The PUMA defectors would have supported Hillary.
It is interesting that you claim Stevenson lost support because he was thought of as a clown. I thought his more common perception was as an all too serious policy wonk a la Dukakis.
By many, Stevenson was viewed as the Oscar Wilde candidate, an image that was encouraged when Nixon described Stevenson as Sidesaddle Adlai, adding that "like all sidesaddle riders his feet hang well out to the left." However, the overwhelming reason Stevenson lost in 1952 and 1956 was Ike's enormous popularity. I cannot recall that Stevenson was seen as a Dukakis-type policy wonk. Unlike Dukakis, Stevenson was, in my reading of history, most widely remembered as the wittiest man who ran for the U.S. Presidency in the last half of the 20th century. He was, on the occasion I mentioned, when he held out the prospect of Ike's dying during his second term (and Nixon succeeding him), thought to have been altogether too serious. As Ambrose noted in this connection, "What most people found distasteful, however, was what Stevenson said, and the way he said it."
".... websites have sprung up to denounce Democrats for picking Obama. One - called PUMA, an acronym for the sentiment "Party Unity My A**" - features postings by Clinton supporters saying they will never vote for Obama, even if it means electing McCain."
There is no suggestion they will vote for McCain. They should certainly be included in the data analysis that prompted my first post in this thread, BUT if, as seems to follow from the above quotation, they simply will not vote at all, I guess they will not be included in 'exit polls' and will not register in the "expanded" likely voters group I pinpointed in my first post.
Granting that they might vote for McCain, they should certainly be included in the sort of analysis I mentioned in my first post. They constitute a 'sore loser' group and it will be of interest to know if such a group could have significant influence. For example, if McGovern had not been nominated in 1972, the 'sore losers' would still have voted against Nixon in all likelihood. It will be of interest to know what proportion of the 'sore loser' group that supported Hillary would vote against Obama, in which case, if the Dems lose, they might, in a close election, be accused of scuttling a Democratic victory (cf. Nader) and Hillary would likely be toast in 2012.