Half Sigma has piqued my interest of late. He makes really bold claims, above and beyond what a cautious reading of data might imply, but, at least he consults the data. That's more than can be said for most political blogging, which seems to draw from the bullshit of the heart. Inductivist is worth a go too (seem's that he's Half Sigma's GSS dealer that started him on the habit). Some of the sample sizes are way too small for my taste, but at least they deal in samples which have an N you can criticize. Read them like you'd read Peter Gammons (I never liked Gammons actually, I was always a Dick Schaap Sport's Reporters kind of guy, but Schaap is dead and I don't have a TV).