Monday, October 8, 2007

Top Four Polling Trends in the Earlies

A quick glance at the trend lines from show some of the second tier candidates lines converging and set to intersect before the 2008 mark. Of course, those trend lines are at least partly a product of whatever regression uses (I don't know what it is). The slopes of those trend lines are anything but set in stone.

Given all of that, trend lines do give us an idea of how the state of the race is changing. That allows us to speculate on what the race will look like in January if nothing changes, so that we can see which campaigns need something about the trends to change.

As I noted a couple days ago, Obama has passed Edwards for second in Iowa. He and Hillary appear to be gaining at the same rate. Edwards, on the other hand, is in a relative state of collapse since his peak in April or May of 2007. In fact, Richardson's modest growth rate may put him past Edwards by January.

This is why Hillary cannot be allowed to win Iowa. Not only is her New Hampshire lead substantial, it continues to grow at an absurd rate. I don't know what all the talk is about Obama's huge momentum in New Hampshire, since he's been in the slow decline there all year. As in Iowa, Edwards is falling apart (at least compared to the other candidates), while Richardson appears almost certain to pass him for third.

This is much more a two-way race than either Iowa or New Hampshire, but otherwise it's pretty much the same story.


Espo said...

I'd be interested to see what these graphs looked like at this point back in '03.

Jonathan Cooper said...

So how much of Edwards' decline would you attribute to the haircut?