It is time for Zogby to quit the game.
I am sure he can afford to join me in Thailand, or wherever.

You moving to Phuket? Say hello to the Gent and give him my best regards!

I haven't decided between Pattaya and Chiang Mai.
What do you think?
Phuket is too pricey for me.

My God, Pattaya? That's about the last place I'd pick. Too many Farangs and too many "touts"
CM would be the obvious choice.

Pattaya is full of those "retired" US Army specialists if you get my drift

Zogby's final pick reveals the depth of his failure:

POPULAR VOTE
Bush 49.4%
Kerry 49.1%

ELECTORAL VOTE
Kerry 311
Bush 213
Colorado and Nevada were “Too Close To Call"

Zogby was off significantly. He hosed this so badly because he based his sample weighting on 2000 exit polling date where people self identified as democrat, republican or independent (39%, 35%, 26% respectively - working from memory there so I may be off slightly but the general ratio is correct).

Exit polls are non-scientific polls, more art than anything else and prone to considerable error as we saw early yesterday afternoon. Then to base his polling samples on such flawed data reveals the partisan nature of Zogby. He knew the problem with exit polls and he also knew the ratios people were identifying themselves at had changed in the afermath of 9/11 yet still went with the oversampled democrats and independents.

Given his track record over the last two election cycles (more than twice as many missed calls as any other pollster), we should never listen to Zogby again.

Not to defend Zogby because he BLEW IT, but it is my understanding that his polls are CALLING on the TELEPHONE POLLS, NOT exit polls.


I think it is clear that a large number of people are now LYING to all kinds of pollsters.

Someone who can devise a better method can do very well. Any ideas?

I one sense I agree with you; Zogby should retire.

Zogby defends himself:

Statement from John Zogby on 2004 Presidential Election Results:

“We feel strongly that our pre-election polls were accurate on virtually every state. Our predictions on many of the key battleground states like Ohio and Florida were within the margin of error. I thought we captured a trend, but apparently that result didn’t materialize.


“We always saw a close race, and a close race is what we’ve got. I’ve called this the Armageddon Election for some time—a closely-divided electorate with high partisan intensity on each side."

Not much to be done about it. Educating ourselves on polling and statistics is the best we can do. Over the last couple of weeks I've been hitting some very hard statistics material (prompted even more heavily with the recent 100,000 Iraqi dead story in Lancet - a fatally flawed report). It allowed me to see the exit polls for what they were even when the MSM was hyping them as though it was the end of the election.

Someone who can devise a better method of educating the masses on this will do very well. It's damned hard to understand and a small part of it is an art rather than science - although the science part of it is sufficient to reject conclusions when someone takes too much artistic license. I have an engineering background with a lot of math and even a few college statistics courses behind me and it totally boggles my mind how they do this stuff. The average guy that struggled through high school algebra is going to have to be pretty dedicated to slog through this kind of material and get anything meaningful. Note that I think anyone can do it, it's just really, really painful.

I'm always reminded of my first statistics professor who told us onthe first day of class that we need to learn about this so we can "seperate the lies from the ###### lies". In retrospect, maybe one of the best professors I ever had...