I heard about this story on Thom Hartmann today:
http://www.statslife.org.uk/...
Essentially it boils down to a statistics professor noticing an unusual trend in voting patterns in states using electronic polling machines with a shift to republicans as voting precincts get larger. The usual assumption is that larger urban areas would trend towards democrats. If you look at the graphs in the paper, they initially trend towards D but after a certain population size they shift back towards R. After reading the article, I'm wondering if there could be any reason that trend could be legitimate and not due to voting machine fraud. I would like to be skeptical of massive fraud and consider all possible explanations. Is there a reason why intermediate size precincts would go towards democrats and the largest size would return to republicans?