Patrick Ruffini has a very smart overview of how flawed the polling has been in the Democratic primary. Overall, the polls have underestimated the winners support by about 5.86 percent, and had an average error of almost seven points. In certain contests, this had huge impacts on the race, because the polling set the expectations. Clinton's win in New Hampshire was so contrary to the polls that the American Society of Public Opinion Research formed a committee to evaluate the discrepancy. In South Carolina, polls showed Obama up by 11 points, he won by almost 30 percent. In Wisconsin, polls showed Obama ahead by 4 percent. He won by 17. But these weren't random swings or unpredictable outcomes. Rather, they all fit a precise pattern, and could indeed be predicted with fairly simple formulas. Ruffini writes, "Now that we are approaching a perfect 50-state electoral breakdown of Clinton-Obama -- marred only by the presence of small-state caucuses -- we see that the results are totally internally coherent based on demographics. In fact, I'm prepared to argue that we would have gotten almost the same margin in each state in a national primary held February 5th. Because the candidates themselves polarized the electorate demographically, momentum and events (and polls) made little difference." Indeed, the absence of a momentum effect, a la Kerry, has been starling. Ohio didn't care what West Virginia did, New Hampshire had no interest in Iowa, and North Carolina was unmoved by Pennsylvania. In this race, demography really has been destiny, and that doesn't strike me as an encouraging conclusion.
