[Spiff]

Margin of sampling error for each is +/- 4 percentage points with a 95% level of confidence. That makes second, third and fourth statistically tied.

16 - 11 = 5

4 does not equal 5

3rd and last are statistically tied. 2nd is not.

And you never made this argument when Thompson was ahead, but statistically tied with Romney. Interesting timing how you just trotted out your poor math.

[colorcountry]

If romney is actually three points under 16 = 13, and McCain is one points over, and Huck and Thompson are two points over, then we have a whole new ball game. Everyone is tied at 13. Sheesh

There are none so blind as those who will not see.

[kevkrom]

4 does not equal 5

Nor does it equal 8 -- within the MOE, Romney could be as low as 12 and Thompson as high as 15.

That said, the extremes of the MOE are more often seen on a 50/50 proposition. For lower values, you are much less likely to see results off the "true" value by as much as the MOE.

All in all, what we're seeing is the net result of the relentless negative drumbeat against Thompson coupled with some key endorsements that Giuliani and Romney have picked up. Now that Thompson has started running ads, picked up the campaign pace, and received the endorsement of the NRLC, I expect to see a turnaround in a few days.

[Petronski]

3rd and last are statistically tied. 2nd is not.

You're just full of lies and disinformation aren't you? Fred is not last in this poll. As for the margin of error, each candidate's numbers are +/-4.

[colorcountry]

BTW Spiff, I am a Hunter supporter. You can see exactly how much polls mean to me.

[MEGoody]

+/- 4% works like this:

Giuliani 27% (meaning his support falls in the range of 31% to 23%)

Romney 16% (support in the range of 20% to 12%)

McCain 13% (support in the range of 9% - 17%)

Huckabee & Thompson 11% (support in the range of 15% to 7%)

So in essence, there is a statistical tie for second place among all the candidates listed except Rudy.