Many "free market" thinkers and others have bought into the theory that betting markets such as the Iowa electronic futures and intrade.com are predictive. The idea is that if people risk hard earned money then the pricing outcome will truly reflect an informed judgement, and the results will be more predictive than even scientific polls.
This has proven to be untrue. A study of these "markets" shows that the prices follow the news rather than lead the news. A good poll for Hillary, her price goes up. Something bad happens it craters.
Were these truly predictive markets they would be more stable in reference to daily news, and sometimes counterintuitive.
One other observation: If somehow the bettors had access to information the general public didn't have yet it would also be more predictive. But with the internet and 24hour news, we are all in the same universe.