Some Democrats I know are getting very jittery, and it’s not hard to see why. At the end of a week that contained a strong debate performance from President Obama, some widely-watched voter surveys still show Mitt Romney in front, and one of them—the Gallup daily tracking poll of likely voters—has him expanding his lead to as much as seven points. On Monday, according to Gallup, the Republican candidate was leading Obama by two points, 49-47. By Thursday, he was ahead by seven, 52-45. On Friday, there was little change: Romney fifty-one per cent, Obama forty-five per cent.
Obviously, these are disturbing figures for Obama supporters. But before
you climb back onto that ledge you were perched on before Tuesday’s
debate, consider the broader picture, which is considerably more
favorable to the President. Taking account of all the polls, rather than
just one, the national race appears to be a virtual tie. At the state
level, despite Romney making strong gains in some places over the past
couple of weeks, Obama’s firewall in the electoral college is
holding—for now, anyway.
Let’s start with the national numbers. The Gallup tracking poll is just one survey out of many, and, for whatever reason—sampling error, faulty weighting, an overly tight definition of “likely voter”—its findings might not be reliable. Absent some corroborating evidence, I simply don’t believe that Romney is leading by six or seven points. Virtually all the other national polls show the race much closer, and some of them still have Obama in the lead. On Friday, the Rasmussen daily tracker had the two candidates tied at forty-eight per cent, which suggests that the debate did give Obama a modest bounce—earlier in the week, Rasmussen had Romney leading by a couple of points.
Three other tracking polls that receive less attention in the media show Obama still leading. In the latest survey from Reuters/Ipsos, he has forty-seven per cent of the vote to Romney’s forty-four per cent: a three-point margin. In Friday’s update from IBD/TIPP, he has a two point lead. And the RAND panel survey, which tracks a given group of voters rather than sampling new ones every day, shows Obama with a three-point lead. Of course, these polls might be off, too. As always, I advise you to monitor a poll-of-polls that averages out individual surveys. For example, on Friday afternoon the Real Clear Politics poll-of-polls had the candidates virtually dead even: Romney 47.1 per cent; Obama 47.0 per cent. The Talking Points Memo polltracker, which juggles the numbers slightly differently, had Romney leading Obama by just half a percentage point: 47.5 per cent to 47.0 per cent.
Another fact to remember is that three days isn’t necessarily enough time to gauge the impact of the second debate. The numbers in the tracking polls are averages based on polling carried out over a number of days, so it takes a while for changes in sentiment to show up. The Gallup poll incorporates seven days of polling; Rasmussen’s is based on three days of surveys. On October 6th, three days after the first debate, Rasmussen still showed Obama with a two-point lead. On October 8th, five days after the showdown in Denver, Gallup had Obama five points ahead. If Obama is enjoying a bounce after Tuesday’s debate, we wouldn’t necessarily know about it yet.
That’s especially true if the bounce itself, not just its effect on the polls, takes some time to materialize. I think this happened after the first debate. Romney got an immediate boost from exceeding expectations, and then over the ensuing week, as the media praised his performance and lambasted Obama, he got a second lift. Since the debate in Long Island was more even than the one in Denver, Obama normally couldn’t be expected to receive much of a second wind. But this isn’t a normal circumstance. On Monday, the third and final debate will take place in Boca Raton, and it will take on aspects of an elimination contest. If Obama does well again, he could get another lift.
Now to the state level, where the election will be decided. As I noted on Monday, in my last update to The New Yorker’s electoral map, Romney appears to have established a narrow lead in some big battleground states, notably Florida and Virginia. In the past few days, that hasn’t changed. In Florida, a CNN/Opinion Research Poll carried out on Wednesday and Thursday showed Romney leading Obama by one point—49-48. A Rasmussen poll that was conducted on Thursday, thirty-six hours after the debate, showed Romney leading 51-46. And a Fox News poll released late on Friday showed the G.O.P. man leading 48-45. Rasmussen also surveyed Virginia, where it found Obama, at forty-seven per cent, trailing his rival by three points. Since Rasmussen’s surveys generally lean a bit to the G.O.P., its figures shouldn’t be taken as gospel, but the trend is hard to ignore. Romney’s five-point margin in Florida was his biggest lead of the year in Rasmussen polls there, and his three-point margin in Virginia was his biggest lead since March.
On the electoral map, I still have Virginia listed as a toss-up rather than as leaning to Romney. That’s despite the fact that the Republican candidate has been ahead in the last four polls. But in three of those surveys, his lead was just a point or two. Even the three-point lead in the Rasmussen poll is within the margin of error. If the next poll or two confirm that the Republican lead is widening, I will change my call on the Old Dominion. But for now, I still regard it as too-close-to-call. If the surge Romney has enjoyed in the past couple of weeks continues, he will probably snatch its thirteen electoral-college votes. But if Obama stages a rally, even a modest one, he could eke out a victory.
For the sake of argument, let’s assume Romney carries Florida and Virginia. According to my map, that would leave him with 248 votes, still twenty-two short of victory. If he also wins Colorado and New Hampshire, both of which I have listed as toss-ups, he would get to 261 votes, leaving him needing nine more to reach 270. That means he would have to win Ohio or Wisconsin, or Iowa and Nevada. The good news for Democrats is that Obama appears to be holding onto his leads in all four of these states. That’s his firewall.
Ohio: There have been four polls this week, and all of them have showed Obama ahead. In a poll of likely voters, P.P.P., which tends to lean to the Democrats, showed him leading Romney 51-46, a five-point margin. Another poll, from SurveyUSA, put the President’s lead at three points. Both of these surveys were carried out before Tuesday’s debate. Another post-debate poll was from Rasmussen. It showed Obama’s lead over Romney at just one point—49-48—but that is the same margin that Rasmussen has shown all month in Ohio. On Friday, Fox News released the newest poll from the Buckeye State, and it showed Obama still leading by three points, 46-43. Taken together, the polls suggest that all the money and manpower that Romney has thrown into Ohio haven’t yet made much of a difference.
Wisconsin: There have been two polls this week, and both of them have had the President on top. Rasmussen, in a poll carried out on Thursday, showed him leading Romney 50-48—a narrow margin. But a poll with a bigger sample from NBC News/Wall Street Journal/Marist College, showed Obama leading by six points: fifty-one per cent to forty-five per cent. That’s virtually unchanged from last month, when the same poll found Romney trailing by five points. The key to Obama’s big lead: he continues to win the support of far more women than Romney. In the latest poll, the gender gap is eighteen per cent.
Iowa: The picture here is a bit more mixed, but still broadly favorable to Obama. Of three polls conducted this week, the President has been leading in two of them. WeAskAmerica, which interviewed residents on Monday, before the debate, showed Obama with forty-nine per cent of the vote and Romney with forty-seven per cent. NBCNews/Wall Street Journal/Marist, which asked questions before and after the debate, had Obama even further ahead of his Republican rival: fifty-one per cent to forty-three per cent, a margin of eight points. As in Wisconsin, this poll suggests that very little has changed over the past month. In September, it also credited Obama with a lead of eight points.
The one dark spot for the White House: a P.P.P. poll released on Friday that showed Romney leading in Iowa by one point. That represents a big turnaround. In a poll taken last month, P.P.P. had Obama leading by seven points. Now he is trailing, 48-49—a statistical tie. And Romney’s favorability ratings are greatly improved. In September, forty per cent of Iowans viewed him favorably, and fifty-five per cent viewed him unfavorably. In the new poll, his favorability rating is forty-eight per cent, and his unfavorability rating is forty-seven per cent—evidently a reflection of his performance in the first two debates.
Nevada: Not much change here from Monday, when I pointed out that Romney, despite making up some ground, had failed to come out ahead in a single poll. Two polls were carried out early in the week, and both showed Obama leading by three points. In yet another Rasmussen survey, he was at fifty per cent and Romney was at forty-seven per cent. A poll that SurveyUSA carried out for the Las Vegas Review Journal showed the President with forty-eight per cent and the challenger with forty-five per cent.
That’s it for now. If things keep going like this. we’ll have to start considering the possibility of Obama doing what George W. Bush did in 2000: narrowly losing the popular vote but winning the election. That, though, is a subject for another post.
Let’s start with the national numbers. The Gallup tracking poll is just one survey out of many, and, for whatever reason—sampling error, faulty weighting, an overly tight definition of “likely voter”—its findings might not be reliable. Absent some corroborating evidence, I simply don’t believe that Romney is leading by six or seven points. Virtually all the other national polls show the race much closer, and some of them still have Obama in the lead. On Friday, the Rasmussen daily tracker had the two candidates tied at forty-eight per cent, which suggests that the debate did give Obama a modest bounce—earlier in the week, Rasmussen had Romney leading by a couple of points.
Three other tracking polls that receive less attention in the media show Obama still leading. In the latest survey from Reuters/Ipsos, he has forty-seven per cent of the vote to Romney’s forty-four per cent: a three-point margin. In Friday’s update from IBD/TIPP, he has a two point lead. And the RAND panel survey, which tracks a given group of voters rather than sampling new ones every day, shows Obama with a three-point lead. Of course, these polls might be off, too. As always, I advise you to monitor a poll-of-polls that averages out individual surveys. For example, on Friday afternoon the Real Clear Politics poll-of-polls had the candidates virtually dead even: Romney 47.1 per cent; Obama 47.0 per cent. The Talking Points Memo polltracker, which juggles the numbers slightly differently, had Romney leading Obama by just half a percentage point: 47.5 per cent to 47.0 per cent.
Another fact to remember is that three days isn’t necessarily enough time to gauge the impact of the second debate. The numbers in the tracking polls are averages based on polling carried out over a number of days, so it takes a while for changes in sentiment to show up. The Gallup poll incorporates seven days of polling; Rasmussen’s is based on three days of surveys. On October 6th, three days after the first debate, Rasmussen still showed Obama with a two-point lead. On October 8th, five days after the showdown in Denver, Gallup had Obama five points ahead. If Obama is enjoying a bounce after Tuesday’s debate, we wouldn’t necessarily know about it yet.
That’s especially true if the bounce itself, not just its effect on the polls, takes some time to materialize. I think this happened after the first debate. Romney got an immediate boost from exceeding expectations, and then over the ensuing week, as the media praised his performance and lambasted Obama, he got a second lift. Since the debate in Long Island was more even than the one in Denver, Obama normally couldn’t be expected to receive much of a second wind. But this isn’t a normal circumstance. On Monday, the third and final debate will take place in Boca Raton, and it will take on aspects of an elimination contest. If Obama does well again, he could get another lift.
Now to the state level, where the election will be decided. As I noted on Monday, in my last update to The New Yorker’s electoral map, Romney appears to have established a narrow lead in some big battleground states, notably Florida and Virginia. In the past few days, that hasn’t changed. In Florida, a CNN/Opinion Research Poll carried out on Wednesday and Thursday showed Romney leading Obama by one point—49-48. A Rasmussen poll that was conducted on Thursday, thirty-six hours after the debate, showed Romney leading 51-46. And a Fox News poll released late on Friday showed the G.O.P. man leading 48-45. Rasmussen also surveyed Virginia, where it found Obama, at forty-seven per cent, trailing his rival by three points. Since Rasmussen’s surveys generally lean a bit to the G.O.P., its figures shouldn’t be taken as gospel, but the trend is hard to ignore. Romney’s five-point margin in Florida was his biggest lead of the year in Rasmussen polls there, and his three-point margin in Virginia was his biggest lead since March.
On the electoral map, I still have Virginia listed as a toss-up rather than as leaning to Romney. That’s despite the fact that the Republican candidate has been ahead in the last four polls. But in three of those surveys, his lead was just a point or two. Even the three-point lead in the Rasmussen poll is within the margin of error. If the next poll or two confirm that the Republican lead is widening, I will change my call on the Old Dominion. But for now, I still regard it as too-close-to-call. If the surge Romney has enjoyed in the past couple of weeks continues, he will probably snatch its thirteen electoral-college votes. But if Obama stages a rally, even a modest one, he could eke out a victory.
For the sake of argument, let’s assume Romney carries Florida and Virginia. According to my map, that would leave him with 248 votes, still twenty-two short of victory. If he also wins Colorado and New Hampshire, both of which I have listed as toss-ups, he would get to 261 votes, leaving him needing nine more to reach 270. That means he would have to win Ohio or Wisconsin, or Iowa and Nevada. The good news for Democrats is that Obama appears to be holding onto his leads in all four of these states. That’s his firewall.
Ohio: There have been four polls this week, and all of them have showed Obama ahead. In a poll of likely voters, P.P.P., which tends to lean to the Democrats, showed him leading Romney 51-46, a five-point margin. Another poll, from SurveyUSA, put the President’s lead at three points. Both of these surveys were carried out before Tuesday’s debate. Another post-debate poll was from Rasmussen. It showed Obama’s lead over Romney at just one point—49-48—but that is the same margin that Rasmussen has shown all month in Ohio. On Friday, Fox News released the newest poll from the Buckeye State, and it showed Obama still leading by three points, 46-43. Taken together, the polls suggest that all the money and manpower that Romney has thrown into Ohio haven’t yet made much of a difference.
Wisconsin: There have been two polls this week, and both of them have had the President on top. Rasmussen, in a poll carried out on Thursday, showed him leading Romney 50-48—a narrow margin. But a poll with a bigger sample from NBC News/Wall Street Journal/Marist College, showed Obama leading by six points: fifty-one per cent to forty-five per cent. That’s virtually unchanged from last month, when the same poll found Romney trailing by five points. The key to Obama’s big lead: he continues to win the support of far more women than Romney. In the latest poll, the gender gap is eighteen per cent.
Iowa: The picture here is a bit more mixed, but still broadly favorable to Obama. Of three polls conducted this week, the President has been leading in two of them. WeAskAmerica, which interviewed residents on Monday, before the debate, showed Obama with forty-nine per cent of the vote and Romney with forty-seven per cent. NBCNews/Wall Street Journal/Marist, which asked questions before and after the debate, had Obama even further ahead of his Republican rival: fifty-one per cent to forty-three per cent, a margin of eight points. As in Wisconsin, this poll suggests that very little has changed over the past month. In September, it also credited Obama with a lead of eight points.
The one dark spot for the White House: a P.P.P. poll released on Friday that showed Romney leading in Iowa by one point. That represents a big turnaround. In a poll taken last month, P.P.P. had Obama leading by seven points. Now he is trailing, 48-49—a statistical tie. And Romney’s favorability ratings are greatly improved. In September, forty per cent of Iowans viewed him favorably, and fifty-five per cent viewed him unfavorably. In the new poll, his favorability rating is forty-eight per cent, and his unfavorability rating is forty-seven per cent—evidently a reflection of his performance in the first two debates.
Nevada: Not much change here from Monday, when I pointed out that Romney, despite making up some ground, had failed to come out ahead in a single poll. Two polls were carried out early in the week, and both showed Obama leading by three points. In yet another Rasmussen survey, he was at fifty per cent and Romney was at forty-seven per cent. A poll that SurveyUSA carried out for the Las Vegas Review Journal showed the President with forty-eight per cent and the challenger with forty-five per cent.
That’s it for now. If things keep going like this. we’ll have to start considering the possibility of Obama doing what George W. Bush did in 2000: narrowly losing the popular vote but winning the election. That, though, is a subject for another post.
Read more http://www.newyorker.com/online/blogs/johncassidy/2012/10/cassidys-count-obamas-electoral-college-firewall-is-holding--just.html?printable=true¤tPage=all#ixzz29xASWfWL
Comments
Post a Comment
Thanks for leaving comments. You are making this discussion richer and more beneficial to everyone. Do not hold back.