情趣用品,情趣用品,情趣用品,情趣用品,情趣用品,情趣用品,情趣用品,情趣用品,情趣用品,情趣用品,...

Anonymous Humanist,

Take a look at the Super Tracker from the peak in June to the present. It's only off by about 2 points in a month, which is about 1/2 the rate of change from the previous month. During this period, the search frequency was also down.

Now compare to the previous two months, from mid-March to mid-May. This is the flattest period of the Super Tracker, and it also corresponds to the period of lowest interest.

I should be clear that I don't expect that every time there is a spike in interest, there is a rapid change in voter preferences. Rather, what I believe is happening is that a spike in interest increases our uncertainty in people's positions. For example, we can expect that the higher the interest in the election over the last week, the worse a 1-week old poll will be at predicting today's opinions on average.

Anonomous Humanist what are you talking about. Tha...

2008-07-28T01:29:41.192-04:00

Anonomous Humanist what are you talking about. That was the biggest pile of crap I have ever seen in my life. Puma On and Off? The point of PUMA is they don't vote for Obama. How can they be on and off? I am stupider for reading your post.

What the hell are you talking about, you are talki...

2008-07-28T01:25:56.383-04:00

What the hell are you talking about, you are talking out of your ass.

re Karl Rove's "I have THE numbers":Of COURSE Rove...

2008-07-28T01:24:20.683-04:00

re Karl Rove's "I have THE numbers":

Of COURSE Rove lied about the Republicans' chances in 2006. Remember, as bad as that election was for them, it could have been worse; there were about a dozen seats (I do not remember the exact number) where the Republicans held on by a nose.

On the other hand, if Rove had said, "You know what, yeah, the GOP is screwed this cycle," yes, he would have been telling the truth, but that would have resulted in an even greater drop in GOP turnout, and a loss of several more seats.

As for Iraq, etc.: that was a policy decision. Rove was a political adviser, and you have to give the devil his due: he made Bush's absurdly boneheaded and (at times) downright unconstitutional policy decisions palatable to the American electorate for longer than anyone else could have.

Rove was by no means perfect; in 2006 he wasted a lot of RNC cash on long-shot races (e.g. Steele in Maryland, as well as several very Democratic House seats), resources that might have made the difference in VA and MO (thus retaining the Senate). But to call Rove a "screw-up" is, I think, a distortion of history.

Cugel, I agree with much of what you say, but 2004...

2008-07-28T01:23:10.748-04:00

Cugel, I agree with much of what you say, but 2004 was hardly the most divisive campaign in history. For more, I recommend the book "Anything for a Vote: Dirty Tricks, Cheap Shots, and October Surprises" by Joseph Cummins.

Modeler,Thanks for joining in for the Loess case!T...

2008-07-28T01:15:49.846-04:00

Modeler,

Thanks for joining in for the Loess case!

The idea that intense interest is correlated with transitions between states is very useful. It is empirically about right, but it is also interesting that it is not quite right empirically: the June-to-July transition was correlated with fairly low interest and, in fact, may have been caused by it.

Another thought related to a previous conversation we had: once we start believing in more steady states, it makes much more sense to have a regression on the polling during those periods. So could Nate please provide us with the 538 regression over the following periods:

1/27 to 3/3
5/3 to 5/5
5/7 to 6/2
6/8 to 6/30
7/5 to 7/24

I am not sure by the way that 5/6 was a major transition. From the data, I have a sense that March numbers are a bit contaminated with polls that were over-exposed to media interest around Wright, while slightly better numbers in May may represent a technical process, namely the gradual elimination of Clinton questions.

In a previous, "humanistic" and qualitative comment in the previous thread I've suggested that Obama's numbers essentially have an "on" and "off" button, "on" when there's the feeling around of the fierce urgency of now and "off" otherwise. I think the "on"/"off" button is worth about 2 points; plus, there are about 2 points in the soft PUMA vote that claimed, never quite sincerely, not to be able to vote for Obama against McCain.

Hence a simple theory:

Feb. is "On", without soft PUMA. It's at +2 Obama.

March-May is "off", without soft PUMA. It's generally speaking +0 Obama, a bit less when the news cycle is dreadful, a bit more as the soft PUMA begin to drift.

June is both "on" and with soft PUMA. It's +4.

July is "off" and with soft PUMA. It's +2 again (but the opposite +2 to February!).

There is always the possibility that there's even more to be squeezed out of the "on" button; but I suspect today's Gallup is too optimistic and that we will settle, at best, at around +4 to +5.

A couple of comments on things I read:1. One post...

2008-07-28T01:14:14.075-04:00

A couple of comments on things I read:

1. One poster said that although the phone polls were generally accurate, the exit polls were way off and that might be an example of the Bradley effect, since people would be more likely to lie face to face. I hadn't heard that before. I thought the exit polls during the primaries were generally very accurate, but I confess I didn't look that closely. Is there a link to an article that covers this directly?

2. The Obama stuttering tape.

I don't normally answer trolls, but... One posted a youtube link to a supposed tape of Obama repeatedly saying the word "uh" during a press conference. The tape was audio only and had no visuals. It was an obvious fake. Its poster said people had said it was a fake, but they should google it and they would see many news stories about it. It even had links to three news stories. The news stories referenced the press conference but didn't say anything about stuttering. Googling Obama stutter didn't land one single quote from a reputable news organization.

Just another smear. Here is the question though. Why? I really don't understand what they have to gain from this one.

It seems to me that the Republicans are trying to spread the rumor that Obama is really a simpleton and that his unscripted debate performances will show him to be a complete moron.

Dems tried this with both Reagan and George W Bush. It failed miserably both times. They set the bar absurdly low and when those two didn't completely embarass themselves at their debates, it looked like they had won easily. I'm not sure the "my opponent is an idiot" attack has ever worked in an actual campaign.

Juris, thanks for the response. I should clarify ...

2008-07-28T01:10:52.170-04:00

Juris, thanks for the response. I should clarify that I was really asking about the general impact of racism on people's voting, not about the Bradley Effect. (In fact, I wouldn't be surprised if the Bradley Effect has disappeared, due to the decline of political correctness since the 90's.)

Our society's underlying racism, of course, has not disappeared: the latest resume experiments I found in a quick Google search were dated 2003 and 2004. I was just wondering if anyone has done experiments similar to the resume ones that could estimate the magnitude of the effect of racism in a hypothetical political contest.

Comments on FiveThirtyEight: Politics Done Right: Today's Polls and Miscellaneous Thoughts, 7/27

情趣用品,情趣用品,情趣用品,情趣用品,情趣用品,情趣用品,情趣用品,情趣用品,情趣用品,情趣用品,...

Anon at 5:30 PM.I do wonder how the kinds of quest...

Er, no. Obama is *well* ahead in the electoral co...

Obama at start of death spiral based off todays Ra...

"If you do not have the competance to win a campai...

KRANKY said...I like that Harry! One wonders if Op...

On the racism front: I wonder, on the basis of no ...

Let me add myself to the Karl Rove is overrated ca...

"I would bet that McCain announces between the end...

Are people intentionally mispelling Charlie Crist'...

A name is just a name. In every election, you neve...

choking as I type...Rove's current map of EV proje...

Juris @ 9:27 PM said... "If I didn't actually have...

re: Cugel @ 8:54 PMyou said, "Although, if I were ...

Modeler, I think the supertracker is wrong about J...

KRANKY said...Because of an aging demographic that...

@ anon 12:29Face the facts, bub: Obama has been do...

Anonymous Humanist,Take a look at the Super Tracke...