Ken writes:

> Comparing the election results to the polling results saved at
>  http://aics-research.com/election2004july.html
>  (I just eyeballed this quickly, so I might have a mistake.)
>
>  States that voted differently in Nov from how they polled using
>  traditional polling methods in July:
>
>      AZ IA MO NM NV OH
>      and TN was tied at that time
>
>  AOL Straw Poll
>  States that voted differently in Nov from what the AOL straw poll
>  indicated in July:
>       AK AR AZ CO FL IA KS KY MO MT NE NM NV OH TN VA WV WY
>
>      That is, all of the above, plus :
>     AK AR CO FL KS KY MT NE VA WV WY

If the implication is that electoral-vote.com poll proved to be more correct
than the AOL "self-selected" poll in the results I placed on our webpage, then
that is correct, but that "correctness" is also misleading. The AOL poll
contained a great deal of inertia, being the sum of the month's voting, so it
didn't move much from day to day. The electoral-vote.com maps were a daily
expression of the most recently released polls on a state-by-state basis and thus
were very volatile. During July, it was possible to observe days when aggregation
of the daily polls were essentially equal to the AOL results, as you can see
in the tracking chart at the site:

     http://www.electoral-vote.com/info/graph.html

I posted those various results at a time in July when Kerry had fallen to his
lowest levels during the month, as marked by electoral-vote.com, thus the
difference seems larger than it was.

Of interest, in regards to how accurately these kinds of things can be
predicted, the electoral-vote.com site had Kerry beating Bush 298 to 231 on the day
before the election:

     http://www.electoral-vote.com/nov/nov01.html

There is a great deal of randomness in this process. Although the end result
is to declare the architects of the victories "geniuses," while the other side
retreats into deep "soul-searching," things could have been very easily
reversed. For the winners to declare a "mandate" is almost always a stretch.

Wirt Atmar

* To join/leave the list, search archives, change list settings, *
* etc., please visit http://raven.utc.edu/archives/hp3000-l.html *