I love the early voting statistics in states like North Carolina and Nevada that break data down by demographics.  As of this morning, 1.85 million voters had voted in North Carolina, more than half the total Tarheel voters in the 2004 election and about 30% of all registered voters.  The breakdown of these voters is 53% Democratic, 29% GOP and 18% independent.  Based on the latest Rassmussen poll in NC, these groups split 85/15, 6/94 and 60/40 for Obama/McCain, respectively (these are approximate since the Rassmussen public release doesn't break out how third-party candidates factor in).  If these numbers are right, we can calculate that Obama currently has a 57.6-42.4% lead over McCain in NC with somewhere between a third and a half of the total votes cast.  McCain will need an awfully big advantage next Tuesday to close that kind of gap...

Clinton's GOTV efforts in South Texas were phenomenal, though whether this is just a reflection her popularity or "Box 13" style South Texas shenanigans is impossible to know.  In Duval County, Hillary got 77% of the vote on turnout close to 50% of registered voters -- primary totals as high as the Texas Sec. of State has ever recorded for a general election.  In Jim Hogg County, she got 76% of the vote, again with close to 50% turnout (and setting a record for votes cast in any election).  In Maverick County, she took 72% of the 9,963 votes cast (0 votes went to GOP primary candidates), again close to an all-time turnout record.  LBJ would be proud.

Limbaugh's unsavory attempts to distort the process by urging listeners to vote for Hillary may actually have made the difference in the close Texas race.  The voters in the Democratic primary were 9% republican and 24% independent, and they voted 47% and 49% for Clinton, respectively.  Obama still won these groups, but up to this point he's had a much greater cross-over appeal than Clinton.  For example, in the Missouri open primary, Clinton got 48% of the overall vote but only 21% of crossover Republicans and 30% of independents.  Similarly, in Virginia, Clinton got 35% of the overall vote but only 23% of crossover Republicans and 30% of independents.  The increased GOP/independent support for her this time could be "the Limbaugh factor"; either way, the race was close enough for it to have made the difference. 

Makes me glad that Pennsylvania is a closed primary.