It's interesting looking at the way some of the prediction markets behaved through this cycle. Up until results started coming in, sites like predictit.org wer

It's interesting looking at the way some of the prediction markets behaved through this cycle. Up until results started coming in, sites like predictit.org were all "predicting" overwhelming Clinton victory. That so far has clearly not panned out. I think it says that these prediction markets aren't so good at actually predicting anything: all they really do is sort of act as a barometer of how media narratives influence popular thought on certain subjects. I do expect that they will correlate with outcomes better than 50/50... but not so much so that the "wisdom of crowds" line is really justified.