BitJam:

Thanks for the great article Mark. Here is a minor correction. You said:

Dots plotted near the lower left corner of the chart have a bigger undecided number, while those closer to the upper right have a smaller number of undecideds.

It would have been better to have said:

... while those further away from the lower left have a smaller number of undecideds.

For example, on a square graph, if one candidate has 100% and the other has 0% the point would be equidistant from both the lower-right and upper-left corners but the number of undecideds would be zero.

Posted on April 17, 2008 3:18 PM

 richard pollara:

Terrific article and very insightful on the late breaking undecided voters. You have to assume that the Clinton people have figured this out as well. That is why I think their debate strategy last night was so brilliant. They have framed the argument to be about Obama's negatives (Wright, bitter, flag, Ayers et al). As the late breakers make their natural move to Clinton, it will not be seen as finding a safe home but a rejection of Obama because of the news of the last 30 days. The argument the Clintons will make to the Superdelegates is that Obama's negatives are insurmountable. The empirical proof is the Pennsylvania vote. It is a gutsy and risky strategy. Her negatives have gone up dramatically in the last month. But given Obama's stunningly bad preformance last night (an MSNBC commentator called it Wimpy this morning. MSNBC!!!) it may well work.

Posted on April 17, 2008 4:35 PM

 BitJam:

@richard,

It was the debate "moderators" who decided to make the first half of the debate all about anti-Obama questions. I think they really helped out Clinton by doing so but it is insulting to call them "Clinton people" because of it.

Posted on April 17, 2008 7:38 PM

 richard pollara:

Bitjam: I am not calling the moderators Clinton people. I was talking about her strategic advisers and debate prep team. They obviously knew, given the publicity these issues have had over the last five weeks, that they were going to come up. She could have elected to play nice and turn the other cheek. That would have been the safer course. Instead she chose to be tough. I think she has now defined the terms of an Obama loss in PA. It was a really smart play by the Clinton people.

Posted on April 17, 2008 8:43 PM

 cinnamonape:

Mark,

Maybe you could include anotrher line on the chart that shows the actual line of total vote differentials necessary to win. That is this would run from the upper left to the lower right from Obama 60%/Clinton 40%...through the 50%-50% point and down to where it's Clinton 60%/Obama 40%. That way the parallels away from that would show the undecideds. It would be pretty clear that the greater the number of undecideds the closer the polls.

Interesting that the top 4 polls for Clinton all fall along a parallel line to each other along that axis.

I can see some merit in Hillary's campaign stressing Obama negatives to sway the undecideds. But that assumes that the undecideds are more swayed by the claims than the negatives associated with attacks. As she has atrtacked her negatives have climbed.

It also may impact turn-out. If only strong supporters come out while others say "a pox unto both your houses" it could actually be detrimental. I think that there are a lot of intangibles involved, but perhaps her campaign has done some assessments of the variable effects of these factors and found it to be a net gainer being nasty.

As well, the debate last night was front-loaded with attacks, so I wonder hgow many people stayed around to watch the second half...in which Hillary was the more poised of the two candidates. Obama seemed a bit rattled.

Posted on April 17, 2008 10:24 PM