I agreed above that not everything on his site was supposed to be agreed with. Though I do agree that with his idea of showing the data behind the polls. A lot of the major polls hid their turnout numbers on the 30th page of their reports. It was also shown that some of the major polls as the election got closer, some polls went from the extremes of D+11 to just D+3 in only two consecutive polls. If that doesn’t raise eye brows, nothing does.
All the major turnout polls trended higher turnout for GOP. Independents trended higher for GOP, too, (including in the results, the GOP got more IND’s). Gallup and Rasmussen had massive 5,000 person samples to base their data on. Those polls were definitely more “scientific” than the subjective guesses of those at CNN, reuters, and others. The fact that the more scientific ones came out wrong only shows that there’s something wrong with the science as of now.