The bounce looks good.

Posted by: Commissar at September 4, 2004 03:01 PM

That is starting to look like the mother of all bounces! And it would seem that it took place over night after Bush’s speech, sept 2 to sept 3

Posted by: CERDIP at September 4, 2004 04:11 PM

Called a bounce for a reason.

Posted by: Lakhim at September 4, 2004 05:10 PM

The more people get to know Kerry, the worse he does. There is a reason he never talks about his senate career.

Posted by: Beemer Boy at September 4, 2004 05:46 PM

Wait, how does Bush’s convention help people get to know Kerry?

Posted by: Lakhim at September 4, 2004 05:55 PM

Lakhim,

Good question.

The strategic implications are based upon the apparent failure of the DNC to adequately define Kerry for most voters. Various polling organizations (and partisan talking heads or party activists opinions are not a reliable source) have stressed the bounce is made up of two major influences. (1) Define your party, platforms and nominee in the most flavorable terms
(2) Define your opponent and his platform in the most unfavorable or negative terms (but the impact is limited to the extent that the description is plausible and accepted, especially among non-commited, independent or wavering voters.)

An objective view of the relative bounces for Kerry (+2 to +4 on average within a week after the convention) versus the Bush bounce (so far its +3, +10 and +11 and counting (Gallup/USA Today/CNN comes out on Tuesday)) would seem to show that the Republicans have taken more advantage of this opportunity.

A number of the critical analyses of the DNC indicates that while the display of unity was apparent, the speeches very good (Kerry got high marks), it is the overall consensus that Kerry still had not offered up very much in the way of who and what he stands for. Yes, he tried and did speak to these, but the softness in the various character polling numbers seem to indicate that his attempts did not work all that well.

Which gets to your question, “How does Bush’s Convention get the people to know Kerry better?” Well, they may not know Kerry better nor know him for what the DNC would like them to know, but the Republicans did paint a picture of Kerry that they surely want the people to believe is the “real” Kerry. I would suspect that some of the drop in the Kerry polling numbers so far may be reflecting the Republicans’ ability to get more people to know their version of Kerry, and it is in turn turning off some of those polled. (A lot of the Kerry polled questions showed a drop and a corresponding increase in the “not sure”, “Nader/3rd party” or “neither” category.

Another danger sign from the Newsweek poll’s background numbers is found in that while among the different polled groups, Kerry’s support is highest (60+%) among non-white respondents. The bad news is that over one third of that group is either supporting or leaning to Bush. While I doubt that Bush will do better than one-third in this voter group, it may translate to disappointment among others in this bloc who would then sit out the election rather than vote for Bush. That makes it a double whammy if that scenario unfolds. Kerry would then have to spend precious resouces and time to try and remotivate a core voting bloc rather than try to increase non-committed voters.

Posted by: steve at September 4, 2004 09:33 PM

Lakhim,

You better get your DU friends to work on this unscientific poll over at AOL…

http://www.electionguide04.com/straw_poll.adp

Bush has every state but Wash. DC and 65% of the popular vote. But it’s not scientific. I’m sure it doesn’t reflect anything. Meaningless. AOL users are, um, normally more liberal than the internet in general, so I’m sure the results are somewhat biased to Kerry.

Posted by: Max Darkside at September 4, 2004 10:26 PM

Great explanation Steve…you must be either a statistician or you stayed in a Holiday Inn Express last night!

I’m seeing this as the second well known poll showing Prez Bush with a large bounce. Give me one more(Gallup or Fox/Opinion Dynamics) and I’ll be sold that Dubya is running away with it.

Posted by: Wayne Fielder at September 4, 2004 10:44 PM

Hmm…methinks something is amiss with the AOL electionguide thingy Max referred to. If you go to the second tab(National Tracking) on the thing it looks like bush jumped into the lead sometime in late june…then look at the two trends…they are mirror opposites to each other.

While I would absolutely LOVE to see this on the morning of November 3rd…I ain’t holding my breath. Yeah…Dubya is gonna carry 60% of California! 54% of New York!!

Posted by: Wayne Fielder at September 4, 2004 10:53 PM

Wayne, I’m sceptical of the AOL poll too. Just funny (both funny ha ha and funny odd).

Posted by: Max Darkside at September 4, 2004 10:57 PM

You forgot the post convention Leno poll:

Kerry took so many shots tonight he is elligible for two more Purple Hearts - Leno

i.e. Kerry’s Vietnam record is now a minus.

Posted by: M. Simon at September 4, 2004 11:26 PM

Another internal in Rasmussen also bodes poorly for Kerry. 25% of his voters think Bush will win. 90% of Republicans believe Bush will win. The ideological perception of the candidates hasn’t moved since the DNC. Kerry’s favorables were 50% compared with Bush’s 54%. Upon reflection, my analysis is the real reason for the bounce is not the perception that Kerry is liberal or unlikeable or unfit for command, but rather he is a loser. Remember, Kerry got the Democrat nomination for only one reason: he could beat Bush. Now that 52% believe Bush will win, just put a giant L on Kerry’s Botox-injected forehead.

Posted by: Rich Blinne at September 4, 2004 11:28 PM

Is it too soon to start talking about COATTAILS?

Reporters are desparately going to need a new story angle in a few days, “coattails” may be it.

Certainly, if things continue as they are, John Kerry is going to hurt Democratic politicians in close races across the country. The Party faithful will learn to hate the guy… and themselves for choosing him.

Then the following story angle for reporters will be Democratic candidates who distance themselves from Kerry.

Posted by: Loyola at September 5, 2004 01:15 AM

I think you are right Loyola…I sure hope so. Is it true that Daeschle is showing an ad with himself hugging the President? That’s the ultimate allegiance to a winner….to save your own skin!!

Posted by: dickmr at September 5, 2004 01:54 AM

That is starting to look like the mother of all bounces! And it would seem that it took place over night after Bush’s speech, sept 2 to sept 3.

Which makes it unreliable.

Posted by: Anthony at September 5, 2004 03:54 AM

Another internal in Rasmussen
Posted by: Rich Blinne at September 4, 2004 11:28 PM
******************************************
those internal polls have had me confused for sometime now even when Rasmussen had Kerry and Bush tied the situation did not make sense to me when I looked at other polls on the site such as

Who is a Better Leader? Bush 49 Kerry 38
Trust on Nat’l Defense? Bush 52 Kerry 42
Trust to Manage Economy? Bush 49 Kerry 45

Today with Rassnusen reporting
Presidential Tracking Bush 49 Kerry 45

they might make sense but they were about the same when Rassnusen was reporting

Bush Kerry
Sept 1 47 47
Aug 31 47 46
Aug 30 47 46

Posted by: Dan Kauffman at September 5, 2004 07:36 AM

Don’t get complacent! I don’t trust the polls.

Posted by: pinkpower at September 5, 2004 01:07 PM

More polls (This won’t be formatted right, but you get the idea):
Bush Kerry
American Research Group 47 48
The Economist 44 45
Leger 51 47
Newsweek 52 41
Rasmussen 49 45
Time 52 41
Zogby 46 44

Posted by: Lakhim at September 5, 2004 02:13 PM

Thanks, Wayne… I am neither a statistician now did my wife kick me out to stay at a Holiday Inn Express. Just a engineer, golfer and former helicopter pilot who likes to look at various pieces of a puzzle and then try to see what the overall mosaic presents.

There are some other good posts from you all as well… I enjoy the various analyses and comments on what if anything significant these mean.

Thanks, too Lakhim for the other polls… the Economist’s poll I would think would bring in some foreign perspectives, but I’ve not looked intyo the trends on that one. Have they been showing a similar Kerry to Bush switch as well?

I was going to offer up the most recent Rasmussen poll numbers, but Rich beat me to it. I did find out, though, that Rasmussen has a weekly tracking poll in addition to their daily poll. The weekly allows for a more comparative set of numbers to the other polls which track on a monthly or bi-weekly basis. The weekly shows Bush up by 2 to 3 points as well.

Rasmussen is now showing the President leading for the first time in all 16 battleground states, which is equally significant.

Finally… I try not to get too dependent with the polls as they are after all reflecting was the sense of respondents three to five days earlier. Anything can happen to cause a major shift now, especially after the RNC demonstrated that there is more volitility in these polls than previously thought.

Posted by: steve at September 5, 2004 04:07 PM

More Rasmussen stuff: Bush is tied in Minnesota and within single digits in California. One thing that speaks to a Time/Newsweek fluke is Rasmussen went down a little bit today when it had more post acceptance speech samples. Now this could be the weekend effect which tends to skew towards Democrats. (This effect did in Zagby before the 2000 Presidentional election.) Even though Rasmussen shows a significantly less bounce, they now say it is Bush’s to lose and the status quo works for him.

Waiting for Tuesday…

Posted by: Rich Blinne at September 5, 2004 08:41 PM

Even though I’m voting for Bush come November….I believe the bounce he got may decline to bout 5% points over Kerry….by election time….

I have to say the RNC convention was a big plus…lot of energy there…where as the DNC was rather boring…yes I watch both….I had no life during those 2 weeks..lol…

also regarding the polls..gotta remember the first one that set off alarm bells at the Kerry camp…LA Times showing bush ahead by bout 4 points overall and a sharp increase over Kerry in specific areas…I even read a few replies on some blogs that the LA times is considered Right-wing now…

I mean geez….to use the old tired line we all have read…if a poll comes out in Kerry’s favor its a significant sign against Bush…and if Bush is ahead…its unreliable?

frankly the truest indicator (poll wise) may be in the weeks ahead more than anything…but they are still polls…in the end only our vote counts

Posted by: terlizzi999 at September 6, 2004 01:32 AM

Please note that this is what Rassmussen has to say since the convention ended:

Election 2004
Date Bush Kerry
Today 47.6 46.4
Sept 4 49.1 44.7
Sept 3 49 45
Sept 2 49 45
Sept 1 47 47

ARG has Kerry AHEAD among registered voters, and that counts better with large turnouts.

The bounce was temporary and shallow. Kerry will be ahead again by the end of the week.

Posted by: ericl at September 6, 2004 08:33 AM

In the same Rasmussen report, the following statement is also found:

“Data released yesterday shows that President Bush has taken the lead in 16-Battleground States and that most Americans now expect him to be re-elected.”

This is a significant data point from the Rasmussen report that was sited to show a decreasing trend apparently in the President’s polls. Above all other polls, the Rasmussen has been showing a very tight race consistently. The Rasmussen polls tend to show less movement and more stability than other polls. When some of the polls were reporting the Kerry post convention bounce at 5 to 7 points, the Rasmussen was only indicating a bounce of about 2 points most of that time.

The Kerry lead did fall some within a week of the convention. What I find interesting in the latest Rasmussen analysis was the absence of any comment about the race still being tight… this has been usually a closing assessment in these reports up until now.

Before classifying the Bush bounce as “shallow” one must wait about a week to see how the trend continues. However, Rasmussen noted that the President’s bounce numbers were somewhat higher and started just prior to the convention and stayed at the higher numbers longer:

“Over the past nine days, Bush has been ahead eight times and the candidates have been tied once. Senator Kerry has not been ahead in the Tracking Poll since August 23.”

Having been following the Rasmussen polls for some time, this trend is significant in that no candidate has yet to post the type of lead that President Bush has maintained since the Democratic primaries ended. As mentioned by others who have posted comments, it is early in the last run up, but the degree of the bounce and the amount of the lead among all voter groups has to be troubling to the Kerry camp. They have their work cut out for them for sure… much more than they had hoped to see.

Posted by: steve at September 6, 2004 09:44 AM

Oops. Looks like the Time/Newsweek double digit bounces are wrong, improperly weighted:
Sample:
Party ID Bush Kerry Und/Oth
Republican 38% 94 4 2
Democrat 31% 14 82 4
Independent 31% 45 40 15
Total 100% 52 41 7

However, if the sample was properly weighted, here are what the results would look like:

Party ID Bush Kerry Und/Oth
Republican 29% 94 4 2
Democrat 33% 14 82 4
Independent 38% 45 40 15
Total 100% 49.0 43.4 7.6

So its more of a 3-5 pt bounce from Time and Newsweek then a 10-12 pt bounce. Sorry.

Posted by: Lakhim at September 6, 2004 11:24 AM

Scott Rassmussen will say at 3:00 that his poll results are too narrow. There was a rogue sample (his words) on Saturday that made it closer. He sees a 3-5% lead for the President despite the 1% difference currently on his web site. He also says that the Newsweek and Time polls oversampled Republicans. When you correct for that they predict a 3-6% lead for the President (confirming Lakhim’s analysis).

Another good piece of data from the deep internals of the Rasmussen poll:

1. It is a trend, not a bounce. Starting from August 24, Bush has slowly built a 3% lead in a monotonically increasing fashion (except for the Saturday sample, today’s sample increased again).

2. It is not coming from “swing” voters. Comparing September 4 7-day-moving average with August 25, Bush gets a 3.1% spread from a 0.1% spread. But, it shows up all in the “certain voters” column, going from 40.6% to 43.5%! BTW, Kerry has a 39.5% certain voters column. This shows that Rove and company were spot on in rallying their base at the RNC.

Posted by: Rich Blinne at September 6, 2004 01:05 PM

I checked the historic data and I found no similar hard base support following the DNC. Kerry seems to be capped at 40%.

Posted by: Rich Blinne at September 6, 2004 01:07 PM

Bush also seems to be developing some coat tails. The generic congressional ballot now shows only a 1.5% Democrat lead from a 6-7% lead (3 day moving average). So, the “bounce” is not merely for the President but for Republicans in general. Republicans tend to do better than the generic ballot so this is very bad news for Democrats and may be the reason why Clinton got involved. It may be OK for Kerry to lose but he wasn’t supposed to take down the party with him.

Posted by: Rich Blinne at September 6, 2004 01:58 PM

More proof of trend rather than bounce. 16 battleground state numbers from 29 Aug. to 6 Sep. (7 day rolling average)

Bush
43.6 43.6 43.8 44.8 45.5 46.6 46.9 47.1 47.2

Kerry
48.2 48.5 48.2 48.1 47.8 47.5 47.1 46.6 46.4

Bush Favorable
48.9 49.3 50.3 50.4 51.0 51.3 51.9 52.2 52.1

Kerry Favorable
53.1 52.9 52.7 51.9 51.4 51.0 51.7 51.1 51.6

Right Track
35.9 35.6 36.2 36.0 37.6 39.7 41.1 41.4 42.0

Wrong Track
59.3 59.4 60.1 60.5 59.2 56.9 55.5 54.9 54.5

No bounce, just slow steady progress.

Posted by: Rich Blinne at September 6, 2004 04:11 PM