).
This isn't to say markets provide no information at all, but just that they aren't any more rational than the people that make them up.
It isn't gambling, in the classic sense of supposedly random bets in a casino, but it is in the sense that you are making a semi-informed bet that certain things will happen, based (if you are an analyst) on a model. But there are plenty of instances (ahem, like now) where the models break down seriously. The models broke down because nobody thought home prices could fall (in real terms) -- and they would have been right if the pattern that has held since the late 1980s held up. It didn't, and if you push the model back to 1950 or so you get a much different picture. But nobody did because it was lot harder to get the data and nobody wanted to hear it anyway -- the market incentive was to behave irrationally,
Posted by: Jesse | October 3, 2008 3:40 PM
With stories like these, I think you can put the fears of overconfidence to rest:
Obama has 40 offices now open in Missouri, and Justin Hamilton, Obama's Press Secretary for Missouri, told us that while he couldn't confirm below or above the published reports of 150 organizers (it didn't come from the campaign), the campaign is only adding to its ground force. Organizers have now recruited 2500 neighborhood team leaders statewide, folks who do the far more effective work than any 30-second ad or yard signs, actual face-to-face contact and persuasion of their neighbors.
Apparently, the Obama campaign from top to bottom is working non-stop to get every possible vote out they can. The McCain campaign?
You could take every McCain volunteer we've seen doing actual work in the entire trip, over six states, and it would add up to the same as Obama's single Thornton, CO office. Or his single Durango, CO office.
Posted by: Greg | October 3, 2008 4:07 PM
The Republicans will win, because they have the support of the voting machine company, and can probably fund more close-in hackers in key districts.
Probably. Still, it's kind of nice to temporarily entertain some hope before being thrown back in the void for another four/eight years.
Posted by: schism | October 3, 2008 6:20 PM
BaldApe mentioned the Bradley Effect, where some races have been confounded by the people who told pollsters they were going to vote for the black man, but actually did not. Ed, I'd be interested to hear if you think that's a plausible scenario in this election. I have this nightmare of the actual vote counts shifting about 5 points toward McCain and Obama actually losing all the tossup states. Possible?
Posted by: Scott Hanley | October 3, 2008 7:41 PM
SLC,
OK, I see what you're saying, and don't disagree.
Posted by: James Hanley | October 3, 2008 9:30 PM
Markets are inherently more reliable than polls, assuming that there is some information to work with. As we see with the financial markets currently, it doesn't work too well when the information us inherently flawed.
The 1948 polling was wrong for a couple of reasons not already cited. First, both Gallup and Roper depended on the 1940 Census in selecting their samples and that data was seriously out of date. The other error was that much of the polling data was collected using mail surveys which were much more likely to be returned by higher-income (and hence more likely Republican) voters. Combined with stopping the polling in early October, the result was almost guaranteed to miss.
Posted by: kehrsam | October 3, 2008 10:08 PM
Actually, political gambling has been the norm in the US longer than not. At the turn of the 19th century, odds and positions were even advertised in major newspapers.
Posted by: Nick | October 4, 2008 4:01 AM
Actually, political gambling has been the norm in the US longer than not. At the turn of the 20th century, odds and positions were even advertised in major newspapers.
Posted by: Nick | October 4, 2008 4:01 AM
Look, Obama is going to win.
McCain can't possibly turn it around!
RELAX! Nothing can go wrong.
Posted by: Julie | October 4, 2008 4:43 AM
This is likely the high watermark of Obama's popularity and polling potential in the run-up to the election in early November. The economy is a mess and it plays into Obama's strengths.
Expect things to fall back in the next few weeks. Obama may not be able to make further gains or consolidate, so he'll have to concentrate on getting his base out on the day, and hope the McCain camp doesn't recover.
By the way, it's the Daily Telegraph, which is a national newspaper based in London, not a London newspaper. (Just one of my niggles).
Posted by: Tony Sidaway | October 4, 2008 10:33 AM
Yep, Missouri has now also become a swing state. http://www.electoral-vote.com has been showing this for weeks.
I thought it was a telephone poll, and Truman's voters didn't have telephones?
Recover? McCain isn't going to fare well in the remaining debates, and Palin is likely to provide more, shall we say, entertainment. All Obama needs to do is saying "it's still the economy, stupid!".
Regarding fraud: firstly, Sen. Chuck Hagel of Nebraska, the owner of ES&S, has enough of Fearless Flightsuit and is for Obama; and secondly, Obama doesn't strike me as someone who'd simply accept being robbed just because raising a stink would be unbecoming of a gentleman such as Kerry.
Posted by: David Marjanović | October 4, 2008 11:44 AM