We also have a new Rasmussen poll for Massachusetts, which -- this is a bad habit of Rasmussen's -- has three different margins listed for Clinton: she's listed as a 15-point favorite over McCain in the text of the survey write-up, a 21-point favorite in the adjoining table, and a 19-point favorite in the headline on Rasmussen's front page. We'll go with the 19-point number but really it doesn't matter; she's not going to lose the state. Let me give you the numbers and then some discussion:
|AL||Survey USA||3/15/08||McCain +27||McCain +18|
|CA||Survey USA||3/15/08||Obama +14||Clinton +18|
|IA||Survey USA||3/15/08||Obama +6||McCain +4|
|KS||Survey USA||3/15/08||McCain +12||McCain +19|
|MA||Survey USA||3/15/08||Tie||Clinton +13|
|MA||Rasmussen||3/18/08||Obama +7||Clinton +19|
|MN||Survey USA||3/15/08||McCain +1||Clinton +3|
|NM||Survey USA||3/15/08||Obama +6||Clinton +6|
|NY||Survey USA||3/15/08||Obama +8||Clinton +13|
|OR||Survey USA||3/15/08||Obama +9||Clinton +6|
|VA||Survey USA||3/15/08||Obama +1||Tie|
|WA||Survey USA||3/15/08||Obama +11||Clinton +5|
|WI||Survey USA||3/15/08||Obama +4||Clinton +1|
In general, Obama continues to run stronger than Clinton in the places where he usually runs stronger than her, although the difference is narrower than in other recent rounds of surveys. Obama fares better in Iowa, Kansas, Oregon, Virginia, Washington, and Wisconsin -- in some cases by small margins -- while Clinton does better in in Alabama, Kansas, Massachusetts (in both polls), Minnesota (this is a reversal from most polling data) and New York. The two Democrats do equally well in New Mexico.
Although my headline is a little melodramatic -- the model now lists John McCain as a 51/49 favorite over Barack Obama, and a 57/43 favorite over Hillary Clinton -- this is actually not a terrible set of polling for either Democrat, showing the Democrats winning a variety of purple and blue-tinged states at one of the worst times in the news cycle for them. Interestingly, Obama has now dropped below 50% in his overall win percentage even while retaining a slight popular vote advantage over McCain, but at other points in the cycle the electoral math has favored Obama and I'm not sure what can be concluded from such trends.
The one state that I have the toughest time figuring is Massachusetts. Our regression model thinks that Massachusetts should love Obama, who fundraised very well there, but he's yet to establish a double-digit lead in any public poll of the state. I have a pet theory that there may be a sort of subconscious association among Bay Staters between Obama and Deval Patrick, who is not particularly popular in the state.