Was there a conspiracy about voting machines in New Hampshire?

A good friend of mine, Robert G. Hansen, speculates whether there is some systematic impact of voting machines on Hillary Clinton's share of the vote in New Hampshire. His discussion reminds me a lot about the discussion that took place after the 2004 and 2000 elections. Bob finds some evidence of a difference in Hillary's share based on whether there is a "machine count" or a hand count. I would have done the empirical work differently and seen if there was a difference in the ratio of Hillary's actual share to her share in the polls with the type of vote counting method. It seems necessary to account for differences from the poll and to see if there is anything systematic in how wrong the polls were in predicting things with respect to the machine counting. All that said, anyone who has read my writings on all this knows that I am very dubious about anything shady going on here.