Much has been made of how the youth while playing a large role in this election perhaps didn't turn out in as large a number as expected. In addition the overall turnout didn't show as much of an increase as expected. I think that the long lines in democratically leaning precincts (and the publicity of such lines) might have led to a decrease in voter turnout among democrats.
In examining the NY times exit polls I noticed the following:
Previous Vote Obama/McCain
37% Kerry 89/9
46% Bush 17/82
11% First time voter 69/30
(6% voted before, not Bush or Kerry- results not included)
Note that this poll included a component of early and absentee voters in addition to the standard election day exits. Thus, Kerry voters should have been included regardless of whether they voted early or on election day.
Thus, it would appear that among those who voted previously, Bush voters out numbered Kerry voters by a 55 to 45 margin. As Bush won by 3 not 10 points, this seems a bit odd! More Bush voters decided to vote for Obama than Kerry voters for McCain but given the polls we had seen and other evidence that was expected. This effect was largely canceled by the imbalance of Bush and Kerry voters.
My theory is that the long lines at democratic leaning precincts may have reduced the democratic turnout and led to this oddity. In addition, perhaps the long lines reduced turnout among new voters (including youth voters) as well. If the Kerry/Bush ratio reflected the actual 2004 results Obama would have won by more than 10 points! If the enthusiasm gap between the democrats and republicans led to an even turnout between Bush and Kerry voters, the lead would have been more than 13 points! This victory could have been even greater had the youth vote increased by a larger amount.
Dick Morris attributes the lack of a blowout to McCain over-performing among white women and a lack of increase among youth turnout. He attributed the victory by Obama to increased turnout due to acorn and move-on registering and bringing a large number of voters to the polls. I think he may have the situation backwards. Obama's margin of victory can be directly attributed to the 34 point margin among the under 30 voters (among those over 30 the election was very nearly tied). Obama's underperformance in other ways is not likely attributable to an attraction to Palin but rather a lack of turnout among former Kerry voters.
Update: A large number of people think this is a problem with the poll. Some have commented that many Kerry voters might have voted absentee/early. It should be noted that those should be included in the poll. (Of course the way this is done might not be perfect). Are there other problems people see with the poll?