Ga. Dems ought to have no chance of winning, and yet ...

Considering Mitt Romney's eight-point win in 2012 in Georgia, the GOP's lock on statewide office, the haughty confidence of GOP insiders going into this election year and the fact that nationally, the Republican Party is expected to do quite well in the 2014 mid-terms, Georgia Democrats ought to have no chance whatsoever in races for governor and U.S. senator.

Yet somehow, in poll after poll, numbers like this keep popping up, in this case from SurveyUSA:

US SENATE:

Nunn: 45 percent; Perdue: 46 percent

GOVERNOR:

Carter: 45 percent; Deal: 44 percent

According to the Real Clear Politics average of the last five polls in Georgia, the governor's race is a dead heat, with Deal up by 0.4 percentage points. The RCP average in the Senate race gives Perdue an advantage of 3.4 percentage points, but that includes a poll by InsiderAdvantage that puts Perdue up 10 points, which appears to be an outlier compared to other polls.

By rights, though, it shouldn't be this close. I wrote months ago that Democratic victories in the two races were plausible but unlikely, and that if Nunn and Carter finished within five percentage points of their GOP opponents, they would have accomplished a lot in moving Georgia toward competitive status by 2016 or 2018. Even now, I'm still a little skeptical that Nunn and Carter are doing as well as the polls indicate, but that skepticism erodes a bit more with each new batch of numbers showing the same basic thing.

If Nunn and Carter are this close, this late in the race, I don't think they're battling for mere moral victories any longer. The real thing is within their grasp, particularly if either of their opponents stumbles in the last few weeks.

From the beginning, Deal was the more vulnerable of the two Republicans, and the news continues to break bad for him. He has been dogged by ethics problems since his days as a congressman, and his penchant for mixing his private business, his family business, his campaign business and his public office into one amorphous operation, with no clear lines of separation between them, remains a problem. It's never good news for a politician when a jury decides that there's been a major political coverup on your behalf, awarding a whistleblower more than $1 million in damages and fees, and when taxpayers are forced to cough up another $2 million to settle additional cases out of court.

On the economic front, things aren't any better. Deal has based his campaign on the claim that he has turned Georgia into the number one place in the country to do business, but that's a hard line to defend when jobless numbers rank the state 50th out of 50. Deal hasn't been a stridently conservative governor along the lines of a Sam Brownback in Kansas and Scott Walker in Wisconsin, and on issues such as criminal justice and juvenile justice reform, he can boast of real progress. But on the whole he has little record of success and offers no sense of vision for what he would accomplish in another four years. Like Gov. Sonny Perdue before him, he has been content to let the state drift, and it is not drifting in a good direction.

To date, Carter has done a good job of highlighting those weaknesses. I don't buy all of his policy proposals -- his recent call to use state pension funds as a source of venture capital, for example, is worrisome. Such a program offers a huge temptation for corruption, and nothing in Georgia's good-ol-boy political culture nor in its system of ethics enforcement offers reassurance that the temptation could be avoided.

What Carter does offer is an understanding that the status quo is not working, that more of the same from the folks under the Gold Dome is going to produce more of the same. After 12 years of such gubernatorial "leadership" under Perdue and now Deal, that's not acceptable.

In the Senate race, neither Michelle Nunn nor David Perdue is as well-defined as their counterparts in the governor's race. That lack of record and definition allows them to present themselves to voters as whatever voters want them to be, and both have done so. From the beginning, Perdue's embrace of a hard-core conservative persona did not seem convincing, and I had hopes that if elected, he would at least be a moderating influence on his party in Washington. But as time goes on, what was once an impersonation of an uncompromising hardliner seems to be hardening into reality, testament to the fact that if you keep on saying things that you don't really believe, eventually you do come to believe them.

From the beginning, I've also had doubts about Nunn's strategy of amorphous centrism, particularly after the leak of that strategy in all its details, but she has carried it off more effectively than I predicted. Early in the campaign, for example, she was close to disastrous in trying to find and hold a middle ground on ObamaCare. Now she is much more confident in explaining that whatever its shortcomings, ObamaCare is not going to be repealed and can only be improved, and that Georgia's decision not to expand Medicaid is hurting hundreds of thousands of its own citizens.

Her biggest challenge is the fact that in a race for the U.S. Senate -- a race that could well decide who controls the Senate -- it is much more difficult than it is in a governor's race to separate yourself from a party and a president that are deeply unpopular with some voters. Because of the nature of her race and the nature of her opponent, Nunn has always faced a more difficult task than Carter, and the polls continue to bear that out.

And of course, you can't discuss surprisingly tight races for the Senate and governor's office without acknowledging that under Georgia law, no one wins an election without getting more than 50 percent of the vote. If no candidate reaches that magic threshold in November, the top two vote-getters go at it once again in a runoff. History says that could prove a substantial hurdle to Democratic ambitions.

In 2008, Democratic Senate candidate Jim Martin finished just three points behind Republican Saxby Chambliss in November and also held Chambliss beneath the 50 percent mark. But in the subsequent runoff, Chambliss won by 14 points. That's a daunting but not insurmountable reality for both Nunn and Carter.

As a further complication, a runoff in a state-level race such as governor would be held Dec. 2, while a runoff for Senate wouldn't occur until Jan. 6, 2015. And if that Jan. 6 runoff helps to decide which party controls the Senate for the next two years ... oh brother.

One final point:

It is of course all about the turnout. Republicans typically turn out better than Democrats in midterms. White voters typically turn out better than black voters in midterms. And both statements are even more true in runoffs, should we get to that point. Democratic-leaning groups are attempting to offset that with an ambitious statewide voter-registration drive among minority voters and other under-represented groups, and are having some success. However, I suspect its impact may be amplified by the GOP response to those efforts, including greatly exaggerated claims of widespread voter fraud.

Those are perceived in the black community in particular as an effort to suppress its participation. From what I can tell, it has provoked anger that is real and pervasive, and may end up boosting black turnout.