I must respectfully disagree.
While SUSAs methodology (primarily sampling) does indeed have a slight left bias, it is a transparent methodology.
Research 2000 however has a more distinct left bias, which varies erraticatlly from mild to severe which is not a simple matter of sampling.
Ya, but Del Ali over at Research2000 always tampers with his results in really obvious ways so you can tell what the poll reallys says if you look carefully..
I'm not as familiar with Research2000 as you are.
I have seen them come up with some REALLY weird results.