Thus Intrade is usually very accurate, predicting presidential elections, economic conditions and academy award winners very accurately. This morning, a ticket that the SC would rule the mandate unconstitutional was going for $7, or a 70% chance they will do just that. But Intrade was very, very wrong. Why? Two theories.
Few Decision Makers. It's very hard to predict anything that a very small number of people decide. Here, there were really only two decision-makers. In fact, most thought it was just one--Justice Kennedy--who indeed sided with conservatives. But no one predicted Justice Roberts would switch sides. Like a sample size that's too small, it's really hard to predict what will happen when so few determine the outcome.
Sample Bias. Go ahead and try to set up an account. If you try to get starting money via a card from your American bank account, you'll see this message:
If your credit or debit card was issued by a US bank then unfortunately you will not be able to use this card to fund your account. This not the decision of Intrade but due to regulations in the US that restrict what US residents can do with their cards.So that puts a bias on who uses Intrade and it makes sense that non-Americans wouldn't be very good at predicting what our highest court does, especially since American's can't seem to predict it, either. It is worth noting that every expert, not just Intrade, was surprised by how the Supremes ruled.
No comments:
Post a Comment