poblano, for those who have lived in a hole the past couple of months is a blogger with a background in research (but not politics) whose posts can be found at dailykos, and now at his own site fivethirtyeight.com . He built from scratch over the course of the primary season a model to predict outcomes of these contests and analyzed all sorts of possible factors. He eventually settled on the conclusion that just a handful of factors are necessary to completely predict results,  and  his model has been entirely horserace-blind. His results have been quite good. In fact he has routinely outperformed most polling organizations, and vastly outperformed the majority of bloggers who shape the web's CW and who are in turn shaped by it.
his last 3 predictions:
PA: Clinton by 7.4  - actual Clinton by 9.3
IN: Clinton by 2 - actual Clinton by 1.5
& NC: Obama by 17.2 - actual Obama by 14.8 (15.0 head-to-head)

This post by Kaus over at Slate captures the audacity of his last two predictions brilliantly:

"Obama by double digits" in N.C.:
Predicted by a blogger using a sophisticated model that ignores ...
what's been happening in the campaign. Like Rev. Wright. I predict this
person is wrong! ... Update: He was right. ... [via Insta] 9:27 P.M.

So here's my question, Josh. You are one of the most thoughtful voices on the big picture questions of the race. Early on, there was a mini-civil war between the momentucrats and the numerocrats (which the numerocrats seem to have handily won), but what I see now goes much further than that and questions the fundamental nature of the horserace interpretation of politics.  I see a guy able to make predictions of who a region will vote for and how strongly by looking at boxes they check on a census and historical political leanings of their location. When he predicts more accurately than by actually asking people what they think, and what's important to them, and even who they say they are going to vote for, totally blind of issues and scandals, I see that as a fundamental game-changer in how we view politics. I see this as the death of the Horserace. You have convincingly written in the past about the perils of liberals ignoring the fundamental nature of the horserace in politics. How do you explain the success of a nearly politics-blind model?

Coincidence? Particular to Obama? Particular to this primary, or this year? Mass hysteria? Very interested in what you think.