Statistics and Trends

It's interesting that if you look at the graph of Iowa Electronic Market futures prices you see the same rapidly increasing spread during September, then trending back together after the first debate at the end of Sept, now spreading apart again. The IEM and Den Beste graphs are remarkably similar.

This is consistent with den Beste's media manipulation hypothesis if futures buyers are taking their cues from media polls. On the other hand, poll respondents and futures buyers may simply be responding independently to the same information, whatever that is. That latter case might still be blamed on the media, of course, since they create the information environment to some degree. That may be especially true when it comes to heavily-hyped news blips and kerfluffles that, in the absence of really important events, may drive poll results pretty dramatically for short periods of time.

The extreme version of this argument would have the media jimmying the information environment and jimmying the polls, those results becoming part of the information environment, which influences the perceptions of futures buyers and subsequent poll respondents, and so on. But even this has its limits. The media can't control the interaction between events of real importance and the gradually developed public perceptions of candidates -- that interaction is what creates the overall trends.

Byron