Since Chris must sleep at some time (I think)…
Alex Kirtland January 8th, 2008
… I’ll alert you to a developing story. [Slate's Daniel Gross: Why were the political futures markets so wrong about Obama and Clinton?]
Thanks to a friend.
~alex
2 Responses to “Since Chris must sleep at some time (I think)…”
Leave a Reply
You must be logged in to post a comment.
The point that has been made time and time again by the prediction market devotees, has been that prediction markets are more predictive/efficient (P/E) than traditional polls.
It is well documented that the polls often get it wrong when an election race is centred around an African/American candidiate. Therefore, if the prediction markets are indeed more P/E than the polls, one would have expected them to factor this point in.
In New Hampshire, they did not do this - they shadowed the polls, and give Obama a 90% probability of winning the caucas. Real money did not reveal the true probabilities. Yesterday Obama was trading on Intrade with a 76% probability of being the Democratic nominee; today, he trades at 40%. Such wild swings taking place in one day, do little to counter the argument that prediction markets are indeed inefficient.
It is hoped, that out of this result, we may see some proper debate of the facts.
“Real money did not reveal the true probabilities.”
The price of this event derivative can be interpreted as the objective probability of the future outcome *(i.e., its most statistically accurate forecast)*.
Also see:
http://www.midasoracle.org/200.....ndicators/