15 States to Watch in the US Presidential Election

[Backlum-Chaam]

Re: 15 States to Watch in the US Presidential Elec

"As CO goes, so goes the nation"? Well, maybe not yet. But it sure is a close race in many states.

It really bothers me that the DNC state has already dipped into McCain's favor. The Republicans' VP snake oil has really shaken things up. Let's just hope middle America starts assessing Sarah more objectively as the dust settles before November...

 

[Sammie13]

Re: 15 States to Watch in the US Presidential Elec

One reason I think state polls are more pertinent than national polls is the race factor.

Obama's most loyal, highest turnout voting block obviously is African Americans. Where do the majority of blacks live? The South! Will Obama win those states? NO!!! So, his most inspired voting block will swing the national polls upward, but that voting block is populated in Republican states, which means they will be a waste to him electorally speaking. Obviously, this is a broad statement, and doesn't mean AA voters in places like Philadelphia, Detroit and Cincinnati can't help swing those states.

 

[CoolBlue71]

Re: 15 States to Watch in the US Presidential Elec

If the Democrats win Colorado, they'll do so thru greater Denver area (Denver and Arapahoe Counties). Jefferson County as well. Ouray County, too. And Larimer County, where Fort Collins is situated. (Larimer went with Bill Clinton in the 1990s, George W. Bush in the 2000s. But not margins too big.)

Pueblo plays a role as well.

If Colorado goes to the Republicans, routinely consevative Colorado Springs is instrumental. Same with eastern regions bordering with Kansas and Nebraska.

I believe this state will be one hell of a race. And we'll get a better idea in four to six weeks of the likely outcome.

 

[Sammie13]

Re: 15 States to Watch in the US Presidential Elec

I know Obama supporters are going to say 'it's just the convention and blahblahblah', but it should be said anyway that tomorrow's USA Today/Gallup Poll, which is 100% post conventions, has McCain now up by +10 points. Their previous poll on Aug 23 had Obama at +3, so a 13 point swing. The good news is that if you break it down to only registered voters, it is only a +4 for McCain. The other good news, if you're an Obama supporter, is that statistically post-convention polls remain essentially unchanged by the actual election some 50% of the time. So, there is a 50% chance that Obama can turn this ship around before it become the Titanic.

 

[MystikWizard]

Re: 15 States to Watch in the US Presidential Elec

I know Obama supporters are going to say 'it's just the convention and blahblahblah', but it should be said anyway that tomorrow's USA Today/Gallup Poll, which is 100% post conventions, has McCain now up by +10 points. Their previous poll on Aug 23 had Obama at +3, so a 13 point swing. The good news is that if you break it down to only registered voters, it is only a +4 for McCain. The other good news, if you're an Obama supporter, is that statistically post-convention polls remain essentially unchanged by the actual election some 50% of the time. So, there is a 50% chance that Obama can turn this ship around before it become the Titanic.

Sammie, perhaps you can share with us the breakdown of this particular national poll. How many people were polled for this one and if you could list how many people polled were from each state and who they voted for, that would be great, too. That way, we can get a better sense of what is going on in each state.

 

[Sammie13]

Re: 15 States to Watch in the US Presidential Elec

Sammie, perhaps you can share with us the breakdown of this particular national poll. How many people were polled for this one and if you could list how many people polled were from each state and who they voted for, that would be great, too. That way, we can get a better sense of what is going on in each state.

This is pretty much all I have:

In the new poll, taken Friday through Sunday, McCain leads Obama by 54%-44% among those seen as most likely to vote. The survey of 1,022 adults, including 959 registered voters, has a margin of error of +/— 3 points for both samples.

It's an astonishing bounce...I don't think it will hold up anywhere near those levels, but still the fact that Gramps could muster this kind of bounce is shocking to me. CNN's poll of polls tonight had Obama +1, but some of the polls included there were pre-GOP convention.

Bang it here for the rest of the story:
http://www.usatoday.com/news/politics/election2008/2008-09-07-poll_N.htm

 

[MystikWizard]

Re: 15 States to Watch in the US Presidential Elec

This is pretty much all I have:

I know. I was looking at it on Real Clear Politics, where it is posted, as well, and has a link to the article.

National polls like these need to have their breakdowns published.

Let's say we put together a hypothetical poll of 500 people, and we poll:

100 from Texas
50 from Florida
50 from Kansas
50 from Nevada
50 from Wyoming
50 from Arizona
50 from Utah
50 from New York
50 from California

See what I'm saying?

We assume from the surface that these polls are evenly distributed amongst the states, but we don't know that for sure, unless we have access to the breakdowns by state.

Certainly, if you were to do a poll with the sample I have above, you would unquestionably get more favorable results for John McCain than Barack Obama.

Now, I'm not saying it wasn't evenly distributed ... but this is important, because people live and die by these polls ... and polls have the power to change perception. The breakdowns need to be published, as well, to preserve the integrity of the poll.

Which is why most of the time, I keep harping on here that I don't listen to the National Polls and rather focus on polling from the individual states, instead.

 

[construct]

Re: 15 States to Watch in the US Presidential Elec

Which is why most of the time, I keep harping on here that I don't listen to the National Polls and rather focus on polling from the individual states, instead.

I have seen four state polls conducted entirely after the RNC--CNN/Time in Ohio, Minnesota, and Iowa (that show improvement for Obama) and Ivan Moore in Alaska (that shows improvement for McCain). The CNN/Time poll for Iowa showed a decline for McCain as well as an improvement for Obama (which is the reverse of the Alaska poll).

If the national movement favorable to McCain is concentrated in states that are moderately to solidly red already, it really means nothing.

 

[MystikWizard]

Re: 15 States to Watch in the US Presidential Elec

I have seen four state polls conducted entirely after the RNC--CNN/Time in Ohio, Minnesota, and Iowa (that show improvement for Obama) and Ivan Moore in Alaska (that shows improvement for McCain). The CNN/Time poll for Iowa showed a decline for McCain as well as an improvement for Obama (which is the reverse of the Alaska poll).

If the national movement favorable to McCain is concentrated in states that are moderately to solidly red already, it really means nothing.

That's precisely my point.

Let's face it. The closer the race- the more money/ratings it is for the Media.

I have a big problem with these so-called National Polls, because they do not publish the breakdowns of their results. Like you said, and what I was implying, if a larger concentration of a poll took place in Red States, it isn't going to mean a damn thing ... except get everyone all worked up, and to keep tuning in tomorrow for the next poll.

We, as the public, deserve to see the breakdowns of these polls, who's being polled, what age is being polled, are the polls evenly distributed amongst the country ... etc.

 

[construct]

Re: 15 States to Watch in the US Presidential Elec

That's precisely my point.

Let's face it. The closer the race- the more money/ratings it is for the Media.

I have a big problem with these so-called National Polls, because they do not publish the breakdowns of their results. Like you said, and what I was implying, if a larger concentration of a poll took place in Red States, it isn't going to mean a damn thing ... except get everyone all worked up, and to keep tuning in tomorrow for the next poll.

We, as the public, deserve to see the breakdowns of these polls, who's being polled, what age is being polled, are the polls evenly distributed amongst the country ... etc.

Well, no, I'm saying something different. I'm assuming that the poll subjects are proportionally divided among the states and correctly weighted for demographics. I'm saying that if a shift is in states already on that side, it affects popular results but not electoral college outcome.

I mentioned four state polls in my earlier post. The only one to get excited about is the Ohio poll.

 

[construct]

Re: 15 States to Watch in the US Presidential Elec

So then some results of the GOP convention would be that

1. McCain is declining in IA
2. Obama is improving in IA, which Kerry lost
3. Obama is improving in MN, where the GOP convention was held
4. Obama is improving in OH
5. McCain is improving in AK just after introducing its governor as a running mate

Sounds like very good news to Obama and bad news for McCain.

But if there is one year I don't easily believe the polls, it's this year.

And

6. Obama is declining in AK.

Now, IA, MN, and AK were already leaning significantly in the same direction that the shift occurred. Ohio was not. Net? It's somewhat good news for Obama.

Now if the RNC shifted opinion in, say, AK, SC, AZ, SD, and TX significantly toward McCain, the results would show a significant shift toward McCain nationally without affecting the probable electoral college result. If the shift was in NC, IN, MT, GA, and MO, it would be more troubling. If it were in NV, ND, OH, VA, and FL, it would mean potentially serious trouble.

 

[hotatlboi]

Re: 15 States to Watch in the US Presidential Elec

If Obama can take Ohio, I think he wins.

If not, it will depend on the smaller battlegrounds like NH, CO, and NV.

 

[Sammie13]

Re: 15 States to Watch in the US Presidential Elec

So then some results of the GOP convention would be that

1. McCain is declining in IA
2. Obama is improving in IA, which Kerry lost
3. Obama is improving in MN, where the GOP convention was held
4. Obama is improving in OH
5. McCain is improving in AK just after introducing its governor as a running mate

Sounds like very good news to Obama and bad news for McCain.

But if there is one year I don't easily believe the polls, it's this year.

No, no, guys...construct and B'back...the 3 CNN polls that your citing, IA/MN/OH, were all conducted Aug 31- Sept 2 well ahead of the Palin announcement or the GOP convention. Those are likely highwater marks, according to CNN, for Obama in those 3 states. In other words, toss them aside now. Only Alaska, to my knowledge, is post-Palin.

 

[Sammie13]

Re: 15 States to Watch in the US Presidential Elec

I know. I was looking at it on Real Clear Politics, where it is posted, as well, and has a link to the article.

National polls like these need to have their breakdowns published.


See what I'm saying?

We assume from the surface that these polls are evenly distributed amongst the states, but we don't know that for sure, unless we have access to the breakdowns by state.

Certainly, if you were to do a poll with the sample I have above, you would unquestionably get more favorable results for John McCain than Barack Obama.

I see what your saying, and it would be great to see state's incorporated into the poll. Nevertheless, you have to assume that there is statistical distribution. After all, that's what they do for a profession and credibility is key. If USA Today, CNN, Fox, etc are paying polling organizations to carry out a poll, you can be sure that the financial officers of those media outlets are checking up on the pollsters to see if they are deliverying accuracy for their money and credibility within the political world.

Further, while I'm not a national poll lover, let's be real here. Once a candidate begins to consistently deliver a 3% lead in polls, they are not going to lose the electoral vote. National polls are a preview of state polls. Let's assume that McCain consistently starts showing 10% leads, you already know the race is over...you're not going to overcome that kind of lead. State polls matter in races that are tight (up to approx. 3%) and after that become somewhat irrelevant.

 

[Sammie13]

Re: 15 States to Watch in the US Presidential Elec

One new poll this morning:

MICHIGAN -- Obama +1 -- Public Policy Polling -- The Wolverine State previously had Obama at +3 in this poll, so not alot of damage from the GOP convention. Obviously, it's tightening up!

 

[Kulindahr]

Re: 15 States to Watch in the US Presidential Elec

If the national movement favorable to McCain is concentrated in states that are moderately to solidly red already, it really means nothing.

I've had that thought as well. But they don't take a large enough sample in a national poll to get a clue about what's going on in the various states. What they ought to do is a full all-state poll integrated to be national as well, so there could be a map showing each individual state and the overall. That would be a lot more useful than the piecemeal approach we're stuck with!

 

[Sammie13]

Re: 15 States to Watch in the US Presidential Elec

A new afternoon poll:

VIRGINIA -- McCain +2 -- SurveyUSA -- Old Dominion finds the MacMan slightly ahead.

 

[hotatlboi]

Re: 15 States to Watch in the US Presidential Elec

That 10% McCain lead poll has to be an anomaly. Washington Post's new poll post convention has Obama at +1 and McCain at +2 among likely voters.

http://www.washingtonpost.com/wp-dyn/content/story/2008/09/08/ST2008090801842.html

 

[MystikWizard]

Re: 15 States to Watch in the US Presidential Elec

A Colorado Rasmussen Poll conducted yesterday has Obama +3.

Colorado is all we need, as long as we don't lose Pennsylvania or Michigan.

http://www.realclearpolitics.com/epolls/2008/president/co/colorado_mccain_vs_obama-546.html

 

[Sammie13]

Re: 15 States to Watch in the US Presidential Elec

That 10% McCain lead poll has to be an anomaly. Washington Post's new poll post convention has Obama at +1 and McCain at +2 among likely voters.

http://www.washingtonpost.com/wp-dyn/content/story/2008/09/08/ST2008090801842.html

I agree. For now, it's an outlier, and should be tossed aside.