Polls that are published during the party conventions can be anticlimactic, representing old news since they won't fully reflect the effects of the convention bounce. And none of the surveys that came out on Wednesday were that newsworthy to begin with â€" although there was one, in Ohio, that had encouraging news for Barack Obama, and another, in Nevada, that was slightly favorable for Mitt Romney.
The Ohio survey was conducted by the automated-polling firm Gravis Marketing, which had Mr. Obama ahead of Mr. Romney by less than 1 full percentage point.
A one-point lead isn't much, and Mr. Obama has gotten some better numbers than that in Ohio. So why does this qualify as good news for him? Because this firm has had Republican-leaning results in the other states that it has polled, putting Mr. Romney up by 2 points in Florida, 1 point in Colorado and 17 points in Missouri, making it several points more Republican-leaning than the consensus of surveys in those states. Once the model adjusts for the firm's “house effect,†it treats Mr. Obama's nominal 1-point lead as being the equivalent of a 4- or 5-point lead instead. Thus, Mr. Obama's chances of winning Ohio rose somewhat based on the survey.
The broader point is simply that Ohio is so important to the electoral calculus that it's good news for a candidate when a polling firm shows him doing relatively well there compared with the other states that it polls. Ohio has a 30 percent chance of being the tipping-point state, meaning that it would cast the decisive votes in the Electoral College. That's as much as the next two states on the list, Florida and Virginia, combined. It's also as much as Colorado, Nevada, Iowa, Pennsylvania, New Hampshire, Michigan and North Carolina combined.
All of these states are competitive. But really, they exist along a continuum of electoral power rather than falling into binary categories of †œimportant†and “unimportant.†Ohio is at the extreme end of that continuum.
The reason our tipping-point calculus rates Ohio so highly is because it would usually suffice to provide Mr. Obama with a winning map, even if he lost many of those other states. If you give Ohio to Mr. Obama, plus all the states where the forecast model now estimates that he has at least 75 percent chance of winning, he's up to 265 electoral votes. That means he could win any one of Colorado, Virginia, Iowa, Wisconsin, Florida or North Carolina to put him over the top.
Mr. Romney is fortunate in this sense to have put Wisconsin squarely into play with his selection of Paul D. Ryan as his running mate; it gives him a few more ways to win without Ohio, although it would still be a daunting task.
…But Don't Forget Nevada
Another state that could be critical to an Ohio-less winning path for Mr. Romney is Nevada. I did not mention it in the list above since the model giv es Mr. Obama exactly a 75 percent chance of winning there. But Mr. Romney's numbers have been a bit better in Nevada lately, including in a Public Policy Polling survey, released on Wednesday, that put him behind Mr. Obama by 3 points there.
Obviously, Mr. Romney would prefer to be ahead in Nevada than behind by any margin. But that is a better result for him than the previous Public Policy Polling survey of Nevada, which put him down by 6 points instead. The firm's surveys are also somewhat Democratic-leaning, although they have been moving closer to the consensus recently.
The pessimistic case for Mr. Romney in Nevada is that the polling spread has been narrow there compared with other states. In the 12 Nevada surveys in our database, Mr. Obama has never trailed â€" but nor has he led in any survey by more than 8 points. A tighter spread in the polls makes a smaller lead more robust, as does the fact that there are very few undecided voters in the Nevada polls.
Still, Nevada is a state that has produced some poor polling in the past, and between a potentially strong Mormon turnout and the state's dilapidated economy, Mr. Romney has some angles to work there.
No comments:
Post a Comment