Nothing wrong with the polls!

I am perfectly satisfied with the analysis offered on Keith Olbermann's Countdown a couple of days after the election. He and Craig Crawford, of Congressional Quarterly, went through the numbers, and if I recall them more or less correctly, it came out that the percentage of the vote predicted for Obama (between 34 and 38) was just about right — he landed at 36%.

The number that was wrong, that had been forecast for Clinton, was 28%, obviously low. But then the number of UNDECIDED or MIGHT CHANGE votes was over 40% or so.

Looks like the majority of those broke for Clinton, maybe because of tears, maybe because of sympathy, etc., or maybe just because a lot of Clinton voters held back until the last minute. And it's reported that lots of those were women over 45 who (anecdotally) felt HRC had been mistreated.

Some independents who voted for McCain did so, they said, to stop Romney, but only after they were assured by the polls that Obama had it nailed. That would explain McCain's unexpectedly high numbers.

In the end, the POLLS weren’t to blame — the interpreters were.