Putting Your Money Where Your Model Is

There is nothing I like better than an academic wager. I fear that this time I’m over-matched: I’ve bet on politics against Columbia’s professor, Robert Erikson, and Bob is one of the leading political scientists in the U.S. But we had to bet, because his models of politics currently give a very different reading of 2008 than mine.

Let me start with some background data, drawing from my latest W.S.J. column:

A straightforward reading of recent polling data makes the presidential election look like a foregone conclusion … Polls from the past two weeks put Sen. Obama somewhere between 4 and 12 percentage points ahead of Republican John McCain. In the poll-of-polls aggregate on RealClearPolitics.com, Sen. Obama’s lead is a massive 5.7 percentage points — as large as it has ever been.

Senator Obama’s lead is larger than the stated margin of error …. Moreover, Sen. Obama is leading in all of the major battleground states, except Virginia and Florida which remain incredibly close.

How should we interpret polling data four months prior to an election?

Political scientists Robert Erikson (of Columbia) and Christopher Wlezien (of Temple) have recently mined daily polling reports from the last half-century of elections, mapping the relationship between early polling numbers and final election returns. At this point in the race, they find that around half of any lead should be discounted, as early advantages tend to dissipate. (You can read the full paper here, or an ungated version here.)

Profs. Erikson and Wlezien point to another reason to be wary of Sen. Obama’s early polling lead: On average, the voting public tends to be more strongly anti-incumbent three-and-a-half years into an administration than they are on Election Day. Based on patterns in previous cycles, the professors suggest that this exaggerated anti-incumbent feeling is boosting Sen. Obama’s lead by around three percentage points.

So if you first halve Obama’s six point lead, then subtract a three point anti-incumbency bias, you are left with a dead heat.

Bottom line:

A naïve reading of polls suggests an Obama landslide; a sophisticated reading points to a dead heat. Prediction markets are somewhere in the middle, suggesting a two-in-three probability that our next president will be a Democrat.

Yet it is the naïve reading of the polls … that dominates media headlines.

There’s a broader intellectual debate here, too. Erikson and Wlezien note that most comparisons of polls vs. prediction markets are comparisons of the naïve reading of the polls versus the prediction markets. That’s a pretty weak test, and so when they ran comparisons of their adjusted-polls versus markets, the markets no longer looked so good.

In fact, over the past four election cycles, the poll-adjusted forecasts did slightly better than the markets. Bob scores this as a victory for the polls over the markets, while I think that four election cycles are too few to overturn the fact that in nearly every domain I’ve seen, prediction markets yield the most accurate forecasts. (You can read more on our debate here and here.)

Given Bob’s model, he should be willing to accept an even money bet: say a bottle of wine for a bottle of wine. Given the markets, I should be willing to bet my two bottles against his one. And so we compromised: I’ve bet three bottles that Obama will win against his two bottles on McCain.

Bob tells me I’ve got the upper hand in this bet: in ongoing research he is finding that polls are slow to reflect economic conditions. Adding in a third adjustment to account for the effects of a weak economy, he concludes that Obama should be the betting favorite.

But he tells me he is willing to accept the bet for another reason: “If McCain wins, I could use that wine.”

The full column, comparing polls and markets through the election cycle, is available here.

H Tran

Turnout is going to be the key. Reasons for people not voting: too difficult to decide, too confident(Obama supporters), too exhausted (Obama again). McCain - cheap wine and a cigar

Jed Christiansen

Nate Silver (of PECOTA sabermetrics fame) has developed his own model for this election at http://www.fivethirtyeight.com. It uses polls, but instead of strictly historical weighting, also goes into demographic movements and more. (My description doesn't do it justice; best to read the site for more.)

What's interesting to me is the result: almost exactly what the PM's are saying for the next election, that Obama has around a 2/3 chance to win.


Put your money on Barack. It's the economy stupid. W ruined the chances for the GOP.


Intertrade where people can bet on the presidential outcome. It favors Obama more than the polls do. I think that when the debates begin, McCain will get killed. People will see a taller, younger and well spoken Obama compared to a shorter, frail looking bumbbling McCain. I like McCain, but don't think he has a chance.



A bottle of wine for a bottle of wine is hardly an even money bet. While a bottle of "two buck chuck" certainly does the trick... its no Plumpjack.

Regardless, this is going to be an exciting election.


Great post!


I don't see mention of Bradley effect.


peter i don't think the bookie odds would be very different from the betting market odds. theory of one price and all.


I think there is a huge black swan on the horizon. Iran is a powder keg waiting to be ignited. Israel is going to bomb them and who knows what will happen. It could be World War III.

Maybe Obama will be able to use what happens to his advantage and maybe McCain will use what happens to his advantage. It doesn't seem clear.

I just think that event will drive the elections more and people aren't thinking about it now, so the polls won't be a good prediction at this point.

Petey Pablo

Mike said: "How else could you explain his “dodging” of all of McCain’s requests for debates?"

Risk aversion. It's in the underdog's best interest to have more debates, opportunities for the tide to turn.

That said (and your political leanings aside), I think Obama's debate skills are being overestimated by many in his camp (while I believe them to be superior to McCain's (in most forums - town hall is an exception), both could stand some improvement in that arena).


Beware of double counting here. You have halved the current Obama lead due to "early overstatement" and then discounted an "anti-incumbency bias". But how much of the former (early overstatement) is *due* (or at least related) to the latter?


I think there is a lot of misplaced overconfidence in Obama's ability to debate. He sees it as well. How else could you explain his "dodging" of all of McCain's requests for debates? Plus Obama fumbled badly when he was presented with tough questions during the debate moderated by Charlie Gibson. The fact is, the media has snowed the public so badly about Obama's true stance issues, that most people don't know up from down. They have had it drilled into them for years that Bush has screwed everything up, and they just want "change." Meanwhile, we are all going to get some change for which we have not bargained. Obama is smart to stay away from McCain as long he leads so much in the polls and has the compliant media firmly in his corner. Eventually, I think (and hope) people actually start asking some serious questions and debate the outcomes of his policy prescriptions of less union oversight, greater environmental regulation, higher taxes, and greater economic interference.



But what do the book-makers say?

I thought it had been shown that the best predictor of any competition is the betting odds set by professionals.

This makes economic sense. For most analysts the prediction is an intellectual exercise - they get paid whether they are right or not. The predictions are forgotten after the fact so even their reputations are not on the line.

But for a professional bookie their livlihood depends on being able to accurately set the odds and skim off the percentage.


Good Call, I really want Obama to win so I just placed a bet online on McCain to hedge my hopes.

I often do that when rooting for my favorite sports teams. I find losing is easier to stomach when I win money.


There are a few "human" factors that haven't been mentioned, and in my opinion, they are huge. In short, I don't think Obama will even be close. Here's why:

1. Obama/Hillary have been dominating the news for a long time, and people have almost forgotten that McCain is even in the race. That enthusiasm will die down and the race will begin to appear closer.

2. McCain is a war hero who has consistently voted with the left. He has shown he can work with the left and that he is not as diehard a Republican as others. In fact, many have called him a RINO or even a Liberal.

3. The hidden dragon in the polling data is racism. Regardless of what many people will say, America is still way too prejudiced to vote for a black man. It's a sad and ugly truth. I live in Ohio, and I can tell you that a great deal of the democrats here are old-school, blue-collar racists who would never vote for Obama. I'm sure Ohio is not alone.

They may tell the pollsters that they'll vote for Obama, but when it comes time to pop the chad, all alone in the booth with no one watching, Obama loses many, many votes.

Take McCain and win some wine. He's a lock.

Me, I'm voting Libertarian as always.



Carl, my guess is that Iran won't happen until AFTER the presidential elections One last parting gift from W.

I think the deciding factor will be televised debates. When people see how decrepit McCain looks, particularly in HDTV, it'll be Nixon and Kennedy all over again. And McCain isn't a polished debater in the first place. Obama will walk all over him.

Personally, I'm undecided, but I think it's Obama's election to lose.

Alvaro Fernandez

I don't recall who said but now being labeled a republican is almost like being labeled a child molester. The Wall Street Journal poll had a huge margin in favor of the democrats when the question was "would you rather have a republican or a democrat as new president" and it dwindled down when the question was McCain vs Clinton or McCain vs Obama. Silent majority refuses to express its true intention on polls.


Correct me if I'm wrong, but John McCain is not the incumbent, right? Yes, he hails from the incumbent party, but surely that's different than actually being the incumbent?

Obviously the Obama camp would like us to treat McCain as if he were George Bush, but that doesn't make it fact.


I think people are overstating whatever advantage Obama might have in the debates. Obama will obviously be more visually appealing, but neither he nor McCain is a very good debater.


This years elections and the predictions reminds me of a quote from Bill Joy of Sun Microsystems, which was "You can drive a car by looking in the rear view mirror as long as nothing is ahead of you."

While I am neither educated in these matters, nor have I done scientific research, I think is happening is a very fundamental shift in how campaigns are funded and how to woo an increasingly significant segment of the American electorate.

More specifically:

* Traditional media increasingly matters less than the internet - the political perspectives of people are increasingly based more and more on forums and blogs like this. Dumb posts are skipped and thoughtful posts are digested. Issues are rarely as simple as they are portrayed in the mass media, so the depth of detail is far greater in this format. If this seems a doubtful trend, consider that the viewership proportions involved here probably mirror the decline of TV (traditional news) and the rise of gaming (internet news) for entertainment. Its more common that you think, but its hard to measure as there isn't a Nielsen ratings system for political blogs or any other genre.

* Funding has fundamentally changed - just glancing at the wiki for the 2004 elections (http://en.wikipedia.org/wiki/United_States_presidential_election,_2004), realize that if only 10% of the votes for a candidate opted to provide only only 3 dollars (thats 75% the cost of ONE gallon of gas) a month during the campaign season, each candidate would draw ~18 million monthly. Thats automated deposits, not a huge string of fundraisers, leaving the candidate more time to campaign.

Specifically regarding Obaman, don't be so quick to disregard the rather impressive administrative/organizational skills, which has provided Obama with his victory over Clinton and very likely a victory in November. Realize Obama was a nobody in the beginning without all the advantages Clinton had, yet it seems he meticulously built a wining political machine that has really overcome impressive odds within his own party.

For the record, I gave money to Ron Paul and I find both candidates largely distasteful (I am a small gov't/fiscal conservative type). I will vote, but I really don't like my choices at the moment.