There has been a lot of talk about the stability of the poll numbers in this year's presidential election due to the polarized electorate, but as we've seen in the last few months, there is room for movement. From mid-May when John Kerry announced John Edwards as his running mate through early August after the Democrats held a successful if lowkey convention, Kerry turned a stubbornly tied race into about a three point lead. Then a perfect storm seemed to develop for Bush wherein the media obsessed over the Swift Boat Veterans for Truth ads, which took up a couple weeks in August leading up to an extremely strong convention performance, after which Bush took the lead by anywhere from 3-10 points depending on whom you believe. It's become clear that even in this tight election, events both unforseen and premeditated can move the numbers a little bit any which way and these will ultimately be the determining factors of this race.
So I wanted to find out, what were the determining factors 4 years ago? Did the candidates get bounces from their conventions? From their VP picks? Were the debates important? And from this information, what can we deduce might happen over the next month and a half? Luckily a friend and contributor to my blog, Wild Democracy Ride, painstakingly tracked every poll in 2000 and kept them in a massive spreadsheet. So here it is, a look back at 2000.
Since neither Gore nor Bush was an incumbent, they each had to fight for his party's nomination. Gore had a relatively easy time of it, dispatching Bill Bradley swiftly in Iowa and New Hampshire, but Bush's fight was a bit harder. He was essentially the presumptive nominee since he had the backing of the party apparatus, but John McCain waged a strong fight, winning the New Hampshire primary and coming close in South Carolina. As a result, Bush emerged from the primaries a bit weathered and worn and what was a solid lead over Gore in January and February became a tight race in early March. From March 8 to April 8, Bush held an average lead of 1.1%.
Between April and June, however, by taking a page out of Clinton's playbook and coming out with strong stances on traditionally Democratic issues such as education and social security, Bush widened the gap between him and Gore who was busy ineffectually attacking Bush without offering any ideas of his own. From April 11-June 13, Bush led Gore by an average of 4%.
In June of 2000, Gore was plagued by questions about whether he committed campaign finance violations in 1996 when he went to, apparently unknowingly, a fundraiser at a Buddhist temple. This allowed Bush to widen his lead to an average of 8% at the story's peak in the second half of June but Janet Reno's refusal to appoint a special prosecutor cooled the story off, allowing Gore to regain some ground. From July 1-July 25, Bush held a modest 2.5 % lead.
On July 25, Bush announced Dick Cheney, the man in charge of finding him a running mate, as his running mate. The USA Today/CNN poll recorded a 10 point net gain for Bush, and the Washington Post/ABC poll had the bounce at 8 points. Leading up to the Republican Convention, which began on July 31, Bush/Cheney had an average lead of 7 points, and they left their convention on August 3 with a solid 12 point lead.
This momentum was cut short on August 8 when Al Gore stunned everybody by announcing Joe Lieberman as his running mate. This gave Gore a bounce, allowing him to cut Bush's lead to 6.5% leading up to the Democratic Convention, which began on August 14. As the convention was going on, it didn't appear as though Gore was receiving much benefit. On the last day of the convention, a daily tracking poll had Bush's lead at 5% but after Gore's acceptance speech that night, and perhaps more importantly, after the kiss, polls showed an impressive jump. In the days following the Democratic Convention, Gore was up an average of 4.5%, a net gain of 11%.
Gore was able to maintain a lead for the rest of August and much of September as a result of convention momentum and negative press for Bush, whether it was his ad that subliminally called Democrats "Rats" or calling a New York Times reporter an asshole when he thought the microphone was off. From September 1-19, Gore had an average lead of 6%. Bush went into damage control mode and with the help of charming appearances on Oprah and Regis and Kathie Lee, he turned the race into a statistical tie. From September 22 to October 2, Gore led by an average of 1%.
The first presidential debate was held on October 3 and despite the fact that Gore was widely considered the winner (by an average of 9%), his behavior at the debate coupled with media coverage that labeled him an exaggerator, denied Gore a bounce. The race was a statistical dead heat leading up to the second debate on October 11. This time, Bush was the clear winner (by an average of 9%) and did receive a bounce. Bush was leading by an average of 4% leading up to the third debate on October 17, of which there was no clear winner (people favored Gore slightly).
Throughout the rest of October, Bush held an average lead of 3.5%. Leading up to the weekend before election day, Bush was still up by an average of 2% and then came the November surprise: the report of Bush's DUI. A CBS Poll released on election day, 11/7, had Gore up by 1%, 45-44 (a net gain of 3% from their previous poll on 11/3). Similarly, Gallup's daily tracking poll showed a net gain of 4% for Gore between 11/3 and 11/7, even though their final poll still had Bush up by 2%, 48-46. And the Zogby/Reuters/MSNBC daily tracking poll had Gore leaping from down 3% on 11/3 to up 2% on 11/7, 48-46.
As we all know, the final (and ultimately irrelevant) popular vote total was Gore 48.4, Bush 47.9.
So what can we learn from this? Mostly stuff we already know but more reinforcement can never hurt:
1. Plenty of time for a major shift
In mid-September, Gore led from anywhere from 6% (Pew on 8/14) to 16% (Newsweek on 9/16) and ultimately won by less than 1%.
2. Debates matter
Prior to the first debate. NY Times/CBS News had Gore up 4%. After the final debate, Bush was up 2%. Prior to the first debate, Newsweek had Bush up 1%, after the final debate, Bush was up 7%. Prior to the first debate, Gallup had Gore up by 2%, after the final debate, Bush was up by 6%.
3. Gallup skews Republican
USA Today/Gallup began their daily tracking poll on September 4. How did their numbers compare with the averages of other polls for the same
periods?
Others Gallup
Sept 4-19 G +5.79 G +5.0
Sept 23-Oct 2 G +0.95 G + 0.64
Oct. 3-Oct. 11 B +0.17 B +0.89
Oct. 12-Oct. 31 B +3.78 B +5.3
Nov 1- Nov 3 B +2.0 B +5.0
Nov 4-Nov 7 B +0.67 B +3.25
Nov 7 G +1.0 B +2.0