Political scientist Michael Franz followed up on my previous post to see whether the prediction markets moved much after various events during the Democratic primaries. Below is his graph.
Most of the movement is due to actual primary results. You can see the effects of Iowa, NH, SC, Super Tuesday, Ohio/Texas, and NC/Indiana.
The other events — Wright, the “bitter” comment, the debate, Wright II — appear to have had very small effects at best. That seems significant given that bettors in these markets are much more attentive to news about politics than is the average voter. If these people aren’t reacting, then there simply isn’t much to react to.
Thanks again to Michael!