Tom Bevan over at RCP asks:

Let me see if I have this right: there have been six polls of the Massachusetts Senate race released in the last 24 hours, five of which show Scott Brown with leads of 5 points, 7 points, 9 points, 10 points and 10 points, respectively. Only one poll shows the race tied. None of the polls show Martha Coakley with a lead.
How then, you might ask, can the Boston Globe justify characterizing the race as a “dead heat?”

The short answer: You can’t. He adds later on:

I’m sorry, but citing the most favorable poll by far for Coakley and suggesting that it is “consistent” with the other polls is such a blatant distortion of the current state of the race that the Globe should be ashamed of itself. That would assume, of course, that the Globe is interested in providing its readers with the truth as opposed to putting its thumb on the scale in favor of a preferred political candidate.