Just as Brian says, the difference between the Obama poll level and the Obama vote total level seems to just be your basic statistical variance. The pollsters underestimated Clinton's level of support. People who were undecided as of the last round of polling seem to have gone overwhelmingly in her direction.
[also note the relevance of this to Wilder/Bradley effect speculations]
The Wilder/Bradley effect is the factor that alluded to in an earlier post, where social pressures might overstate the support for black candidates. In particular, the effect seems to be that rather than declaring their opposition to a black candidate, voters will choose the more ambiguous category of undecided. Thus, the rule of thumb among some political professionals that when you have a black candidate running against a white candidate, take the undecideds and add them to white candidate's total. Wikipedia actually has a pretty good entry on the effect.