The US election that is. It’s been a thrilling three weeks. We’ve seen the Palin/convention bounce that put McCain ahead in the polls for about ten days or so, the Palin effect fade and, as the financial crisis hit over the past week, Obama move to new highs. It’s been like riding in the front car of a roller coaster (which I, unsuspectingly, did the first time, in Houston).
Further analysis. When the third party candidates are taken into account–and, as 1992 and 2000 showed, it’s not clever to ignore them–Obama’s lead has been wider since the two candidates crossed in the polls about a week ago. Today, Obama’s RealClearPolitics average lead is 4.2%, but 4.8% with the third party candidates added in.
And this trend is likely to continue. Today, the Rasmussen Reports daily Presidential Tracking Poll (which tends slightly to favour the Republican candidate) has Obama 5 points ahead.
Most importantly, the tide is shifting in Obama’s favour in the states where he needs the votes most. In recent days, RealClearPolitics has moved Minnesota, New Mexico, Colorado and Michigan from toss up to “leaning Obama.” Bush took New Mexico (5 Electoral College votes) and Colorado (9 EVs) in 2004.
Today Rasmussen shows McCain’s lead down to a single point in Ohio and Florida. Best of all, Obama has raced to a 5-point lead in Virginia, solidly Republican in recent presidential elections, but a top Obama target with 13 EVs. Even giving McCain Ohio and Florida, RealClearPolitics Electoral College Map with no toss-ups has Obama ahead 286 EVs to 252. Pollster.com has Obama at 229, McCain at 174 and 135 hanging in the balance.
There has been some movement to, or continuing support for McCain in states which weren’t likely to be competitive, where polls are rarely taken before the campaign proper, and where there aren’t many EVs at stake: Montana, West Virginia and Maine.
What next? Some argue that with all this volatility, today’s debate has emerged as a “crucial test“. But they always say that, despite previous debates having limited impact. Close analysis of past elections’ data reveals that the effects of the debates are in the order of two to three percentage points. In a race this close, those couple of points could make all the difference.
Tags: US elections
Jafapete's Weblog 
Leave a Reply