While Barack Obama earned another four years in office last Tuesday, the true election night winner was Nate Silver. Using complex statistical models to aggregate the results of dozens of different political polls, the blogging mathemagician was able to correctly predict with uncanny accuracy the voting results in all 50 states — a remarkable accomplishment, especially when compared to the predictions of other more experienced, more politically-connected pundits.
A far less sophisticated but nearly as impressive feat was carried out by Intrade.com. The online prediction market was able to pre-determine the outcome of 49 of the 50 states. The only mistake? Florida. Given the fact that the margin there was so close it wasn’t officially declared until Saturday, Intrade could probably be forgiven. Over the past three presidential elections Intrade has only missed on three states.
How does Intrade work? Not by relying on wizards like Nate Silver, but by relying on the wisdom of a whole bunch of normal people. Much like the stock market, Intrade lets investors buy shares of possible future events, say Barack Obama being re-elected. If the event does come about, winning investors receive a payout and losing investors receive nothing. Intuition tells us that the more people participate, the more likely the outcome is to be wrong since very few people know enough about politics to make reliable predictions. But somehow the opposite happens. The more people invest, the closer the market value gets to the actual result. While the average investor doesn’t know what’s going to happen, aggregating the gut-feelings of all those investors generates remarkable accuracy. Another weird example of the whole being greater than the sum of its parts.
This phenomenon is analyzed by James Surowiecki in his excellent book “The Wisdom of Crowds.” Says Surowiecki, “under the right circumstances, groups are remarkably intelligent, and are often smarter than the smartest people in them.”
So how does a market make better predictions than political polls? The biggest factor may be the money. While people tell pollsters whatever they want to without repercussion (like “Yes, I’m likely to vote” or “I’m moving to Canada!”), Intrade investors have to put their money where their mouths are.