The results aren't too surprising; to some degree, site popularity seems to be driven by a paper's circulation and reputation, as well as the size of the area they serve. So up in at the top you find the Times and USA Today and the Washington Post and the LA Times.
A little hometown shout-out for the Houston Chronicle, which came in fifth. The Chronicle is not a great newspaper, largely because of the economic challenges all papers who aren't the big nationally known elite face: they don't spend enough money on reporting, so they wind up relying heavily on wire services and articles from other papers for things that you'd expect a newspaper in the country's fourth-largest city to cover on their own.
They have, however, been very aggressive in using the web, and as a result their site is dramatically better than the paper itself. They've also been way ahead of the curve in incorporating user-generated content: a section of reader-written blogs has morphed into a community space where readers can not only blog but post pictures and interact with one another.
(Full disclosure: I was one of the initial crop of Chronicle reader bloggers (writing about politics) and still blog there.)
That seems to be paying off in site traffic; I note that the Dallas Morning-News site, produced by a generally superior newspaper in a slightly larger metro area, is way down the list. (Deservedly so - it's pretty mediocre.)
You can see signs on this list of who's doing well adapting to a new media environment, and who's not. We haven't reached the winnowing phase yet - when paper publishing simply doesn't make economic sense anymore - and I don't think that's coming too quickly. There's still too much appeal to paper. But we may start seeing some victims among papers who aren't keeping up with the web.