Bill.N. Wrote:
-------------------------------------------------------
> Thomas Sowell Wrote:
> --------------------------------------------------
> -----
> > I know this has to be depressing for you Bill
> > knowing your background. But it is over.
>
> Was it over when the Germans bombed Pearl Harbor?
> (If you don't get the joke, watch Animal House.)
>
> Seriously though, while I'm sure you enjoy the
> idea of me gnashing my teeth on the latest poll
> numbers the reality is that I've been discounting
> NATIONAL poll numbers for some time. Did it back
> at the beginning of the year when someone was
> boasting that "Generic Republican Candidate" was
> pounding Obama. Did it when polls seemed to show
> Obama pulling away, and am doing so now. My
> position on this election is now as it has been
> for months, namely it comes down to Pennsylvania,
> Ohio and Florida.
>
> And much as I enjoy your Gallup poll conspiracy
> theory the Gallup numbers at that point were not
> much different that what other pollsters were
> reporting at the time.
Your Belushi reference shows your comedic genius my friend. Excellent reply! I was disarmed immediately as laughter ensued.
But alas we disagree. The only poll Gallup compared to right before Axlerod intervening was Rassmusen which was using the same weighted sample as Gallup at the time. Axlerod makes a few contacts then all of the sudden Gallup shifted their weighted sample from app. 4 points +Obama to 9 points +Obama bringing them more in line with WSJ/NBC and others.
The polls showing Obama way up were using samples weighted as high as 5 points higher than the 2008 election model which would say that there is even more enthusiasm for Obama supporters to vote now than 08. That simply isn't the case.
It's just like this tremendous swing in most polls to Romney since the first debate. There is no way millions of women in Ohio and other swing states suddenly changed their allegiance virtually overnight after one debate. Rather the polling organizations showing Obama way up before the debate shifted their weighting closer to reality.
In this way I do believe certain pollsters try to create news rather than quantify real information. As we get closer to election day they adjust so as to preserve whatever reputation for accuracy they have.
It is comparable to the BLS unemployment statistics. They dropped off a few million people out of the work force to get their 7.9 or whatever. Now if you are one of the 24 million Americans that are either unemployed, underemployed or not looking period is a little number like 7.9% from a nebulous entity like the BL really going to effect your vote? I think not.
Same thing with the polls. It wasn't a month ago all the pundits, even conservatives, were saying, "it's over". Romney's toast. that made most Romney supporters feel like crap which is exactly what many polls were designed to do. Suppress Republican enthusiasm. This would work if Obama had just a few things he could trumpet other than OBL and GM.