*****EVERYBODY THIS THREAD IS CLOSED*****

Go to this one, it’s newer and has free beer!
Another Dull Debate or Blue Touch Paper, Lit?

well, I never said it was going to be pretty

If your 50/48 poll had a margin of error of +/- 3%, that doesnt mean that the true result is equally likely to fall anywhere between 47 and 53 – it’s more likely to fall nearer to 50 than it is nearer to either 47 or 53 because of the way results are normally distributed (within a bell curve shape)

Roughly it means that there’s a two in three chance that the true value is between 48.5 and 51.5, and a 95% probability that the true value lies between 47 and 53.

The margins of error that all pollsters use everywhere are mostly hogwash. It’s based on a theoretical minimum margin of error that would occur were

a) were the pollsters using only pure random sampling.
b) The entire population was equally likely to be polled by a pollster at any given time.

Pollsters dont use pure random sampling and we know that B is wrong (for instance, younger people are less likely to be at home in the afternoon than little old ladies watching Matlock reruns)

But here’s the tricky bit – in theory, any additional tweaking to that sampling that a pollster uses (such as engaging in quota sampling, then weighting the results for age, gender, income, partisan ID etc etc) theoretically causes the margin of error to blow out, and blow out substantially.

Yet in practice, with good pollsters, it doesnt actually blow out at all and remains around the theoretical minimum.

This doesnt mean that theory is wrong, it just means that certain polling data manipulations utilise known human behavioral patterns to improve polling accuracy in ways that make this basic type of probability theory redundant to solving the problem at hand.

The big problem here though is that not all pollsters are good pollsters, and even the good pollsters get it wrong sometimes. Good polling that delivers high accuracy is as much of an art as it is a science.

“……..there is something about the latest PBoC which should indeed cause worry. For me one of the bad-case scenarios that we have most to worry about is a sudden reversal of hot money inflows, large enough that it puts liquidity pressure on the formal and informal banking systems. This is clearly not a problem yet, but the shift in a matter of months from massive inflows to moderate outflows is not confidence building.

As a related aside, and I am now straying into areas about which I need a lot more information, by coincidence I had two meetings yesterday – one with a world famous Harvard economist and a group of PKU professors, and the other with a group of traders and bankers – in both of which South Korea suddenly became the topic of conversation. I am no expert on Korea but the kinds of things I was hearing raised all my Latin-American-bond-trading hackles. One of the academics said he thought that Korea would come under tremendous liquidity pressure in the next three months. If there are problems once again in Korea I would lay pretty serious odds that capital flight will become a serious problem all through East Asia.”

……..
The outlook for China and the East Asian economies is very troubling. In 1997, the developed economies helped lift the East Asian economies out of their collapse. This time, the lifeboats are already full of bankers and hedge funds. Keep your eyes on China….the real home of the boom of the last 20 years….

Frivolity first. The premiere of Oliver Stone’s movie “W” was held next to our hotel last night. Papparazzi abounded. It was hard to see the celebrities through the people, but I managed to spot Ted Danson. OK, I know that’s pretty lame, but at least I saw someone moderately famous.

I thought about being a groupie outside tonight’s debate location, just to try to see Obama, but the debate’s being held on Long Island, rather than Manhattan.

Anyhoo, you need to read this article from today’s NY Times re the latest Times/CBS polling.

Over all, the poll found that if the election were held today, 53 percent of those determined to be probable voters said they would vote for Mr. Obama and 39 percent said they would vote for Mr. McCain.

Voters who said their opinions of Mr. Obama had changed recently were twice as likely to say they had grown more favorable as to say they had worsened. And voters who said that their views of Mr. McCain had changed were three times more likely to say that they had worsened than to say they had improved.

The top reasons cited by those who said they thought less of Mr. McCain were his recent attacks and his choice of Gov. Sarah Palin of Alaska as his running mate.

There’s much more yummy goodness in that link.

http://www.nytimes.com/2008/10/15/us/politics/15poll.html?_r=1&hp&oref=slogin

Olbermann: McCain, suspend your campaign

Huh???

“We’re completely surrounded. Don’t let one of them get away!!”

But they still rock.