The Gallup Critique and the Margin of Error.

There have been several people who have jumped onto the Gallup tracking result (which currently shows Romney ahead nationally by 6.)

Nate Silver of fivethirtyeight.com offered a critique on Thursday, and several others have jumped into the fray.

Silver discusses Gallup's recent history:

Usually, when a poll is an outlier relative to the consensus, its results turn out badly.

You do not need to look any further than Gallup’s track record over the past two election cycles to find a demonstration of this.

In 2008, the Gallup poll put Mr. Obama 11 points ahead of John McCain on the eve of that November’s election.

That was tied for Mr. Obama’s largest projected margin of victory among any of the 15 or so national polls that were released just in advance of the election. The average of polls put Mr. Obama up by about seven points.

The average did a good job; Mr. Obama won the popular vote by seven points. The Gallup poll had a four-point miss, however.

Everything he says about the past is true.  But how do we know Gallup is an Oulier this year?

In the final RealClearPolitics average in 2008, the Average was Obama +7.3, while the final result was Obama +7.6.  But Gallup (and Reuters the other poll that was Obama +11) made up part of that average.  Without them they would have been further off (by about 0.3 for each poll).

Another point is that this was the final poll results vs. the final results.

Looking at the 2008 polls roughly two weeks before the election (on or about Oct 20th).   The average was still around 7.1 and Gallup was right at 7.   In fact at that point Pew Research was showing Obama ahead by 14.  (Pew's final number was Obama +6).

In the 2008 final 15 Polls that made up the RCP final average, 5 of them were at Obama +7 or Obama +8.   5 more were at +6 or +9 (1 point off in either direction).  The other 5 were +11, +11, +5, +5, and +2.

But here's the key fact:  THE FINAL RESULTS WERE INSIDE THE MARGIN OF ERROR OF EVERY SINGLE POLL IN THE FINAL AVERAGE.

We look a lot a the actual result margin, but the Margin of Error applies to each statistic, not the margin of victory.  So if Candidate A gets 54% and Candidate B gets 43% in a poll, but the margin of error is 2%, then the final results can be A 52-56% and B 41-45%.   That was the final Gallup poll and although Silver calls it a disaster, technically they had it within the Margin of Error.

So looking at the Current Gallup Survey, it shows Romney with 51% to Obama's 45% with a MOE of 2%.  This means it could be as close as Romney 49% Obama 47% or as wide as Romney 53% Obama 43%.   At this point Romney is beyond the MOE for victory.  Let's see if that is true on November 7th before calling Gallup out on it.