The silly season is here again. Polls are being touted as the infallible indicator of the actual electoral outcome. I think not.
Now Obama could very well win. The fall financial crisis was a godsend to The One. But the race is surprisingly tight, and the infamous month of October is not done.
Let's take a look at some other poll predictions,
Poll: Advantage Kerry in Ohio RaceThis story, dated 19-Oct-2004 put Kerry over Bush 50 to 47 in the critical swing state. Actual outcome: Bush 51 Kerry 49. Understated Bush by 4.
More updates as the spirit moves me
And in late October 2000, CNN reported that CNN/Time poll: Bush holds edge Bush was leading Gore by 6 points. As we know, the election was virtually tied, with the final outcome showing Gore up by about a half percentage point. (The poll probably does not reflect the impact of the enormously damaging "drunk driving" story, skillfully sprung by a Democratic operative in the last week of the campaign.)
The 1948 Classic: The Man on the Wedding Cake vs The Man from Missouri
The 1948 election was a classic- and a pollster's nightmare. Suave urbane Thomas Dewey, the favorite of the then dominant northeast elites (sound familiar?), seemed to be the candidate from Central Casting. Polls were overly reliant on telephone sampling, which overrepresented - well, dominant northeastern elites and their like-minded brethren around the country. Scrappy Harry Truman fought a vigorous, shamelessly partisan campaign and carried the day.