It is certainly possible.  But these results would contradict not only other state polls, but national trends as well.  Hillary may win these states, but I have a hard time believing by these number.  12 points in CA?  I think not.  11 in Missouri?  No way.  16 in MA?  Unlikely.

SUSA is ranked 14th among the major polling organizations in terms of accuracy, well behind Zogby, CNN and ABC, all of which paint a prettier picture for Obama.

with the respondent pressing a button.  Ghat might account for their inaccuracy

It's cheap, and sometimes effective.  But it's very "hit or miss."  They're either right on the money, or WAY off.  My hope is that they are way off, obviously.

for the Virginia Senate election 2006 -- with Webb at 52% and Allen at 44% the day before the election.  

That would have been nice.

Rasmussen nailed the Virginia election at 49%-49% -- 100% on the mark.

The main problem with SurveyUSA is that it does a very bad job of capturing undecideds and leaners.  The New York and Missouri numbers look right.  Massachusetts may play along similar lines.  I think California will be close to a 50-50 race.  

Compared to the last SUSA poll in California, Hillary ticked up 3 points (which is within the MOE).  So that's not surging, that's staying the same or moving up slightly.

http://www.surveyusa.com/clien...