Politisite Political Projections - North Carolina

Politisite Political Projections – North Carolina

Web Posted on May 5, 2008
 By Albert N. Milliron, Chief Political Correspondent,
Iron Mill Interactive Media, Inc / Politisite.com

Hello all you pundits, politicos, politstes, pollsters, politisites, political junkies, and casual political readers.  This is your place for North Carolina Poll data.  See also our PPP_ Indiana Hoosier Pick?

Check back Tuesday at 4:00 pm  ET   for exit polling data for Indiana and North Carolina. Then at 7:00 pm when the polls close we will bring you real time election results right from the election department of the Secretary of States offices from both states.

Politisite uses a Trend Line / Momentum Model in determining possible outcomes of elections. If you have ever traded in the stock market you are aware of these terms. Our model is based on the same ideas.

Trends & Momentum: Remember in statistics, one or two data points do not make up a trend. It takes more than 5 or 6 data points, in this case, polling data from pollsters. “The Trend is your Friend” is the title of a book on how to buy stocks by Peter Lynch. Mr. Lynch was portfolio manager of Fidelity Magellan, a mutual fund that had a 20% return per year over its life time. Lynch used a momentum/trending model by asking folks who read the book, “where do you shop”, “what products do you use”. His idea was that if you chose one store or product over another, other are probably making the same decision. Why do you like a certain candidate? What programs are they offering? If you like them for whatever reason, others probably have the same ideas.

I believe that anyone who does polling should go the extra step and look at trend lines and momentum. Had pollsters considered trending in New Hampshire, they probably would have had the numbers correct. Pollsters report data one or two days after it has been gathered. So what you get is data that is where a candidate was in the past. They do not consider where the candidates numbers may be. They don't deal in speculation, we do. Our model projects where a candidates numbers will be on the day of the election. Thats why we call our model a projection rather than a prediction.

Real Clear Politics has trend lines below the polls so one can see how the candidates are moving. You can also use these graphs to gauge where a candidate will be on election day. Why is momentum very powerful? Picture an snow ball rolling down a steep hill. The snow ball gets bigger and moves faster as it rolls down the hill. It is very difficult to stop a rolling train. Here is where physics comes in, what did Newton say?. “an object in motion tends to stay in motion with the same speed and in the same direction unless acted upon by an unbalanced force.” An object on our case is a political candidate. There is immediate resistance like the Rev. Wright situation in the Obama campaign. Gradient resistance which causes things to stop, like a rolling ball at the bottom of a flat hill. To look at gradient resistance take Mayor Guiliani. He was considered the front runner in the Republican party. He lost 3 contests then took a break to try to get momentum going again in Florida. The resistance was so large that he was unable to regain momentum. His big mistake was stopping his campaign through several primaries. Newtons first part of the law of motion is “ An object at rest tends to stay at rest” If you are stopped, it takes much more power to get one going again. This is why the almost dead McCain was able to pass him. McCain was stopped but won a big primary, South Carolina, this was enough power to get him started. Enough momentum to pass Guiliani in Florida which has a large retired population from New York, New Jersey, and Ohio. Guliani believed he would win due to that demographic. Get the idea? O.K. Now lets get started.

The Polls Trends  from North Carolina

SurveyUSA

http://www.politisite.com/images/surveyusa.jpg

Figure 1. . Obama = Blue Clinton = Red

Public Policy Polling

http://www.politisite.com/images/ppp.jpg

Figure 2. Obama = Blue Clinton = Red

Rasmussen

http://www.politisite.com/images/rasmussen.jpg

Figure 3. Rasmussen Obama = Blue Clinton = Red

Insider Advantage

http://www.politisite.com/images/insideradv.jpg

Figure 4 Insider Advantage Blue = Obama Red = Clinton

American Research Group

http://www.politisite.com/images/arg.jpg

Figure 5. America Research Group Obama = Blue Clinton = Red

 

The data shows Senator Clinton with a Clear lead in November 5 of 24 percentages points. A reversal took place at the beginning of the new year were Senator Obama and Senator Clinton were tied. Obama gains momentum after winning the the Iowa Caucus win on January 3. Senator Obama continues momentum to a 10 point gain after winning the South Carolina Primary. Obama has hovered in the 50 percentage range since February 11th. Senator Clinton lost momentum following the loss in Iowa. She wins New Hampshire which gave her a temporary boost until South Carolina. Her numbers are the only ones that have changed. Clinton has gained momentum following the Rev. Wright issues and her coming out with a gas tax summer holiday which Obama is opposing. Oil prices hit a new high on May 5th. Clinton has gained 5 percentage points over the last 28 days. Based upon these polls it appears that Obama number are steady while Clinton's numbers are rising steadily. Recently polling has Obama up by 8 percentage points. Remember in Pennsylvania we said we think Obama polls better Then outcomes and Clinton polls lower then outcome. The case where this conclusion was wrong was South Carolina. North Carolina has about half the African American population as South Carolina. There is about 8 percent of voters whoa are still undecided. We think that Clinton is gaining votes due to recent momentum at a rate of 1.5% to Obama 1%. .

Conclusion: Obama wins 53% to Clinton 46.5%    Obama +6.5

What Others are Saying

Zogby has Obama up by 14% (Zogby got the PA Number right)

Pollster.com

SurveyUSA

http://www.surveyusa.com/

24-Hours Till NC Votes are Counted, Obama Clings to Small Lead Over Clinton: On the eve of the North Carolina Democratic Primary, with 25% of votes already cast, Barack Obama has no breathing room in his hope to defeat Hillary Clinton in popular votes, according to SurveyUSA's 8th and final tracking poll, conducted exclusively for WTVD-TV, the ABC-owned TV station in Raleigh. On the final day of the fiercely fought campaign, it's Obama 50%, Clinton 45%. There is no foreseeable outcome in North Carolina, regardless of which candidate wins the popular vote, where one candidate collects significantly more Convention delegates than the other. Therefore, the exact final vote totals have much more symbolic importance than real importance. The popular vote is remarkably stable: In 6 SurveyUSA polls released since Super Tuesday, Obama has polled at 50%, 49%, 49%, 50%, 49%, 50%. The contest is stable among men, where Obama leads by 11. The contest is stable among women, where the two remain tied. In Charlotte and Western NC, there is the slightest momentum for Clinton. In the Research Triangle, in Southern NC and in Coastal NC, there is slight offsetting momentum to Obama. The net is a wash. If Obama wins, it will be entirely from the 19% of voters who describe themselves as Liberal. Clinton leads by 9 among Conservatives and leads by 8 among Moderates. If Obama wins the popular vote, it will be because of his 16-point advantage among Liberals. Clinton has increasing momentum among voters age 50 to 64, where she has gone from 30% in January to 51% today, her highest showing. Among those age 65+, Clinton leads by 20 points; the more seniors who vote, the better Clinton does. But there is offsetting momentum among younger voters, some of whom may be first-time voters, and not all of whom may show-up at the precinct. 1 in 4 of SurveyUSA's likely voters say they have already voted. Among those who say they have already voted, Obama leads by 16 points. Among those who say they will vote on Primary Day, Obama and Clinton are effectively tied.

Public Policy Polling