or like this (I hope this isn’t the same poll quoted in the first article):
Romney: 47%
Obama: 46%
that Intrade’s tracking of the presidential election, which I wrote about previously, would not look like this:
WTF?
giving B-Rock a 57% chance of victory to Mitt’s 41%.
However you care to count it, with the tumult of headlines and economic headwinds going on, I wouldn’t give an extra 1-in-5 chance of winning to the incumbent when money’s involved here.
So what’s happening here? Remember, one of these two candidates will win. There are, to be sure, 4-and-change months left until the election. But what can swing one way can just as easily swing another, and so one really must wonder if this is a wonderful arbitrage opportunity, or if someone’s manipulating Intrade and making its value as a prediction marker into excrement.
PS. 56.6% + 40.6% = 97.2%. So there’s definitely at least 2.8% of arbitrage opportunity in this market. Perot ain’t running this time around.
Intrade is a prediction market. In short, it aggregates the views of individuals into a composite probability of an event occurring by using a trading pricing mechanism. The prices reflect the implied probability of something happening. For instance, if Event A is trading at 2.90/share, then the market says the probability of that event occurring is 29%. I follow it to get a sense of what may be coming down the pipeline.
In particular, I follow the presidential markets. President Obama has, with the exception of the latter part of 2011, had fairly solid numbers for re-election. However, his prices have been dropping recently and dramatically, so I wanted to look at why this might be, and whether the market was functioning efficiently enough to be regarded as a credible indicator of the Fall ’12 race. Intrade provides charts and day-to-day data on trading in its markets.
This chart shows the recent history of the intrade prediction on the re-election of President Obama.
Obama Re-election price history
There are a few major jumps with apparent ties to a few news events, highlighted below. I researched some of these events on the wikipedia calendar summaries. Macro concerns–a la Europe, and so on–probably weighed in on late 2011, as I couldn’t find any single events responsible for the trough in that area.
Prices, with some real-world events
Interestingly, though the price has been more or less constant over the last few months (less late May/June), trading volumes have been jumping dramatically.
Trading Volume Summary Stats
Now, obviously the volume increase is at least partially attributable to the nearing presidential election. However, the massive jumps in volume (e.g. ~7%-200% swings) with the more-or-less negligible price change (from 60-54, a 10% change) tells me that more and more people from both sides of the aisle are getting involved and putting money on the table in this market.
As more people are putting money in the market, the market should function more efficiently per the research sited in the wiki article on prediction markets. That is to say that the price should be a more accurate indicator of the “wisdom of the masses” and a better reflection of what will actually happen.
Therefore, the price/probability of Obama’s reelection dropping in the face of increasing volumes is a bad sign for Obama’s re-election. Anything can happen, but if this trend continues (and I expect that it will), we’ll see dramatically increasing volumes, and we’ll see the Obama re-election probability dropping to sub-50 in the next 2 months. This will indicate a losing campaign in the fall.
Conditions for Obama winning, or staying above 50 in Intrade: improved unemployment figures and improving economic indicators, e.g. the S&P increasing. But with Europe about to blow up (again, or finally?) and a dropping stock market, I don’t see that happening in the near term.
An interesting study might be to correlate trading volumes with media coverage citations for both GOP candidates and Obama, but I’m not sure where to get that data.
****Update
Also, as a teaser– Right now, 7 June, Obama is trading at 52.7. Romney is trading at 43.3. Those added together equal 96% probability. Now, we know that one of those two is going to win in November. There is no 3rd option. Where is the other 4% of probability hiding out?
Falling Outside the Normal Moral Constraints 