Is it Random Noise, the Palin Effect, or the Bailout Bill?

In analyzing the presidential race to this point I have focused my attention primarily on national survey data.  As I noted in an earlier post, that survey data indicates that Obama increased his lead over McCain by approximately 6% in the period from Sept. 15 through Oct. 3. It now appears that the credit-crisis induced movement has crested, and there are even signs that the race is beginning to tighten again. Four of the five national daily tracking polls show a slight movement toward McCain over the last three days.  These include Rasmussen, which shows Obama losing 2% to drop to a 6% lead over McCain;  Hotline, which shows McCain gaining 4% to move within 1% of Obama, GW Battleground, which shows McCain cutting his deficit from 7% to 4%, and Zogby, which shows Obama losing 1 point to lead McCain by 2%.  Note that all these movements are within the margin of error and may simply reflect random noise rather than any real underlying movement toward McCain.  In this vein, note that a fifth daily tracking poll run by Gallup shows Obama actually increasing his margin over McCain in this period from 9% to 11%. How do we explain the difference between Gallup and the other polls?  There is always some random variation in polls so one cannot be sure why one poll seems to be an outlier. But notice that Gallup is the only one of the five tracking polls that is still sampling registered, as opposed to likely, voters.  At this stage in the race, most pollsters switch to a likely voter sample.  Gallup has not done so, and that may explain why it continues to show Obama increasing his lead.  Generally, as I noted in an earlier post, McCain does better in likely voter polling.

If – and it is a BIG if – McCain has stemmed the Obama surge, what is the explanation?  One is tempted to attribute it to Palin’s debate performance on Oct. 2. All of these tracking polls were taken on Oct. 5-7, beginning two days after the debate, or about the time that Palin’s performance had been dissected by the media and circulated among the water cooler crowd.  It may be, then, that Palin’s debate performance halted the McCain slide.

However, I think a more likely explanation is the passage of the bailout/rescue bill which George Bush signed on October 4.  That momentarily stemmed the financial panic and staunched the bleeding from the McCain campaign.  However, the recent financial turmoil that necessitated intervention by the central banks in several nations, as well as the roller-coaster stock market, threatens to erode any gains McCain may have made since the bailout bill was signed.

Whatever the explanation for the slight movement toward McCain in the latest national tracking polls, Obama is taking no chances.  I watched his speech at a campaign rally in Indiana today.  Arrayed behind him on stage were ONLY women – all white, middle-aged women.  That was not by accident.  My guess is his internals are picking up the first indications of a potential Palin effect and he realizes that married women are the key swing group in this election.  Recall his debate performance – he emphasized those issues (education, health care) that are particularly important to women.

Palin, meanwhile, has been holding rallies in key battleground states, starting with Florida where she held four rallies, and then on to North Carolina and Pennsylvania.  When she works the rope lines at these rallies, invariably she is surrounded by women, many holding babies.  But is she having an impact?  And will it be enough to push the financial news off the front page?

We should have a better answer in the next few days as the news of the financial intervention by the central banks begins to have an impact among voters. In a later post I’ll check the tracking numbers in the states Palin has visited.  In meantime, however, I want to turn to examine where the candidates stand in the Electoral College.  I’ll do that in my next post.