We obviously won’t know the final results of this election until after November 2.  However, in evaluating this race and all the recent polls, it would be well to keep the following in mind.

1.  Most voters have already made up their mind.  Those who have not yet decided to vote for an incumbent almost always vote for the challenger.  This is a well documented fact.  For this reason any incumbent failing to poll above 50% this late in the race is in jeopardy.  Think Barbara Boxer and Harry Reid
2. All of the polls are based on “expected voters.”  All that a pollster needs to do to make it look like Democrats are catching up is assume that Democratic participation will be the same as in previous elections.  However, that has not been true for a long time, at least as far back as the Scott Brown election in Massachusetts.
3. In recent elections all of the polls have grossly underestimated the Republican vote and also grossly underestimated the tea part influence.   That is why no one was predicting Joe Miller beating Lisa Murkowski in Alaska.  No one was predicting Christine O’Donnell beating Mike Castle in Delaware.  The list goes on and on.  Not only did these candidates win, they won by large margins.  For example:  the polls showed Joe Miller down by double digits less than a week before the election.  Some of this can be explained by brilliant last minute media buys by the Tea Party Express, but at least some of it is the result of the polls being really wrong.
4. Nothing has happened to change this election.  There has been no October surprise and Obama’s campaigning has done more harm than good.  The more campaigning by Obama, the more this becomes a nationalized election and the more Democrats lose. 
5. Democrats have only made gains in a couple of states where the Republican candidate has screwed up.  Classic examples are New York Governor and Delaware Senate.  Democrats may pick off one or two candidates, but they cannot stop a rising tide.
6. The numbers to really watch are Gallup.  They are predicting a 17 point Republican advantage based on normal turnout.  They consider an average turnout to be about 40%.  But, they are still predicting a 12 point Republican advantage if there is a high turnout of about 55%.  But, in my personal opinion, that this year is different and if there is a high turnout it is likely to be even more Republican.  If that is true, then anything could happen.
7. None of the polls are really taking the Gallup results into consideration.  Even Rasmussen does not do that.  The reason is that the pollsters are terrified of being wrong.  If they are wrong they don’t want someone accusing them of having made a material difference in technique. 

The bottom line is that this election will probably be unlike any other.  Just think of all that has happened this year.  Scott Brown, a Republican, won Ted Kennedy’s old Senate Seat.  Several incumbent senators have been picked off in primaries by underfinanced unknowns.  Others, like Evan Bayh and Christopher Dodd, didn’t even bother to run for re-election because they knew they were going to lose.  There is a high tide and a storm surge coming that will lift all ships.  The only question remaining is the ultimate size of the tsunami.

TDM