I have written about the polling data coming out of this race a few times now and always with skepticism. There was a sudden bounce for Obama around the time of the economic crisis and until late last week that bounce was holding. Something strange happened though late last week into early this week, a set of polling data came out showing that the race was in Obama’s favor but within the margin of error. Increasingly it the polls showed tightening with independent voters becoming somewhat more receptive to McCain.

The difficulty is that there are still polls showing Obama with huge gains. Most of these polls assume large turnouts among African Americans, young voters and a shrinking Republican population. The question becomes, are they oversampling and overestimating the turnout with youth and the effect of African American turnout or are they dead right?

A Pew Research Center poll released yesterday shows a 15-point lead for Obama, a result based on relaxed criteria for when to consider an African American respondent a likely voter, said Andrew Kohut, president of the center. He said the poll shows that roughly 12 percent of the electorate this year is black, up from 2004, with a similar increase among younger voters. Kohut defended this approach, saying there are historically high levels of interest in this contest among both demographic groups. At the same time, he added, “we’ve consistently shown less enthusiasm and engagement among Republicans than is typical, and the composition of the electorate shows that.” (Washington Post

Likewise state polling appears to have the same problem. Many of the estimates are placing Obama well above three hundred in the electoral vote. The same problems with the national polls exist with the states and so these polls are showing traditionally Republican states, including McCain’s homestate of Arizona, close to turning over for Obama. This seems unlikely unless you consider that in the last few years large portions of the population have moved. There has been a resurgence in the city populations of many Southern and some midwestern states as people seek jobs. Is this shift enough when mixed with the dislike of the president to turn conservative voters over to a liberal Democratic candidate? It could be or it could also be a sign of trouble in predictions.

This normally wouldn’t be too much of a problem as it wouldn’t be the first time Americans have had an election called a few weeks ahead by pollsters and pundits only to see a massive shift of expectations near election day. The race this year is altogether different largely because of race.

Many pundits are already building up the case that if election results don’t mirror the polls it will be because of voter racism. These voters seemingly would have told pollsters they were undecided or supporting Obama because they felt that was the socially correct answer, when left alone in the voting booth though some believe they will switch.

It seems likely with all of the evidence that Obama will win Tuesday, but there is enough reason to doubt the polling to hold off. Unfortunately many in the media and online haven’t done so, instead giving the perception that there is no discrepancy in the polls and instead that the race is already decided. My personal fear is that the polls are wrong and the only conclusion offered on election night will be one that puts race front and center instead of the perceptions of pollsters.

Lets just hope for the best.