New Poll shows 15 point lead for Obama

[Undesired Walrus]

http://bloomberg.com/apps/news?pid=20601087&sid=agCTbSDJ83rc&refer=home

Appears to be beating McCain 48 percent to 33 percent, with the independents pulling in 7%.

This said, I remember a 20 point lead for Kerry at one time.

 

[Pookster]

Here the history of some other elections with one candidate having an early big lead ...

http://politicalticker.blogs.cnn.com/2008/06/25/poll-shows-obama-with-a-double-digit-lead/

“Historically speaking, when June polls show a tight race, the race usually remains tight all the way through November. But when June polls have shown a big lead for one candidate, that lead has often melted," Holland said.

"Bill Clinton was leading Bob Dole by up to 19 points in June, 1996; Clinton won by eight. Michael Dukakis had a 14-point lead over George Bush the elder in June, 1988; Bush won by seven. Jimmy Carter was up nearly 20 points in June, 1976 but in November eked out a two-point win. And Richard Nixon managed an even smaller victory in 1968 even though he had a 16-point margin that June," Holland noted.

I typically don't pay too much credence to polls before the convention. They can change so dramatically over the coming months. But, I do think they're accurate for a point in time. What may make this year different is how close people are paying to the election compared to others in the past. It's hard to gauge that right now though. Also, there are wildcards in this election that haven't been there in the past -- a woman and a Black man as serious/major contenders.

 

[Denver]

It just seems to me that a democratic sweep over the republican candidates is just such a slam-dunk at this point, that all the polls, newspinions, conventions, and hundreds of millions of dollars in advertising and campaigning is just a huge waste. Am I just caught up in some version popular opinion, or does this seem likely to others as well?

 

[Pookster]

It just seems to me that a democratic sweep over the republican candidates is just such a slam-dunk at this point, that all the polls, newspinions, conventions, and hundreds of millions of dollars in advertising and campaigning is just a huge waste. Am I just caught up in some version popular opinion, or does this seem likely to others as well?

It's too easy to get caught up in the moment. For example, I'm a big softball fan, and like baseball almost as much. I was watching the CWS last night with my husband and Georgia was up 5-0 early in the game, and Fresno St was using a patchwork of pitchers because they had used up their best over the weekend. The Georgia players looked very confident about wrapping up the series with a huge win. It looked like it would be an easy blowout for them. However, they ended up losing 19-10. As Yogi once said, it ain't over till it's over.

 

[dudalb]

It will tighten up as the election gets closer.
There are also signs the Media infatuation with Obama is finally wearing out. You are beginning to see some less then laudatory reporting on Obama and his campaign.
But the big danger with stories like this is Democratic overconfidence. If they think they will have a cakewalk to the White House that is a huge opening for the GOP.

 

[dudalb]

And then of course there is always this little lesson on thinking you have it in the bag because of polls:

 

[Brainster]

I typically don't pay too much credence to polls before the convention. They can change so dramatically over the coming months. But, I do think they're accurate for a point in time. What may make this year different is how close people are paying to the election compared to others in the past. It's hard to gauge that right now though. Also, there are wildcards in this election that haven't been there in the past -- a woman and a Black man as serious/major contenders.

Well, the woman's gone, right? And there really is no metric for telling whether people are paying more attention this year than in elections past.

Gallup shows the race even at 45-45. The Bloomberg poll contains one major red flag:

...according to the poll of 1,115 registered voters...

Registered voters? Those polls always skew for the Democrats, because those gosh darn Republicans are more reliable about voting. Likely voters are what most pollsters survey, unless the intent is to get a headline.

 

[BenBurch]

There is only one poll that matters. And I assure you we WILL win this one. And we WILL come out of it with 2/3 of the House, and enough of the Senate to have our way with it. (Remember the nuclear option on filibuster? Expect to see it exercised.)

 

[corplinx]

The RCP index has Obama by 7. If someone has Obama by 15, the news is probably that their poll methodology might need looking at.

 

[Pookster]

Well, the woman's gone, right? And there really is no metric for telling whether people are paying more attention this year than in elections past.

Yep, she's no longer running for President. But it has been a wildcard in this election that has brought a lot of attention that isn't normally seen in primary seasons in the past. As I alluded to, it is hard to gauge how close people are paying attention, especially compared to the past. But I'd bet this season ranks pretty high (at least for the Democrats) compared to previous ones because of the players involved. Now we have the first Black major party candidate. I think it's a safe bet that attention will be a bit higher this year than most any in the past.

Gallup shows the race even at 45-45. The Bloomberg poll contains one major red flag:

Registered voters? Those polls always skew for the Democrats, because those gosh darn Republicans are more reliable about voting. Likely voters are what most pollsters survey, unless the intent is to get a headline.

The polls are all over the place right now. Groups use varying methodologies to do their polls. Many attempt to predict turnout within the various parties as part of that methodology. Many have their own definition of what is a "likely voter". Many pollsters report both registered and likely voters. They don't tend to vary much between them though, but they can. Which methodology/defintion is more accurate? Time will tell. Again though, I put little credence in polls right now. It's like predicting the final score of a softball game in the first inning. Anything can happen.

The one "poll" I put any real faith in at this time is the Iowa Electronics Market (and similar predictions/futures markets). But it's still even a bit early for it as I expect many more traders to participate in the coming months which could change things dramatically. It has an amazing track record though.

 

[Pookster]

The RCP index has Obama by 7. If someone has Obama by 15, the news is probably that their poll methodology might need looking at.

RCP averages polls that use different methodologies. It doesn't have any real value other than as a gee whiz thing, IMO. I haven't looked, but I bet the RCP index is including one or more of those "Obama by 15" polls. If you question those polls methodologies, then you can't put any faith in the RCP index.

 

[Puppycow]

RCP averages polls that use different methodologies. It doesn't have any real value other than as a gee whiz thing, IMO. I haven't looked, but I bet the RCP index is including one or more of those "Obama by 15" polls. If you question those polls methodologies, then you can't put any faith in the RCP index.

The RCP actually does exclude polls with dubious or unknown methodologies from being a component in that average (although it may publish them with an asterisk).
Of those "scientific polls" that use reasonable methodologies to get a truly representative sample, Since we can't be sure which methodology is best, it seems to me that an average of multiple polls is better than just one as the total sample gets larger.

 

[Puppycow]

The one "poll" I put any real faith in at this time is the Iowa Electronics Market (and similar predictions/futures markets). But it's still even a bit early for it as I expect many more traders to participate in the coming months which could change things dramatically. It has an amazing track record though.

Meh. It seems to track conventional wisdom. The prediction markets totally didn't predict that Clinton would win New Hampshire (neither did the polls, or, apparantly Clinton Campaign insiders). Just goes to show there's no ct crystal ball.

 

[BenBurch]

Puppycow, yes, a pool of polls usually works best. I look forward to Professor Pollkatz starting his election tracking soon as he does a good job with those.

 

[Pookster]

The RCP actually does exclude polls with dubious or unknown methodologies from being a component in that average (although it may publish them with an asterisk).
Of those "scientific polls" that use reasonable methodologies to get a truly representative sample, Since we can't be sure which methodology is best, it seems to me that an average of multiple polls is better than just one as the total sample gets larger.

I don't doubt they exclude dubious polls, but mixing polls with different yet reasonable methodologies gives you something of little to no value. It's likely less accurate than just looking at any one poll. Polls are greatly influenced by the specific questions asked. Many ask the same questions in slightly different ways. This can obviously lead to very different results, but it doesn't mean the poll isn't valid for it's specific purpose, intent, and the questions asked.

 

[Pookster]

Meh. It seems to track conventional wisdom. The prediction markets totally didn't predict that Clinton would win New Hampshire (neither did the polls, or, apparantly Clinton Campaign insiders). Just goes to show there's no ct crystal ball.

The Iowa Electronics Market has been pretty much spot on with every Presidential election it has done ... so far. That's when you also have a higher interest and volume of traders. It'll be interesting to follow again this year and see how accurate it is.

 

[Brainster]

The Iowa Electronics Market has been pretty much spot on with every Presidential election it has done ... so far.

Much as I enjoy the IEM and InTrade, I have to disagree here. IEM had Bush winning the day before the election in both 2000 and 2004, but there were wild gyrations on both election days. Gore actually ended up winning the IEM in 2000 (because their metric was the popular vote), and Kerry was up at least as high as 80 cents on election day 2004 before crashing and burning.

I agree with Puppycow; the IEM and InTrade do a great job of synthesizing the conventional wisdom.

 

[Pookster]

Much as I enjoy the IEM and InTrade, I have to disagree here. IEM had Bush winning the day before the election in both 2000 and 2004, but there were wild gyrations on both election days. Gore actually ended up winning the IEM in 2000 (because their metric was the popular vote), and Kerry was up at least as high as 80 cents on election day 2004 before crashing and burning.

I agree with Puppycow; the IEM and InTrade do a great job of synthesizing the conventional wisdom.

But it was right in the end. But, I concede that it's not very reliable as a predictor leading up to a very close election.

 

[dudalb]

There is only one poll that matters. And I assure you we WILL win this one. And we WILL come out of it with 2/3 of the House, and enough of the Senate to have our way with it. (Remember the nuclear option on filibuster? Expect to see it exercised.)

So you want basically want the Democratic Party of have a defacto dictatorship of the US?
Screw that.
And I promise you this: If the Democrats do get total control of the congress, they will behave in such a way that 2010 will become a repeat of 1994. Guaranteed.

 

[Pookster]

And I promise you this: If the Democrats do get total control of the congress, they will behave in such a way that 2010 will become a repeat of 1994. Guaranteed.

Democrats already do have control of Congress since 2006. Does that mean 2008 will be a repeat of 1994? Or is two years too soon?