Lately, Obama's been getting some numbers in the polls that reveal he's losing the bounce from his 2012 DNC. Indeed, Nate Silver from the New York Times predicted that the President would lose his edge near the end of the month. Here's Silver's model predicting how the Convention bounces for each party will fade after two or three weeks:
It's clear from looking at the curve that Romney's bounce from the Convention would end soon because the Democratic National Convention followed the week after the Republican National Convention. What is also seen in the image is that the model predicts Romney to lose his Convention bounce by the start of the 2012 DNC. I should mention that the NYTimes blogger is accounting for inflated poll numbers, so he accounts for any inflation in the polls by subtracting the average bounce as given from previous polls from Presidential Elections. What that means is that if the image above shows Obama is ahead by two or three points in the FiveThirtyEight forecast, then the real poll numbers are higher because Silver is subtracting any misleading noise that results from polling within two or three weeks after a party Convention. In short, Silver is holding Obama's poll numbers to a higher standard because the DNC is providing a boost to the incumbent's polling.
Oddly enough, Obama's decline started last Friday when the industrial production numbers were released (Silver takes into account the industrial production number because that number ties into the economy--which voters look at when their casting their votes). Then the following day Silver posted that Obama's number held steady on Saturday, but declined on Sunday. The decline is expected by the FiveThirtyEight model and should not come off as a surprise. In fact, looking at where Sunday's date falls on the graph (9/16), Obama is half way through his decline, meaning, the numbers will get worse for the President. What the graph also shows is that Obama's bounce from the DNC will dissipate around next week on 9/28.
As of this Monday the polls held steady from the day before, however, on Tuesday the President took another hit to his poll numbers. What's more interesting about the numbers that Silver released this Monday is that he confessed that this election could be a repeat of the 2000 Presidential Election, wherein one candidate won the popular vote and the other candidate won the Electoral College. What this means is that the next president will be the winner of the Electoral College, not the popular vote--given historical precedence. Moreover, the NYTimes blogger points out that the incumbent could win the popular vote, but lose the Electoral College.
No comments:
Post a Comment