The stakes are high for this last of the big Democratic primary days. If Clinton surges ahead, she will undeniably be a serious contender again, and if Obama maintains his lead, he can cement his status as the frontrunner, and push Clinton further to the margins.

Using the normal methodology (averaging national and state-focused blog results), the blogosphere predicts a very narrow Clinton lead in both states. It is with some reservation, however, that I post this as a final prediction, because of several mitigating factors– the primary reason being that if only Ohio and Texas blogs are taken into account, the numbers are even closer, but Obama maintains a small lead in both.

I believe a lot of Clinton’s traction in the national blogosphere comes from her recent media blitz—SNL, The Daily Show, et cetera; as well as Rush Limbaugh’s recent advice for Republicans to vote for her in the primaries. The question is if all this national coverage will actually influence voters in the states.

Our methodology for blogosphere primary predictions worked much more consistently early on in the races than later (and in general, Republican primaries were also easier to predict). One reason for this might have to do with the saturation of the data during the multiple races on Super Tuesday. But a more interesting hypothesis is that the national trends, coverage, and dialogue, are no longer as big of influencers on the state-level political process. Perhaps more and more people are ignoring the national pundits and TV appearances, and focusing on the local campaigns, speeches, and issues.

The two-data-point methodology was put in place to ensure that the small sample of local political blogs was not skewed, and it was balanced out by the national conversation. Maybe though, as the races go on and on and are focused longer on particular states, the national conversation is becoming less relevant to the results. Or perhaps, after the saturation of Super Tuesday, the collective opinions of the national blogosphere will be more accurate than a smaller local sample.

Even though the aggregate data predicts a Clinton win and the numbers are very close, I’m going to go out on a limb and put my own personal faith in the state-based projection– and call both races for Obama. As the race goes on, I believe the importance of national punditry in state races is diminishing.

But regardless of the end results, tonight will certainly be interesting.

(We ran data for the almost-settled Republican race as well, showing an overwhelming McCain lead—69% in Texas, and 81% in Ohio.)

Tags: Analyzing Consumer Generated Media, Blog Analysis, Mass Media/Culture //