||[Nov. 8th, 2006|12:36 pm]
Most news outlets are still calling the senate races in Virginia and Montana too close to call, but Tradesports
seems to have made up its mind. GOP retaining senate control has a mere 10% probability as of now, and surprisingly, hit that value last evening and has held steady.Will tradesports trump conventional predictors once again?
(To me the track record of prediction markets is not very surprising, given my core belief that the wisdom of the masses, as manifested in the free market, is more efficient than the wisdom of a few, a regulated/planned economy.)
Also, the probability of Obama becoming president
if he gets the nomination is around 70% (simply dividing the value of 2008-PRES-OBAMA by 2008-DEM-NOM-OBAMA). Right now the volume of trading on that contract is too low to mean anything, but assume, hypothetically, the following:
- Prediction markets are always accurate
- Shares on all contenders for the dem/rep nomination are actively traded
Does this mean it is an optimal strategy for each party to simply nominate the candidate that maximizes the conditional probability of success? I suspect the answer is no, but I'm not sure why.