Hat tip to the Australian Lowry Institute think-tank for this. A mix of experts commenting on the issue, on-line summary or a podcast:http://www.freakonomics.com/2011/09/...of-prediction/