Late Deciders and May 6th

by: fladem

Fri May 02, 2008 at 17:19


About a month ago I wrote here about a pattern among late deciders that the exit polls were showing.  As I noted then, Clinton had consistenly won among those who decided in the last 3 days and in the last 24 hours.   Conversely, Obama had consistenly won among those who decided in the last week.  Here is the data updated to show that Pennsylvania continued the pattern:

However, it is critical to note that this relationship DOES NOT HOLD FOR STATES BELOW THE MASON DIXON as this table shows:

Pay particular attention to the average for the last 72 hours.  

All of this is of obvious significance given the primaries in Indiana and North Carolina.  As I will show on the flip, the last 4 days differ signficantly from any pre-primary week in this cycle.   Before this week, Obama has tended to close on Clinton in polling.   But this week in both North Carolina and Indiana it has been Clinton who has been gaining....
fladem :: Late Deciders and May 6th
The tables at the bottom of this article show how the last two weeks have fluctuated in this cycle.  Again, note that Obama has consistently outperformed his numbers in Southern States, while Clinton has outperformed her polling outside of the Mason Dixon.

Shortly after the South Carolina primary, I calculated that Obama had received a 13 point bounce as a result of his blowout victory.  Clinton's bounce in the national polls seems of a similar magnitude:

Pollster Margin Pre-PA Current Margin Swing
National Average 9.7 -1.5 11.2
Gallup 10 -2 12
Rasmussen 8 -2 10

Of course, there is no national primary.  Here is the trend in Indiana (also showing polls that have prior results):
Pollster Margin Pre-PA Current Margin Swing
Polling Average 2 -6 -8
Downs Center 5 -7 -12
ARG -5 -9 -4

Clinton's lead of 6 points heading into the weekend is larger than her lead at the same time in Ohio (was 2.8, won by 10) and Pennsylvania (lead by 4.7, won by 8.6).  Given what we know from exit polling about how states like Indiana have finished, it would be reasonable to expect a Clinton win of 10 to 15 points.  However, in Indiana Clinton has already picked up 6 points.  It will be interesting to see if she can maintain the momentum.

North Carolina, as I have shown, is a very different case.  First, here is the summary:

Pollster Margin Pre-PA Post-PA Recent Swing
Polling Average 15.5 12 8.2 -8
ARG 10 11 1
Rasmussen 23 14 9 15
IA 15 -2 5 -10
PPP 20 12 -8
SUSA 10 5 -5

The final weekend in North Carolina will be dramatic.  There is a well set tendency for Obama to outperform in Southern States.  Obama's 8.2 point lead is, at this point, smaller than he had in South Carolina (led by 15 going into the weekend, won by 25), Virginia (led by 17, won by 29) but larger than his lead going into Alabama (trailed by 3, won by 14).  

However, Clinton is clearly gaining both Nationally and in North Carolina.  The stage is set for her to get within 5, and doing so while at the same time winning Indiana may be enough to seriously damage Obama.  But this very expectation may be enough to effectively end this race.   The evidence has consistently shown that Obama outperforms in Southern States, and if that pattern holds he should win by 10 or more.

Polling Results Averages:


Tags: , (All Tags)
Print Friendly View Send As Email
Late Deciders and May 6th | 12 comments
Doesn't this indicate a huge win for Clinton in Indiana? (0.00 / 0)
You write: "Given what we know from exit polling about how states like Indiana have finished, it would be reasonable to expect a Clinton win of 10 to 15 points.  However, in Indiana Clinton has already picked up 6 points.  It will be interesting to see if she can maintain the momentum."

Are the last two sentences meant to suggest that Clinton's picking up six points might somehow mean she will not pick up as many points as usual in the last few days? I don't see why that follows.  Couldn't one as readily conclude she will pick up even more in the remaining days?

Yes (0.00 / 0)
that's fair.  The question I have is whether the votes she picked up this week are ones she usually wins in the last 24 hours.  

I don't know the answer.  

[ Parent ]
Great post, but MD is below the Mason-Dixon line, (0.00 / 0)
not above it.
Culturally (0.00 / 0)
though, I don't think it is.  Maryland votes more  New Jersey than North Carolina.  
[ Parent ]
I'd like to have some theory (0.00 / 0)
why this is true before putting too much faith in this pattern.
I keep asking about this pattern (0.00 / 0)
I keep asking people to come up with theories why Obama would do well in the last week, but badly in the last few days.  I wondered whether there were some kind of last-minute mailings going out, that weren't being reported.

It's even more mysterious to me if the pattern truly only exists for certain regions.

Pennsylvania (4.00 / 1)
Pennsylvania voters who decided in the last week went for Clinton, 55-45, right?  

Conversely, Obama had consistenly won among those who decided in the last week.  Here is the data updated to show that Pennsylvania continued the pattern:

Unless I'm missing something (very possible!) I don't understand the statement above.

Identity politics? (4.00 / 1)
The clearest explanation of this voting pattern is identity politics.  The states in the south have a higher percentage of blacks in the population.  Mississippi at 37% has the highest proportion of blacks of any US state; Georgia has the highest number of black residents.

FWIW, 21.7% of North Carolina's population is black and, according to the state's voter statistics, 38% of the state's registered Democrats are black (NC has roughly 20 times as many black Democrats as black Republicans).  That's a lower total than most other southern states but a higher total than the northern states. AL (26.3%), MD (30%), MS (37.1%), LA (31.7%), Sc (29.0%) all have a higher percentage of blacks than NC.  The most similar state is VA (19.9%).  That would indicate a small Obama bounce in North Carolina.

Very odd (0.00 / 0)
that the phenomenon of "momentum" seems to be returning to the race after having been seemingly absent ever since February 5th. Couldn't have happened at a better time... :S
Questions re numbers and theory (4.00 / 2)
1.  You need to clarify what the numbers refer to. You don't seem to put the base referent in ...it takes a moment to realize the when doing the pollster shifts you are using Obama as the base.  So with these other numbers there are a bunch of possible meanings....whichneed to be clarified.  

Example The last week in Penn..  what is 45 for Obama, 55 for Clinton and what does the percentage of 6 mean.  I want to make sure I really understand your numbers.

2.  I think you may be conflating the time in which all these southern states voted with how southern states will vote. That is these states voted when Obama was on an upswing in terms of national polls and winning. He was proving to voters ...both black and white ...that he could win and that synergistically drove up his vote in these soutern states. I think in most of them African Americans turned out in greater than ever expected numbers and drove up his poll numbers.  

3.  I am not sure that this is the same kind of time for him in terms of voting.  Pennsylvania did change the race...psychologically in terms of media and voter perceptions.  Two.    Rev Wright came back like an expected bad penny which will affect lots of the white voters he attracted in the earlier batch of southern states. Also it is not clear to me that the Rev Wright controversy will not have some kind of negative effect on black turnout...whether out of some kind of paralysis or something else  ....it may energize the black community or enervate them.  He is also taking them for granted as he courts white voters.

You could be right or wrong...but I think there is some reason to think that the Southen pattern you found will not hold in the post Pa primary/Rev Wright/"elitist" environment.

"Incrementalism isn't a different path to the same place, it could be a different path to a different place"
Stoller

NC happens to be between (4.00 / 1)
SC and VA...the only two states where exits polls show late goers going for Obama
She wins last 3 days: DEBATE (0.00 / 0)
She wins voters every time she debates.

She has always arranged that in the last couple days before each vote, to be in a situation where she shines on policy: in a debate or, if she can't get Obama to agree to one, a similar forum for her to show her superior skill and detailed understanding of how good policy can help us.

Obama is better in inspirational speech, and that appeals to other types of voters. But where she is the stronger candidate is in one on one conversational discussion of policy.

Right before SuperTuesday: Kodack Theatre debate
Right before Ohio/Texas: Ohio debate, (+Daily Show)
Right before PA: (second hour of) ABC debate(otherwise silly)
Right before Indiana/NC: (Obama spurned the Lincoln/Douglass flatbed truck debate spurned by Obama, so) so she will take town hall questions on Sunday.

Obama is not being as smart schedualing a MTP interview opposite her as he is a bit boring in one on ones. His fans like to see their own great hordes get in line validating their judgement.

And also, she got a chance to show this expertise on O'Reilly/Faux.

While not in the last 3 days, but the last 6, but the ratings for that episode jumped through the roof: 6 times his audience, so even though its Faux, she was able to reach potential voters and make clear to them that she knows America worked with 90% top tax rates in the 40's and 50's. O'Reilly's audience remembers those good old days but this good government policy is news to them having been so propogandised by the VRWC.

She shamed the YOYO party into agreeing that the rich should pay higher taxes. She did this by owning up to being(now)among those rich who should pay higher taxes, and making her equally rich O'Reilly squirm and admit he too should pay higher taxes.

Since that selfishness is the essence of what Republicans sell to their YOYO's, this was a terific GE battle too.

She made it morally wrong for the rich to continue to evade reponsibility.

John McCain vetoes every Environmental Bill already.

Late Deciders and May 6th | 12 comments
Donate to Open Left







QUICK HITS

Friends of the Earth thanks the OpenLeft community for the ideas you generate and your contributions to the progressive movement.
As an anti-spam measure, there is a 24-hour waiting period after registering before new users can comment.
blog advertising is good for you
blog advertising is good for you
LIBERAL INSTITUTIONS
STATE BLOGS
SEARCH

   

Advanced Search

Home | RSS/XML Feed | Register | Contact

  © OpenLeft 2007-2009, Powered by SoapBlox