[colorcountry]

Margin of sampling error for each is +/- 4 percentage points with a 95% level of confidence.

That makes second, third and fourth statistically tied.

[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.

[org.whodat]

That makes second, third and fourth statistically tied.

True and 22% undecided. Poll doesn't mean spit!

[curiosity]

Romney's numbers aren't the story here. They haven't changed much and appear to be fluctuating within a band of about +/-5. Of course that isn't surprising since he's not running a national campaign but rather an early state strategy.

The big story here is Thompson's big decline, dropping from close to 30% down to just over 10%! That's huge, both statistically and economically.