Nate Silver agrees with me that much of that shocking 2% swing can be explained by systematic differences between sample and population: survey respondents included too many Clinton supporters, eve…