Obama '08?

realjohnboy wrote:

I know I am pissing into the wind but, as a life-long red-neck, I get very weary of yankees and westerners looking down their holier then thou noses and finding it so easy to use "race" and "south" in the same sentence all the time. Look in your own neighborhoods, damn it.

Soz and Nimh have been discussing the recent polls, which of course are national polls. Not state specific, where the electoral votes are.

Virginia went for Bush over Kerry 54% to 46% in 2004. But in 2008 we stand a good chance of going blue (not done in my life time except by LBJ).

Obama, if nominated, has a good chance in VA, NC, SC, GA and others, particularly if the repubs choose Rudi or Romney. He would have a tougher fight against Huckabee and an impossible chance against McCain.

I don't see Hillary as having much traction in the South if she were the nominee.

If the South is deemed critical to the Dem's chances, they should look there for a VP. Edwards probably wouldn't want it so it would be someone lesser known like VA Gov Kaine, who is one smart dude.

The best choice of a running mate for Obama would be Bill Richardson of AZ. That is the dream ticket for me.

How is that for a mid-night ramble.

I agree on the racism issue. Being a Southerner, living in the East for over 37 years, racism here is overt and subtle, but it's here and these yanks will argue you down that they are not racist, yet you see no minorities in any high management positions or leadership, no matter how qualified. Housing discrimination, etc., you name it, theres elements of it everywhere! Until the conventions, I'm leaving my options open. NJ is a so-called "blue" state, but it isn't liberal, trust me. 2 Cents

0 Replies

nimh wrote:

Surely the Obama campaign has someone who is able to track the polls and calculate the averages, so their candidate doesnt go out making some false claim on the basis of cherrypicked polls? And if it does, why did he do exactly that anyway?

Why shouldn't a candidate be allowed to cherry-pick among the polls? Don't polling firms invite that sort of cherry-picking? After all, each polling firm represents its product as being accurate (within a certain statistical degree of error), and so there should be no need to average those results with the results of other polls.

Averaging is something that statisticians outside of the polling industry insist upon, not the pollsters themselves. Nobody at Zogby or Gallup is saying "our poll results should only be read in conjunction with the results of everyone else's polls." And if the polling firms are representing the results of their polls as accurate, why can't politicians do the same?

0 Replies

JPB wrote:

Admitted drug use is being touted as a plus in the media I've been watching - demonstrates honesty and willingness to be upfront and open. It will come up in the general election but doesn't seem to be a big deal.

I wonder what drugs the other candidates would admit to using in their lives if they had to list out all the drugs they've ever used over their lifetimes and not just the illegal ones.

Also, would/should a politician's past reliance on legal drugs such as sleeping pills or diet pills or anti-depressants make the same difference to anyone's vote? Should the public be informed of these things or is that considered too private?

0 Replies

I don't think voters care about drug use.

Nobody believed Clinton when he said he never inhaled, but we elected him anyway, understanding why he couldn't admit to it when the War on Drugs was still going strong. Everyone knows Bush was a coke head (among other much worse things) in his youth but let him off the hook even though he refused to admit it while at the same time asserting that his past was behind him. I don't think voters would think twice about a politician who admits to having used drugs in their youth. In fact, I think it might go to their credibility. It's a non-issue, and has the extra bonus of being likely to backfire on anyone wishing to make it one.

0 Replies

Ah, I missed Joe's snark a few pages back..

joefromchicago wrote:

nimh wrote:
The site identifies LaFollette as "the populist governor, U.S. Senator, and presidential candidate from Wisconsin who founded the Progressive Party and spent his career battling the corrupting, impoverishing and anti-democratic influence of big moneyed interests over government and public policy".

Well then, it's official.

Well, that plus the other citations you conveniently skip -- you know, the biographer of LaFollette and stuff...

0 Replies

realjohnboy wrote:

The best choice of a running mate for Obama would be Bill Richardson of AZ. That is the dream ticket for me.

Close, but no cigar. Richardson is from (ahem) New Mexico.

0 Replies

More from Joe:

joefromchicago wrote:

Why shouldn't a candidate be allowed to cherry-pick among the polls?

Because it's misleading.

I mean, sure, I know they do, and they will; but it is misleading, so luckily people will in turn point out they're not being honest.

joefromchicago wrote:

Don't polling firms invite that sort of cherry-picking? After all, each polling firm represents its product as being accurate (within a certain statistical degree of error), and so there should be no need to average those results with the results of other polls.

How is that relevant? When the presidential candidate conveniently cherry picks a poll to his liking and presents it as a truthful representation of reality, it's OK as long as he is parrotting the PR of the polling firm in question when doing so? Dont see the logic here.

Meanwhile, not to beat the semantics out of this dead horse, but in the case at hand your question is actually irrelevant. Obama didnt say "I have a better chance at beating the Republicans," basing that claim on one poll in accordance with the poll's claims of representing reality, as in your scenario. He said he was doing the best of all candidates in the polls - and since he was only doing better than Edwards in one poll, with another poll in the same week showing the opposite and other recent polls split between them, that just wasn't true.

joefromchicago wrote:

And if the polling firms are representing the results of their polls as accurate, why can't politicians do the same?

Because it's not enough for a candidate for the Presidency of the US to live up to the standards of a commercial polling firm's PR. Better for him to live up to the standards of critical citizens - those who notice when they're cherrypicking polls, for example.

0 Replies

A sadly interesting tidbit from the Iowa campaign:

http://www.timesrepublican.com/page/content.detail/id/500947.html?nav=5005

0 Replies

From a blog by Dave "Mudcat" Saunders:

To begin with, I don't know who is going to win the Iowa Caucuses. The absolute truth is the pundits and others can guess all they want, but nobody really knows. There are simply too many pieces to the puzzle, and it's all those pieces that make the Iowa Caucuses the most intriguing electoral process in American politics.

My intention, and I want to be clear that I don't have permission, is to speak out for all the Democratic presidential campaigns by saying that a win (or a strong showing in some of the candidates' cases) in Iowa is of critical importance. That said, I am constitutionally incapable of sitting back in good conscience and quietly listening to Hillary Clinton's words to the people in Iowa and then her 180-degree spin to the national press.

Hillary Clinton has a serious political problem. From the get-go her campaign pushed her to the press and her contributors as the "inevitable" candidate. Her problem is if you bill yourself as the "inevitable" nominee in a presidential nominating race, you had better be right because there is no fall-back position.

Now that her "inevitability" is under serious scrutiny, she is running around Iowa telling Iowans that she loves them, they are important, and she has to have them, while at the same time, she is spinning to the national press that Iowa is not so important that she has to win there to secure the nomination.

The absolute truth is with a campaign team that looks like the 1927 New York Yankees and with all the resources she has thrown at Iowa, a bad showing from Hillary can only reinforce what so many of us have been saying all along. If she can't win Iowa, she has absolutely no chance to win the general. It's that simple.

Next Thursday night when she doesn't win the Iowa Caucuses, she will immediately say that Iowa is not all that important, which in my mind is not only denigrating all those Iowa Caucus goers who have taken their time to take a great look at all of the candidates, but it also gives all Americans a great look at what's to come.

After next Thursday night, Hillary Clinton will resort to the only conceivable option she has left. With the huge coffers she has accumulated from raising funds from the many who have a financial stake in the next administration, she's going to try to flood the media markets for February 5 and buy the damned thing. It's that simple

To paraphrase Mr. Lincoln, "You can fool some of the people all the time, but in the case of the Clinton campaign, everybody can see this trick coming." My greatest hope is the blogosphere and the national media won't let them get away with it.

0 Replies

nimh wrote:

More from Joe:

joefromchicago wrote:
Why shouldn't a candidate be allowed to cherry-pick among the polls?

Because it's misleading.

Citing the results of a poll is misleading? Really? Even when polling firms do it?

nimh wrote:

joefromchicago wrote:
Don't polling firms invite that sort of cherry-picking? After all, each polling firm represents its product as being accurate (within a certain statistical degree of error), and so there should be no need to average those results with the results of other polls.

How is that relevant? When the presidential candidate conveniently cherry picks a poll to his liking and presents it as a truthful representation of reality, it's OK as long as he is parrotting the PR of the polling firm in question when doing so? Dont see the logic here.

The logic is pretty simple. Polling firms are in the business of taking polls. They go to great lengths to insure the fairness of the questions and the accuracy of the results. Clearly, the polling firms think the results that they get mean something. And just as clearly, so do you, otherwise you wouldn't be citing those poll results in post after post. So why should politicians be any different? Why should they be required to average poll results (even though the use of differing methodologies among the polls means that averaging is, at best, only a faute de mieux solution to perceived inadequacies in the poll results) when no one else -- except the statistical geeks outside of the polling industry and the campaigns -- does it?

Now, of course, if the results of any single poll are meaningless, then I can see your point. But you've never taken that position, so I still can't see why Obama should take it.

nimh wrote:

Meanwhile, not to beat the semantics out of this dead horse, but in the case at hand your question is actually irrelevant. Obama didnt say "I have a better chance at beating the Republicans," basing that claim on one poll in accordance with the poll's claims of representing reality, as in your scenario. He said he was doing the best of all candidates in the polls - and since he was only doing better than Edwards in one poll, with another poll in the same week showing the opposite and other recent polls split between them, that just wasn't true.

I'm not sure what Obama said. I imagine that he said all sorts of things on the campaign trail. But then I wasn't responding to Obama, I was responding to you. You said that he was cherry-picking a single poll result. That's what I'm responding to.

nimh wrote:

joefromchicago wrote:
And if the polling firms are representing the results of their polls as accurate, why can't politicians do the same?

Because it's not enough for a candidate for the Presidency of the US to live up to the standards of a commercial polling firm's PR. Better for him to live up to the standards of critical citizens - those who notice when they're cherrypicking polls, for example.

PR? You mean that polls aren't reliable or meaningful? That it's all just a matter of public relations? Then what is the use of polling?

0 Replies

joefromchicago wrote:

nimh wrote:
More from Joe:

joefromchicago wrote:
Why shouldn't a candidate be allowed to cherry-pick among the polls?

Because it's misleading.

Citing the results of a poll is misleading? Really? Even when polling firms do it?

That's different. Pollsters decide which poll to cite before the results are in: it's their own. This is not as accurate as averaging over a number of polls. But there is no cherry-picking involved, so at least the pollsters report no information that is systematically skewed.

Campaigns, by contrast, look at a number of polls after the results are in, then cherry-pick the one whose results they like best. This is not as bad as outright lying, and nimh didn't say that it is. But it's misleading because the poll that any one candidate likes best will report results somewhere in the tail of the statistical distribution of polling results.

0 Replies

Thomas wrote:

joefromchicago wrote:
Citing the results of a poll is misleading? Really? Even when polling firms do it?

That's different. Pollsters decide which poll to cite before the results are in: it's their own. This is not as accurate as averaging over a number of polls. But there is no cherry-picking involved, so at least the pollsters report no information that is systematically skewed.

Averaging over a number of polls is not necessarily more accurate than taking the results of a single poll. After all, in order for the average to be meaningful, the different polls must all be alike. But if the various polls use different methodologies, or if they ask different questions, then averaging the results is simply averaging apples and oranges. The resulting average is little better than a guess.

Thomas wrote:

Campaigns, by contrast, look at a number of polls after the results are in, then cherry-pick the one whose results they like best. This is not as bad as outright lying, and nimh didn't say that it is. But it's misleading because the poll that any one candidate likes best will report results somewhere in the tail of the statistical distribution of polling results.

So what? If a poll represents itself as an accurate reflection of reality, then why isn't a candidate entitled to accept it as such? Just because there's another poll out there that comes to a different result? I'm sure you don't apply that reasoning in other areas. For instance, if two economists come to different conclusions about supply-side economics, am I supposed to average their conclusions, or can I choose one of them based on my judgment of which is the more accurate?

0 Replies

Looks like Obama overtook Hillary in New Hampshire.

0 Replies

hmmm... I'm not so sure that averaging polls is more accurate than using a pre-selected poll, Thomas.

I haven't spent any time looking into those who average polls but, unless they are taking multiple errors into account, the errors associated with the averages are much greater than those of individual polls.

Usually, a poll result is expressed in terms of a margin of error at 95% confidence (say, 3 percent). If you average 5 polls, each with a MOE of 3% at 95% confidence, the MOE on the average is more like 15%. The confidence of the average would also be affected. Each MOE has a 5% chance of being wrong (confidence is almost never reported with poll results). If you take the result of numerous polls then there is a one-in-twenty chance that the number you are using is inaccurate beyond the MOE. This inaccuracy isn't exactly cumulative to the average because the effect of the wrong result will be diluted as part of the averaging, but it still affects the overall outcome.

0 Replies

What JPB said. There are two reasons why I agree with JPB; 1) she's a statistitian, and 2) I've noticed the polls done by FOX usually favors the republican candidates more than the others whether it's about Performance or Electability. Polls would see more accurate by excluding FOX from the mix.

0 Replies

joefromchicago wrote:

Thomas wrote:
joefromchicago wrote:
Citing the results of a poll is misleading? Really? Even when polling firms do it?

That's different. Pollsters decide which poll to cite before the results are in: it's their own. This is not as accurate as averaging over a number of polls. But there is no cherry-picking involved, so at least the pollsters report no information that is systematically skewed.

Averaging over a number of polls is not necessarily more accurate than taking the results of a single poll. After all, in order for the average to be meaningful, the different polls must all be alike. But if the various polls use different methodologies, or if they ask different questions, then averaging the results is simply averaging apples and oranges. The resulting average is little better than a guess.

Using the moving average of the same poll diffuses these affects somewhat. If CNN takes a weekly poll by the same procedure (sampling, questions, etc.,) and calculates a moving average of their results then they are more accurate than taking polls from numerous entities and averaging them together. Error bands would still need to be adjusted, but a repeated measures over time is less problematic than multiple measures under different protocols.

0 Replies

PollingReport.com

PRESIDENT BUSH - Overall Job Rating in recent national polls

--------------------------------------------------------------------------------

,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,, Dates.... %.... %.... %.. Disapprove
.

FOX/Opinion Dynamics RV 12/18-19/07 36 57 6 -21
.

NBC/Wall Street Journal
12/14-17/07 34 60 6 -26
.

USA Today/Gallup 12/14-16/07 32 65 3 -33
.

Diageo/Hotline RV 12/10-14/07 33 62 4 -29
.

ABC/Washington Post 12/6-9/07 33 64 3 -31
.

CNN/Opinion Research 12/6-9/07 32 66 2 -34
.

Gallup 12/6-9/07 37 57 5 -20

0 Replies

JPB wrote:

Using the moving average of the same poll diffuses these affects somewhat. If CNN takes a weekly poll by the same procedure (sampling, questions, etc.,) and calculates a moving average of their results then they are more accurate than taking polls from numerous entities and averaging them together. Error bands would still need to be adjusted, but a repeated measures over time is less problematic than multiple measures under different protocols.

Yes, I agree. But then I doubt that nimh or Thomas were talking about moving averages for the same poll. They were talking about averaging different polls taken at approximately the same time.

0 Replies

joefromchicago wrote:

nimh wrote:
More from Joe:

joefromchicago wrote:
Why shouldn't a candidate be allowed to cherry-pick among the polls?

Because it's misleading.

Citing the results of a poll is misleading? Really? Even when polling firms do it?

I'm going to bed now, but on a quickie note, I dont really understand the confusion here.

If one poll substantiates your assertion, and another concurrent poll refutes it, than to say that "the polls substantiate my assertion" is misleading.

Same as with any cherry picking of data.

Agreed?

0 Replies

OK - one more.

joefromchicago wrote:

Now, of course, if the results of any single poll are meaningless, then I can see your point. But you've never taken that position, so I still can't see why Obama should take it.

Either you must have read very few posts of mine about polls, or you are of short memory. This is after all the exact point I've been hammering on about ad nauseam on this forum. Yes, it is foolish to go on any one single poll - as I have repeated over and over on thread after thread, as others will attest with memories of headache...

Why do you think, in your thread about how Hillary and Giuliani would be the worst choices for their respective parties, I brought a set of graphs and stats that reflected, not any one single poll, not any two or three single polls, but a systematic tracking of polls over several months? Here, re-read the post, as it might further illustrate my problem with the kind of statement that, in this case, Obama made. Notice how individual polls featured results that varied enormously, veering out one way or another, and how I therefore only based my interpretation on rolling averages and other such ways to distill general trends.

Polls are extremely informative things, but their nature means that yes, any one single poll is, if not quite meaningless, an irresponsible basis for any assertion that reaches beyond the most blatant fundamentals (if a mainstream poll shows you ahead 60 to 30, you can safely say you're looking pretty good, but that's about it). There's ways to validly use polls, and ways to abuse them; cherrypicking the one that fits your agenda from among dissenting ones is the latter.

I invite anyone who's interested in this stuff, by the way, to my Polls, numbers and pretty graphs thread.. there's already a lot of relevant information and reflections there, not just about who's ahead and who's down, but also about the nature and pitfalls of polling itself.

0 Replies

Related Topics

Quick Links

My Account

able2know