We’re not quite at the point where “who would you vote for if the election were held today?” poll numbers are predictive, though we’re getting close. For most of the year, my mantra has been that these polls have ZERO predictive power about the November election whatsoever, and that if you’re trying to interpret them as a way to tell who is more likely to win, you’re best off completely ignoring them. At this point I’d temper that a little bit, but only a little bit. I think these “who would you vote for if the election were held today?” numbers really start becoming somewhat predictive about two weeks before election day.
<
p>Of course, there are plenty of other reasons to believe Obama is more likely to win than McCain, but that has been true all along, even when McCain was supposedly “ahead”, and when they were tied.
marcus-gralysays
The election is being held today and will continue to be held for the next month.
Probably people sure and comfortable with the choices they make, not recent converts to Obama. I imagine in early voting, Obama is sweeping die-hard Democrats.
… and a voter in a precinct that had lines around the block in 2004 to the point that judges had to get involved, I’d have all the motivation necessary for early voting that a long time support of Obama has in the same precinct.
<
p>I’ll grant that, outside that situation, you’re probably right.
… but only a very small percentage of the total are voting “today”, so what I said above holds true. These polls are not yet predictive, for the most part. (which is different from, say, a month ago, when they weren’t predictive to even the slightest extent)
chrissmasonsays
As of 10:50 today Pollster.com now has Ohio blue with Obama capturing 320 Electoral Votes. Throw Virginia and Nevada in there and Obama takes it with 338. If he holds onto this lead, one might call it a mandate. I’d call it a mandate for change.
jim-gosgersays
For the most scientific treatment of polls and projections of Electoral Votes try fivethirtyeight.com:. Nate Silver, one of the statistical geniuses behind Baseball Prospectus, has applied his mathematical mind to the national and statewide polls. His numbers show Obama with a big lead and getting bigger. He even has an article on how Georgia is in play for Obama give the large registrations of black voters (not reflected in the polls). I expect the the race to get closer as we get closer to election day, but this will be a very big win for Obama/Biden. BTW, Nate also is projecting the Senate races. At this point 60 Senate seats is a reach, but 55 or 56 is likely.
jconwaysays
Id be cautious. Polls are showing an Obama landslide which would be a true mandate at this point, but three weeks ago they showed a McCain-Palin victory. I would say it is far more likely that Obama wins a smaller victory. Kerry states plus CO, NM, and at this point OH.
<
p>Nate Silver is awesome, a fellow UofCer, and he is coming on my college radio show so I like him for personal reasons. But he also has the best predictive model for electoral votes. He dismisses partisan polls, dismisses polls that have been proven inaccurate, and he runs over 1000 simulations for each state daily. A solid site.
<
p>As for the Senate I personally am angry at the DSCC for wasting time and money on Tom Allen in Maine (a lost cause, he’s been down by 15pts since the convention) while not spending enough to put Kay Hagan, Ronnie Musgrove, and Jim Martin over the top in NC, MS, and GA. If just two out of three of those states turns over and all the other races fall the way they ought too, then we can get 60 seats. Schumer should ditch Allen and spend in the south.
We’re not quite at the point where “who would you vote for if the election were held today?” poll numbers are predictive, though we’re getting close. For most of the year, my mantra has been that these polls have ZERO predictive power about the November election whatsoever, and that if you’re trying to interpret them as a way to tell who is more likely to win, you’re best off completely ignoring them. At this point I’d temper that a little bit, but only a little bit. I think these “who would you vote for if the election were held today?” numbers really start becoming somewhat predictive about two weeks before election day.
<
p>Of course, there are plenty of other reasons to believe Obama is more likely to win than McCain, but that has been true all along, even when McCain was supposedly “ahead”, and when they were tied.
The election is being held today and will continue to be held for the next month.
Probably people sure and comfortable with the choices they make, not recent converts to Obama. I imagine in early voting, Obama is sweeping die-hard Democrats.
… and a voter in a precinct that had lines around the block in 2004 to the point that judges had to get involved, I’d have all the motivation necessary for early voting that a long time support of Obama has in the same precinct.
<
p>I’ll grant that, outside that situation, you’re probably right.
… but only a very small percentage of the total are voting “today”, so what I said above holds true. These polls are not yet predictive, for the most part. (which is different from, say, a month ago, when they weren’t predictive to even the slightest extent)
As of 10:50 today Pollster.com now has Ohio blue with Obama capturing 320 Electoral Votes. Throw Virginia and Nevada in there and Obama takes it with 338. If he holds onto this lead, one might call it a mandate. I’d call it a mandate for change.
For the most scientific treatment of polls and projections of Electoral Votes try fivethirtyeight.com:. Nate Silver, one of the statistical geniuses behind Baseball Prospectus, has applied his mathematical mind to the national and statewide polls. His numbers show Obama with a big lead and getting bigger. He even has an article on how Georgia is in play for Obama give the large registrations of black voters (not reflected in the polls). I expect the the race to get closer as we get closer to election day, but this will be a very big win for Obama/Biden. BTW, Nate also is projecting the Senate races. At this point 60 Senate seats is a reach, but 55 or 56 is likely.
Id be cautious. Polls are showing an Obama landslide which would be a true mandate at this point, but three weeks ago they showed a McCain-Palin victory. I would say it is far more likely that Obama wins a smaller victory. Kerry states plus CO, NM, and at this point OH.
<
p>Nate Silver is awesome, a fellow UofCer, and he is coming on my college radio show so I like him for personal reasons. But he also has the best predictive model for electoral votes. He dismisses partisan polls, dismisses polls that have been proven inaccurate, and he runs over 1000 simulations for each state daily. A solid site.
<
p>As for the Senate I personally am angry at the DSCC for wasting time and money on Tom Allen in Maine (a lost cause, he’s been down by 15pts since the convention) while not spending enough to put Kay Hagan, Ronnie Musgrove, and Jim Martin over the top in NC, MS, and GA. If just two out of three of those states turns over and all the other races fall the way they ought too, then we can get 60 seats. Schumer should ditch Allen and spend in the south.
last night. Pretty cool stuff.