Posted on | February 20, 2012 | 8 Comments
Yes, we will be resuming the Federal Review Composite Poll this year. As it has always been, it will be a meta analysis of polls, state-by-state. Unlike the pedestrian Real Clear Politics average, it will be based upon a weighted analysis of polls based upon margin of error, sample size, respondent type. We have done this since 2000, and have predicted the winner in each of 2000, 2004 and 2008, only getting 2 states wrong – but noting those were toss-ups (Wisconsin in 2004, North Carolina in 2008). So, stay tuned.
Based on our historic analysis, we believe that a Romney / Obama matchup would yield a Romney victory in excess of 300 electoral votes. But it’s early. Way early. Santorum could compete, but it would be a nail biter. When we get closer to having a GOP nominee, we’ll start running the numbers and posting our analysis.
In the meantime, why don’t you predict how the battlegrounds will come out. Ohio. Pennsylvania. Michigan, Virginia, North Carolina, New Mexico, New Hampshire, Iowa, Wisconsin, West Virginia, Nevada, Indiana, Tennessee, Arkansas. Florida.