Discussion»Statements»Rosie's Corner» National polls don't mean anything but still they're interesting. Head to head the deadhead loses in every match. Surprised?
I note you cite one poll (singular), not multiple polls.
Without knowing the methodology or history of the poll, it's difficult to know if it has any systemic biases. For example, Rasmussen polls consistently skew more pro-GOP than other major polling firms.
Hence, it's possible both RosieG and you are equally "correct."
Rosie--the polls just quoted on Morning Joe are the one I just posted.
There is a poll from the Dec 16 that is similar Trump beats all but Biden. You said the polls results are the same for all candidates so that leaves the USA Today poll (Dec 17 also the newest poll)
I think polls may have been more interesting in the past when the US electorate was less strongly polarized and the way real-world events shifting polling had a significantly stronger relationship to how people would actually vote.
Also, unless presidential election polling is weighted according to Electoral College votes, its predictive power is very poor. I once did the mathematical exercise and figured out a candidate could "lose" the popular vote something like 80 million to 37. Not 80 million to 37 million, but 80 million to 37, as in one more than 36 and one less than 38, and still be elected President. If every voter in the 37 least-populous states in the United States stays home, except for one voter in each state who all vote for the same candidate, that candidate wins, no matter how many people in the remaining 13 states vote for someone else.
Of course, in the real world, we don't see those extremes. But as the 2016 election showed, the polls can miss small fluctuations at the state and local level which have large Electoral College effects.
my2cents 