Picking the Next President: What Are the Odds?

My latest Wall Street Journal column about the election is titled “Making Sense of Disagreement.” I start with a simple question: who is the current favorite to win the election?

Polls taken since the Palin pick have typically suggested that McCain is the new favorite. But sorting out an underlying trend from the usual convention bounce is a tough task. And indeed, the latest poll-of-polls has the two candidates roughly even, with this morning’s numbers suggesting a trend back towards Obama.

My usual starting point is to turn to Intrade.com, whose markets had shifted last week to install John McCain as the new favorite.

Why start there?

Political insiders are known to follow the Intrade market closely, partly because the large sums of money being bet give the opinions of these traders greater credibility. Greater volume also makes it more difficult — or at least expensive — for politically-motivated traders to manipulate this market.

This morning, Obama holds a slight lead at Intrade — a reversal from recent days. This isn’t the only place to look though:

But there exists another major political exchange: Betfair. This British alternative also takes large real-money bets, but will not accept trades from U.S. customers. Across the pond the Democrats remain a robust favorite, with a 58 percent chance of winning the White House compared with the Republicans’ 42 percent chance.

The Iowa Electronic Markets also rate the Democrats as the favorites by a roughly similar margin. However, the Iowa market is linked to the winner of the popular vote, not the winner of the White House, and so perhaps it is not surprising that the Democrats are performing better on this score.

And in fact, there are plenty more markets also offering odds:

Various online bookmakers (and, in England, street-corner betting shops) are offering odds — and all rate Sen. Obama as a strong favorite.

There are myriad other play-money prediction markets that also point to an Obama victory. For instance, the market run by NewsFutures.com suggests a 53 percent chance of the Democrats taking the White House, while a competing market run by Inkling suggests a 61 percent chance for the Democrats.

(Note that the numbers above have shifted farther towards Obama since my column was written.)

How do we make sense of these competing numbers? In truth, I can’t:

The current variation in price is larger than I have ever seen in my years of studying prediction markets. The forces of arbitrage that would typically eliminate these differences have been handicapped by the legal restrictions preventing U.S.-based traders from using overseas markets. I also suspect that this week’s financial strife may be a pretty big distraction for many people who otherwise might be active traders. Even so, it is surprising that enterprising European- or Australasian-based traders are not exploiting the current arbitrage opportunities.

I’m left wondering just which set of prices to trust. Let me weasel my way out: I don’t know.

But I do know that all of these markets agree that the race is close, and if anyone has the edge, it is Obama. You can read my full column here, and the related thoughts of Emile Servan-Schreiber, the NewsFutures founder, here. (Disclosure: I’m an academic advisor to NewsFutures.)

Ben Friesen

http://www.fivethirtyeight.com/ shows much more information than the other polling composite places. They put an Obama win at 73%.

http://www.pollster.com shows there being a 269-269 tie in the electoral college if the votes were cast right now, but they hedge their bets more.

http://www.realclearpolitics.com has a composite that's incredibly opaque, but they also lean towards Obama winning.


I'm surprised at the frequency with which Intrade and other internet-based prediction markets are used in the mass media as some sort of harbinger for political victory. By my calculations on Intrade, the TOTAL DOLLAR VOLUME of trades for an Obama victory SINCE HIS MARKET WAS CREATED has been approximately $2.5 million. Approximately 9 months of trading and he's at a mere $2.5 million. This is hardly a large amount; in the stock market, a single stock which has a DAILY DOLLAR VOLUME volume of $2.5 million is considered to be quite small.

My conclusion is that Intrade and other political prediction markets are not reliable sources for making a meaningful statement about the probability of victory in the presidential election. They simply do not have enough trading activity and are too small.

Mike B

I would like to once again mention the possibility for people using Intrade to hedge against a McCain victory when things got close. This would skew the price in McCain's favor.

Mark Truman

Does anyone know how they account for the lack of land lines? It seems like it would be an enormous hole in their projections?



You've hit the nail right on the head. From this morning's news:

A new Nielsen Mobile study indicates that one in five U.S. wireless households could be wireless-only by the end of 2008. Already 20 million households, or 17%, are abandoning land-line phones in favor of mobile ones. That's up from only 8.5% in 2005.

That's got to skew the data quite a bit!


With a reported half million new donors in August to the tune of 66 million, and the seemingly limitless supply of funds being willingly funneled into the Obama campaign by average and wealthy amerians alike, I don't think it is as close as they predict.

Those phone polls are notoriously unreliable as well. Hardly any one in my age bracket (18-30) has or uses a land line. Those polls must certainly be skewed towards SAHM's and the elderly.


Going by the limit book liquidity, Betfair is pretty weak on the bid side for Obama. Hitting the bid for $1000 would knock the bid down 10 cents(ignoring hidden liquidity), whereas the ask side is much stronger. I would therefore trust Intrade a bit more. I don't think the arbitrage opportunity is that great, because of transaction and especially execution risk(unless you have a program that simultaneously trades both markets)


I will be voting for Barack Obama, but frankly the pessimistic side of me is saying McCain will win.

First, there are a lot of Americans who do not want the country to be run by a black man. No one is going to admit their racism in a poll, but I know many people who have privately admitted to me their fear of having "them" in power.

These people are educated and urbane; not some high school dropout redneck. Racism is still very entrenched in American society.

Second, with more people voting for American Idol than the US Presidential election, a candidate’s personality trumps any discussion over issues.

Barack Obama has an extraordinary background; is very well spoken and has a rational perspective on this country’s situation. He also has nothing in common with a large section of the American electorate.

Republicans in this election will not be voting for John McCain but for Sarah Palin. And a mother of five that has shot a moose and promises to shake up the old boy network in Washington speaks much louder than any eloquent and logical speech Obama could give.



I can only say, from these betting parlors to God's ears. We will not survive another four years of Republican rule.


@1: Even with Obama's recent fund raising the McPalin+RNC war chest is roughly twice that of OBiden+DNC. Obama's decision to reverse himself and forgo matching funds means that he starts 90 Million behind while people who want to donate to McCain can still donate to the RNC.

You are free, of course, to fudge your opinion on the current polls with whatever facts you hear but donations is a poor choice given the current Republican advantage in funds and the huge cost of Obama's staff.

To the people putting any weight on the European only numbers; you are foolish with a capital F. Their social structures are so different from those in the US that the concept of McPalin winning is beyond imagination for many of them. They can't even conceive of the kind of life a person in the Western US lives and Alaska might as well be another planet (I've lived there so I know.) There is a spirit in the American psyche to which Palin speaks that most simply can't understand.




Do you really think that 9/11 was a black flag operation? Wow.


@15: Matthew, having lived, worked and travelled extensively in the US and Europe (based in the UK) you are wide of the mark with your assessment of how different life is in the Western US. They are far more similar than you try to imply.

As for the American 'spirit' to which Palin speaks, such talk is absolute nonsense. Palin has been selected because she appeals to certain sections of the electorate that McCain doesn't. Generally the type of voter that she has been selected to appeal to is religious and insular.

As a European I initially had a preference for McCain and thought Obama was not up to the job. My opinions on the presidential candidates has not changed significantly but the addition of Palin to the McCain ticket has convinced me that an Obama victory is 100% better for both America and the world.

D T Nelson

Speaking of reasons to not trust the polls, I suspect we will see a reverse Bradley effect among women voters: Some percentage of them will tell the pollster they are going to vote for the Democratic ticket, because that's what they think they should say, but then voting for the Republican ticket to elect the woman.

Troy Camplin

To what extent are the Europe-based companies being skewed by their own preference for Obama?


I was referring to polls like the Rasmussen poll, which the above editorial includes a reference to, and it is a telephone poll. If you took the time to look at them before responding...

Jacob Lipton

Surely betting in Britain (where I live) is skewed by what people are betting on. When most gamblers are favoring one candidate, the odds on that candidate are driven down (because the bookmakers do not want to risk large losses, and because if people are more inclined to bet on one candidate the bookmakers can lower the odds and still receive bets).

Because the British public heavily favors Obama (I think he would get up to 80% of the vote here) they are more inclined to bet on him. I'm not sure exactly why this is - perhaps just optimism - but I do know that it is a phenomenon that is especially noticeable when English or British sports teams play against foreign ones. British bookmakers will be in serious trouble if England ever manage to replicate their 1966 world cup win!

Paul K

#9 (Amanda) - they are referring to the opinion polls and not the betting markets. The opinion polls show it about even, where the betting polls show it tipped to Obama in smaller or greater amounts.

The question is how valid a phone opinion poll can be these days given that a sizable portion of the voting population does not have a landline that the pollers can call? Of course, if you looked at the landline numbers in close states, I suspect it is far less sizable. I think you will find more mobile only households on the coasts. But, opinion polls seem to be less and less accurate, so Justin is looking for more accurate predictors, and betting markets are one such.


If Intrade, Betfair, and bookmakers involve real money, and if there are differences, this appears to provide an easy opportunity for arbitrageurs to profit without risk.


Obama is so obviously the superior choice. It should not be close, but it will be because:

This is the NEW WORLD ORDER... don't put it past the neo-cons and their corporate masters.

You think false flag operations(911), packing the judiciary, phony wars, hubris, the accumulation of wealth by a select few, the assault on the middle class, police state, disregard for the constitution, total surveillance, rigged elections, voter disenfranchisement, voter fraud, financial crisis, lack of education, media bias & manipulation, religious verve, class warfare... etc... are all just coincidence?

Welcome to the NWO.

Do you really think they will let Barack win???

It doesn't matter if 80% of the country votes Obama. The fake news gives you fake polls saying how close this fake election is. They are priming you for disappointment.

but... I hope I’m wrong!!!

EVERY PERSON interested in $ MUST SEE the documentary: "Money as Debt" or even "zeitgeist"...




"I’m wagering considerably on Obama so that if he wins, I can afford to pay all the new taxes that are sure to follow."

I am so sorry that you are making over $250,000.....you have my deepest sympathies.

getting back on topic, I wonder how much the "bubba vote" will actually matter, that is people that trend Democratic, poll Democratic, but when it comes down to selecting an African American for President, they....can't.