I think you are referring to the Gallup polls ~ and this last year they were biased beyond belief ~ possibly inadvertently, but maybe vertently! Anybody who’s been making business decisions based on Gallup data probably ought to reconsider ~ seriously ~ check http://www.freerepublic.com/focus/news/2965660/posts?page=155#155
I was referring to Gallup, among others (including Rasmussen and internal GOP polls), which had Romney with a slim lead in the national popular vote the day before the election. Considering that Gallup's bias is usually in favor of the 'Rats, a slim Romney lead even in Gallup might logically be interpreted as a slightly larger lead in real life. Only problem is that Gallup, nor anyone else doing the polling, had any mechanism for accounting for the Dummyfraud seen on a larger basis than in previous presidential elections.
No wonder why the Romney campaign insiders were reported as "shellshocked" when the alleged vote totals came in on election night.