So what do we know after two days of early voting?
Not a lot -- but something.
We know turnout was huge during that first weekend -- almost 60,000 voters turned out, 15,000 more than in '08 -- and we know that the second week is usually much bigger than the first (it was 50 percent greater in '08).
We know the Democrats are getting out their voters -- better than four years ago, in fact.
We know the Republicans are turning out much better than in '08 (a very low bar because John McCain had abandoned the state and it was a wave election for the Democrats) -- but are handicapped by a large (127,000) voter registration deficit and thus have a smaller pool to draw from.
It is silly to extrapolate from a couple of days. But, on the other hand, we have to figure both parties were trying to show off on opening weekend -- and Democrats have a 13,000-vote lead in Clark. It is likely that the percentage edge will diminish every day -- but the raw vote edge could be formidable if the GOP can't turn around the pattern of the last two cycles.
The real question, of course, is whether both candidates are holding 90 percent of their voters and where the unaffiliated or third-party voters are going. If, as some data suggests, President Obama is holding nearly all Democrats, Mitt Romney will need to win independents in a landslide.
Finally, what is the predictive power of early voting once all of the data is in?
In 2008, Democrats had an 83,000 raw vote lead in Clark County. That was good for a 52 percent to 31 percent. When the first numbers were posted on Election Night, including early and mail ballots, the lead for Barack Obama over John McCain was 102,000 votes in Clark -- 61-37. So not only did Obama hold the Democrats, but he picked up indies and, perhaps, some crossovers. That was, as I have said, a wave election. Obama won Nevada by 12 points.
In 2010, the Democrats had a 23,000-vote lead after early voting -- 46-37. When the first votes were posted on Election Night, Senate Majority Leader Harry Reid had 34,000-vote lead -- 54-42. Once again, the GOP gained nothing on Election Day -- indeed, just the opposite.
Yes, it was Sharron Angle. But, on Election Day in Clark, the Republicans did not turn out their voters very well and Reid had more than the firewall he needed. And he won the state by 6 points.
So, it's reasonable to assume, because 70 percent of Clark voters likley again will cast ballots before Election Day, that there are not enough voters left to make much of a difference either way. (I also wonder with the much-hyped effort by the GOP whether there will not be as much of a difference on Election Day. The Republicans could have an advantage on Election Day and still not cut much into the firewall -- or, as they did the last two cycles, lose ground because of the Democratic machine.
So watch these numbers in Clark very closely, folks. There will be a firewall again, based on the first two days. How big is the only question -- and how resistant to a blazing GOP effort in rural Nevada and on Election Day.
Current numbers for the entire state are here.