Most Accurate 2004 Poll

[irvin88]

McCain has cut into Obama's lead for a second day and is now just 1.1 points behind. The spread was 3.7 Wednesday and 6.0 Tuesday. The Republican is making headway with middle- and working- class voters, and has surged 10 points in two days among those earning between $30,000 and $75,000. He has also gone from an 11-point deficit to a 9-point lead among Catholics.


About IBD/TIPP: An analysis of Final Certified Results for the 2004 election showed IBD's polling partner, TIPP, was the most accurate pollster of the campaign season. Learn more at www.TIPPonline.com.

[Cajuncowboy]

I already posted this and it was roundly condemned by the leftists.

:lmao:

Truthfully, you will see the polls start to shrink each day I think because in the end the pollsters want to get it right because it makes them look like idiots when they don't And if they keep up with these huge double digit numbers when everyone knows it's very close they will end up looking like fools.

Like they did in 200 and 2004.

[masomenos]

I already posted this and it was roundly condemned by the leftists.

:lmao:

Truthfully, you will see the polls start to shrink each day I think because in the end the pollsters want to get it right because it makes them look like idiots when they don't And if they keep up with these huge double digit numbers when everyone knows it's very close they will end up looking like fools.

Like they did in 200 and 2004.

Yeah projection margins always shrink as election day approaches, it is odd for it to make a shift to the other side this late though. We shall see.

[Heisenberg]

There are 3 threads about this poll already. The major problem with it has already been pointed out and with the problem corrected, the poll falls right in line with others.

Look at the internals. Notice the 18-24 age group. McCain leads by 50. Good luck with that one.

[silverbear]

McCain has cut into Obama's lead for a second day and is now just 1.1 points behind. The spread was 3.7 Wednesday and 6.0 Tuesday. The Republican is making headway with middle- and working- class voters, and has surged 10 points in two days among those earning between $30,000 and $75,000. He has also gone from an 11-point deficit to a 9-point lead among Catholics.


About IBD/TIPP: An analysis of Final Certified Results for the 2004 election showed IBD's polling partner, TIPP, was the most accurate pollster of the campaign season. Learn more at www.TIPPonline.com.

ROTFLMAO... I'm gonna rain on your parade again, Irvin...

First off, I'll repeat, guys like you who rely on the results of ONE poll, because you like what they have to say, are intellectually dishonest... you're propagandizing, trying to convince people (and probably, yourself) that the race is closer than it is...

But more important, fivethirtyeight.com, that well-regarded website that takes a CLOSE look at the polls, has an article up calling the methodology of the poll you're touting here, into serious question... I'll post the article here, but basically they're saying that these folks severely undersampled the 18-24 year old vote, a demo that heavily favors Obama:


That's right ... IBD/TIPP has John McCain ahead 74-22 among 18-24 year olds. Who knew the kids were groovin' on J-Mac these days?

IBD/TIPP puts an asterisk by this result, stipulating that "Age 18-24 has much fluctuation due to small sample size".

Indeed, there may be some fluctuations when looking at small subgroups like these. That's why I generally don't pick on a poll if, say, it has John McCain winning 18 percent of the black vote when he's only "supposed" to be winning 7 percent or whatever. Fluctuations of that magnitude are going to be relatively common, mathematically speaking. In fact, they're entirely unavoidable, if you're taking enough polls and breaking out the results amongst enough subgroups.

But fluctuations of this magnitude are an entirely different matter.

Suppose that the true distribution of the 18-24 year old vote is a 15-point edge for Obama. This is a very conservative estimate; most pollsters show a gap of anywhere from 20-35 points among this age range.

About 9.3 percent of the electorate was between age 18-24 in 2004. Let's assume that the percentage is also 9.3 percent this year. Again, this is a highly conservative estimate. The IBD/TIPP poll has a sample size of 1,060 likely voters, which would imply that about 98 of those voters are in the 18-24 age range.

What are the odds, given the parameters above, that a random sampling of 98 voters aged 18-24would distribute themselves 74% to McCain and 22% to Obama?

Using a binomial distribution, the odds are 54,604,929,633-to-1 against. That is, about 55 billion to one.

So, there is an 0.000000002% chance that IBD/TIPP just got really unlucky. Conversely, there is a 99.999999998% chance that one of the following things is true:

(i) They're massively undersampling the youth vote. If you only have, say, 30 young voters when you should have 100 or so in your sample, than the odds of a freak occurrence like this are significantly more likely.
-or-
(ii) Something is dramatically wrong with their sampling or weighting procedures, or their likely voter model.

My guess is that it's some combination of the two -- that, for instance, IBD/TIPP is applying a very stringent likely voter model that removes you from the sample if you haven't voted in the past two elections, which would rule a great number of 18-24 year olds out.

A pollster could get away with a turnout model like that in 2004 (when IBD/TIPP did well in estimating the national popular vote), when the split in the youth vote was relatively small between John Kerry and George W. Bush. They can't get away with that this year, when the split is much larger.

But the basic takeaway is this: you should absolutely not assume that just because someone has published a poll, they have any particular idea what they're doing. Pollsters should be treated as guilty until proven otherwise.

You make this too easy, man... :D

[silverbear]

I already posted this and it was roundly condemned by the leftists.

As it should have been, their methodology is highly questionable...

[AbeBeta]

As it should have been, their methodology is highly questionable...

Silverbear -- don't know if you know this site but they also provide a projection like 538.

http://election.princeton.edu/

Today in fact, they have an article on single polls vs. aggregates. This is a nice companion to 538 - they use slightly different methods but are pretty consistent in terms of outcome. 538 is based more in methods common in fields like economics whereas the Princeton election consortium uses methods common to psychology and medicine

[silverbear]

Silverbear -- don't know if you know this site but they also provide a projection like 538.

http://election.princeton.edu/

Today in fact, they have an article on single polls vs. aggregates. This is a nice companion to 538 - they use slightly different methods but are pretty consistent in terms of outcome. 538 is based more in methods common in fields like economics whereas the Princeton election consortium uses methods common to psychology and medicine

No, I hadn't heard of it, Abe, and I'm looking forward to checking it out in depth... thanks for the heads up...

[zrinkill]

I am sorry ...... but what good is a poll that says the guy you are supporting is barely losing.

[jimnabby]

About IBD/TIPP: An analysis of Final Certified Results for the 2004 election showed IBD's polling partner, TIPP, was the most accurate pollster of the campaign season. Learn more at www.TIPPonline.com. (http://www.TIPPonline.com.)

Today's IBD/TIPP poll is out: Obama 45.8, McCain 42.3, for a 3.5 point Obama lead.

[AbeBeta]

Today's IBD/TIPP poll is out: Obama 45.8, McCain 42.3, for a 3.5 point Obama lead.

What is remarkable about this is that they still have a crazy youth vote reported -- 30% Obama, 64% McCain. Obama leads with all other groups except 65+ where McCain is shown with a 4% lead. So based on this poll, McCain is doing great among 18-24 voters. That dog don't hunt.