Why is everyone so worried about a poll that's statistical TIE?

[Brotherjohn]

So what if the Zogby poll has Bush up by 4! Statistically, that's not a change at all; they're still tied (MOE=2.9pts).

People are fretting that why, after Kerry won the third debate, has BUSH gained momentum. Well, he HASN'T. Statistically, nearly every poll is a tie.

The big picture shows that, since the debates began, Kerry has turned a statistically significant deficit (in most polls) into a tie. Nothing has happened to reverse that momentum.

It will take a few more days for the acknowleged fact that Kerry SWEPT the debates to reflect in poll numbers. I suspect Kerry will have another poll surge (similar to after the first debate) over the next week or two. This will leave him up by 5-10pts, and it will be too late for Bush.

[Nicholas D Wolfwood]

1. All of the polls are bullshit

Do not worry and do not read them.

[onehandle]

2. Tie=Kerry Win.

It's simple. Apply the 2000 math to the current "polls" and Kerry wins easily. Gore was behind as much as ten points at this time in some polls. And he won.

Even=President Kerry.

[DaveinMD]

3. its just

one weird day in the sample. This happens in tracking polls because the daily sample is quite small.

[FlashHarry]

4. The problem isn't the poll, it's the headline.

Dumb people see "Bush up by four points in new poll" and they think, "gee, this guy's winning. I'd better vote for him. I don't want to vote for a loser."

To paraphrase Tommy Lee Jones, "A person is smart. People are dumb."

[DaveinMD]

5. most of the polls

show Kerry tied or in a slight lead. Not worried about one tracker. The momentum is in our favor.

[MercutioATC]

6. Actually, a MOE of 2.9 puts Bush ahead...

MOE is the total margin of error. You can't apply it to both numbers.

That said, I agree with your basic premise. Day-to-day polls are going to fluctuate. There's very little to be gained by getting worked up over it.