Reading this Matt Yglesias post, and noting that the two latest polls out of Pennsylvania (1, 2) show a 13 percent Obama lead, to believe that John McCain will win in Pennsylvania -- indeed, to believe he has any chance of winning in Pennsylvania -- you must reject the basic concept of random statistical sampling. And sure enough, there are a lot of people who trust their gut sense that Pennsylvania is conservative more than they trust the fact that most polls show a double-digit lead for Obama in the state. Political commentary does not attract data-driven types. But if Obama does win Pennsylvania by 13 points, the media should have to answer for why they ever covered the state as if it were competitive. Surely the "the McCain campaign said so" is not enough.
