Turnout among the voting eligible population in 2008 was 58.7%, and turnout in 2010 was 41.7%: a population-wide midterm drop-off of 29.0%. Among voters under 30 (if turnout parallels 2010) drop-off is estimated to be 46.8% or 9.6 million--enough to decide several elections including in the Senate and the House.
Less youth are registered to vote during midterm elections (~50% versus ~60% for Presidential elections), but the greatest discrepancy between midterm and Presidential years is clearly in turnout. For Presidential elections, 4 in 5 registered youth voters actually go vote, but in midterm elections, only 1 in 2 vote:
Further:
[According to exit polls] the 2010 youth electorate was largely a subset of the 2008 electorate: 13% went to the polls for the first time, meaning that 87% of youth were repeat voters. This suggests that the 2010 election did not bring out new young voters, nor did it sustain the entire young electorate of the 2008 presidential election.This suggests that it may be most effective to target midterm drop-off youth voters. Campaigns certainly do this but growing the electorate through voter registration has long been a Democratic priority.
Source: CIRCLE. "The Youth Vote in 2010: Final Estimates Based on Census Data."