I am rethinking things after this year's elections. The problem in Georgia is that I don't think black voters are going to remain this D after Obama. And the white voters moving in are much more likely to be culturally Midwestern than in FL, VA or even NC. So I just don't see a D path in Georgia with an 85/15 black vote, a 65/35 Hispanic vote and a 55/45 Asian vote, which is probably where we are heading post-Obama. The more libertarian positions D's will need to take to hold on in CO/NV/VA etc. will eventually turn off some moderate minorities.
On the other hand, culturally Western white voters were one of the only Dem bright spots this year, and AZ is acting like a tilt R purple state for downballot D's. Obama did some unique things to alienate the state, so once he is out off office, D opportunities should only grow there.
Exit polling both in 2010 and 2014 suggested that black support dropped a bit, but only to where it was in Georgia pre-Obama, which is 90% Democratic. Yes, that'll impact things considerably, but consider that even with that drop, Nunn basically got the same percentage of the vote that Obama got in 2012, and 3-4 points more of the white vote during an election with abysmal turnout. I'm still waiting to see all of the county-by-county breakdowns on racial turnout (should be available from SoS soon), but it appears that the biggest drop-off in this cycle was among white, progressive, younger voters.
I've seen two sets of estimates for white D support in 2010 - one showed 20% and the other showed 23%. It's important to consider that the white level of support in this very low turnout election (more so than 2010) was at or above 2010 levels. Considering the state became two points more Democratic when compared to 2010, I tend to think the closer estimate for 2010 was whites @ 20% Democratic. I think there's a much bigger increase (among whites, mainly) waiting to be seen in 2016 with full turnout and a (presumably) non-black nominee.