Crooked Timber looks to squash a spreading error – the idea that electronic markets predicted the US election better than the polls did. The actuality – that the markets were as inaccurate as the polls – indicates to me both the reliance of markets on available information and common wisdom, rather than anything more ethereal; and the primacy of polling data in the US political newsline. Are there alternative means of sampling public views than opinion polling? How reliable is it? Where is the political bias?