Prediction Markets and Obamacare – Intrade Got It Wrong

In my classes on strategy, I like to discuss how to prepare for the future. I’m careful to say that nobody can predict the future (though we often think we can) but we can prepare for it. In class, we look at various ways to think about the future, including scenario planning and prediction markets.

Prediction markets operate much like sports betting “lines”. In sports betting, the bookie simply wants to divide bettors into two equal camps. The amount bet for a team to win should equal the amount bet for the same team to lose. If the line is “Team A wins by 7 points” and too much money accumulates on one side, then the line moves up or down to even out the two sides. No matter which team wins, the bookie pays off the winners from the losers’ bets and keeps the service charge, known colloquially as the “vig”.

Prediction markets work essentially the same way, except you’re betting on the likelihood that something will happen or not happen. What’s the likelihood that Republicans will capture the Senate in the November election? What’s the probability that Greece will exit the euro before the end of the year? A number of prediction markets exist, especially in Britain where they seemed to first gain popularity. I tend to consult Intrade, which is probably the most popular prediction market in the United States.

Until recently, Intrade had a great track record in predicting political events in the U.S. It predicted, for instance, that the Republicans would not win the Senate in 2010, even as a number of pundits predicted exactly the opposite. This week, however, Intrade made a spectacularly wrong prediction. It suggested that there was a 75% probability that the Supreme Court would invalidate the individual mandate in the Affordable Care Act, also known as Obamacare.

So, if you can’t trust the crowd, whom can you trust? Should we turn back to the pundits?  The New York Times says experts may be a better source for confidential, closely held decisions. Bloomberg points out that this is not the first time that Intrade got it wrong. Additionally, the blogosphere went wild — just google “Intrade got it wrong”.

What does all this mean? Perhaps it puts us back to the original point: you can’t predict the future, you can only prepare for it. By the way, as of this morning, Intrade says there’s a 56.2% chance that Barack Obama will be re-elected. Anybody want to place a bet?