July 13, 2004
THE FIRST BOUNCE?
Bush 48.5% – 283 EV | Kerry 49.5% – 255 EV
EV without Toss Up states (under 2% margin): Bush 222 | Kerry 234 | Toss 82
Pundits have long pointed to the Veep selection as the time of the first real bounce in this relatively stable race. Edwards was selected on July 6, and we now have a full week of polling on the national level and state level (though many of the new state polls are from before the announcement). Was there a bounce?
Yes, kinda. Most polls show an increased lead for Kerry, but generally so small and within the range of past fluctuations in polling results over recent months. The Composite Poll shows a net gain for Kerry in one week of 2.7%. His biggest one week net gain was in the May 18 Composite Poll when he picked up 4.1% against Bush. That week was also Kerry's (or either candidate's) biggest lead since I began this analysis on March 9. Kerry led by 2.7%.
Since June 29, Kerry has gained almost 3.5% in the Composite Poll. But he continues to lose points in the Electoral College. We'll get to those in a minute.
The best way to see if Kerry's selection of Edwards gave him a real bounce is to look at polls of particular polling organizations and see how they have moved. For this reason, we have not used the one day "quickie" polls conducted by NBC, Time and CBS, each of which used a different pollster, had a high margin of error, and presumably had different polling methodology than earlier polls commissioned by the same news organizations.
Kerry/Edwards bounced a total of 6 points in the CNN/USA Today/Gallup poll, from being down by 1 on June 23, to up by 5 on July 11. But this represents little change from the June 6 Gallup poll, when Kerry also led by 6. So this could be a statistical fluctuation and not real movement.
Zogby shows no net bounce for Kerry. On June 5, he had the race a two point lead for Kerry. On July 7, a two point lead. Newsweek now gives Kerry a 3 point lead. On May 14 he only led by 1, but on April 9, Newsweek had Kerry up by 4, so he's down from that high.
Rasmussen also shows a statistically insignificant increase for Kerry (less than 1%).
But everyone, except for the AP/Ipsos poll (Bush by 4 on July 7 and by 1 on June 9), shows an increased margin for Kerry over immediatley preceding polling. And the Composite shows a bump. So, I think it's safe to say there really is a bounce here. It just isn't significant, but perhaps we shouldn't expect more with such a low number of potential voters reporting they are undecided each week.
So, what's happening with state polling. Well, state polling lags the national poll, so if this is a real bounce for Kerry I suspect we'll start to see it at the state level next week. Many of the new state polls in this week's analysis were Rasmussen polls, taken throughout the month of June. Thus, they reflect the race not as it stands on June 30, but they combine results obtained throughout the month. If Bush had a substantial lead in the first three weeks of June, but lost support during the last week, the Rasmussen month long polls will have a hard time showing that.
This week incorporates new polls in 27 states. Highlights include Arkansas (Bush +2 and Kerry +1), Illinois (Kerry +16), Iowa (Kerry +4), Maine (Kerry +1), Michigan (Kerry +2, +7 or +3), Minnesota (Kerry +9), Missouri (Bush +3), New Jersey (Kerry +10), North Carolina - post Edwards selection (Bush +15), Ohio (Bush +4), Pennsylvania (Kerry +5), Virginia (Bush +4), Washington (Kerry +9).
A note about Florida: Rasmussen's Florida poll gives Kerry a 5 point lead, while SurveyUSA gives him a 3 point lead. Because my analysis is designed to show trends and not necessarily today's snapshot through a weighted average of state polls (giving more weight to newer ones, but some weight to older ones), I still don't have Florida in the Kerry column. Why? Two Florida polls that are only a couple of weeks old showed Bush leading by 7 and 10 points. In this analysis, Florida is trending for Kerry, it just needs another poll to validate the two latest polls to push it over to him. Or another poll favorable for Bush to validate the June polls.
Now some data:
This Week’s Polls
AP/Ipsos (Bush 49 | Kerry 45)
CNN/USAToday/Gallup (Bush 45 | Kerry 50)
Zogby (Bush 45 | Kerry 47)
Newsweek (Bush 44 | Kerry 47)
Rasmussen Week Average (Bush 45.7 | Kerry 47)
Iowa Electronic Vote Share Market (Bush 50.9 | 47.5)
Last Week’s Polls
Fox News/Opinion Dynamics (Bush 47 | Kerry 40)
CBS (Bush 43 | Kerry 42)
CNN/USAToday/Gallup (Bush 48 | Kerry 47)
Rasmussen Week Average (Bush 45.4 | Kerry 45.9)
Iowa Electronic Vote Share Market (Bush 53.0 | Kerry 46.1)
Bush State Gains
Ohio
Kerry State Gains
New Mexico
Oregon
Toss Ups (slight lead of less than 2%)
Florida *NEW* (Bush)
New Hampshire (Bush)
Ohio (Bush)
Pennsylvania (Kerry)
Wisconsin (Bush)
Full data, maps and graphs are updatedon the Composite Poll page.
