30 August 2008

The DEMOCRATIC NATIONAL CONVENTION - in retrospect

The DEM convention in retrospect.

 

For just one blog entry, I would like to step mostly away from statistics and do a personal assessment of what has just happened this week.  Sometimes it does good to step back and breath and just assess, both objectively and emotionally.

 

The week from 08/25-08/29 has been historic in ways I never thought I would see within my lifetime. I wish to elaborate here:

 

1.) A major party, the DEMOCRATIC party, has just nominated a black man to be it’s candidate. The fact that a black man has achieved, against all odds, a nomination, regardless of which party, is cause to celebration. For me, it means that we as a nation have grown. For me, it also means double so much, when we consider that the DEMOCRATIC party was the party of slavery just roughly 150 years ago and that thereafter the DEMOCRATIC party was not willing to let black voices be heard in the south. The entire saga of the civil rights act in the 1960s, including the massive defection of former democrats in the deep south to Barry Goldwater in 1964 once the DEMOCRATIC party latched on to the idea of civil rights, proves this point in a stark way.This shows an amazing turnaround in the DEMOCRATIC party worth noting.  Even republican pundits critics, among them even Pat Buchanan, express respect for Barack Obama.

 

2.) A major party, the DEMOCRATIC party, was on the verge of nominating a woman for the presidency. That the party was so evenly split between both Hillary Clinton and Barack Obama gave the GOP a lot of ammunition to shoot with, but in spite of this, the presumptive GOP nominee’s rating have stuck in the mud in the low to mid 40s the entire time. Rumours of a divided convention were just that: rumours.

 

3.) In an honest gesture of extreme grace, Clinton released her delegates, both she and her husband, former President Bill Clinton, gave Obama ringing endorsements and will be campaigning with him.

 

4.) Obama’s acceptance speech at the Ivesco arena, in front of 84,000 participants within the arena, with thousands more outside the arena, where the line for getting into the stadium stretched for over 6 miles, was watched by at least 38 million american viewers, exluding those who watched through C-Span and excluding those who watched from abroad (such as I), so the more accurate figure must be over 40 million american viewers.

 

“The audience estimate of 38.3 million means that Mr. Obama’s speech reached more viewers than the Olympics opening ceremony in Beijing, the final “American Idol” or the Academy Awards this year, the Associated Press notes.

Furthermore, the four-night Democratic convention ranks as the most-watched convention of either party, Democratic or Republican, since Nielsen began measuring conventions in 1960.“ –NYT

 

5.) John McCain from the REPUBLICAN party  has picked Alaska Gov. Sarah Palin to be his running mate. This is the first time that the REPUBLICAN party will have a woman on the ticket. There will be much discussion about this pick (including the timing and the location for the pick) and the press will be vetting her extensively, but all sides agree that this interesting pick brings many pros and many cons with it. More on this next week.

 

But we now know what both tickets will look like:

 

Barack Obama / Joe Biden  (D)

 

-vs-

 

John McCain / Sarah Palin (R)

 

6.) This is the first time since 1952 that both conventions have been held so late in the season and the first time ever that both conventions have been held back-to-back. Which will make for some statistical mayhem, but at the end of the day it doesn’t matter as the electoral trends will show themselves quite clearly between the debates.

 

7.) CONVENTION BOUNCES:

A number of sources have written much over convention bounces. Here is the GALLUP take on the issue, that the bounce is usually about 5 points. But 538 also has a fascinating, worth-the-reading write up on the whole issue of bounces. The McCain team put out a press release before the DEM convention, saying that it expected Obama to get a 15% bounce. You can find my statistically driven response to this type of propaganda here. Here a quote from the end of my write-up:

 

FACIT: looking that the averages, it would be more logical and fair to assume that an Obama bounce will be between 6-8% and the McCain bounce would be between 4-6%, if statistics are our only guide. However, with race a factor in this election plus a disgruntled core of Clinton supporters, plus the war in Iraq plus the price of gas plus the housing market crisis, plus plus plus…., these statistics may all just fly in the wind of history. If Obama pulls off a 15% jump, then I would be amazed, for I am of the personal opinion that the nation is far too divided among political trench lines for this to happen. Add to that the fact that this is the longest campaign in American political history, there is indeed the possibility that voter fatigue may already be setting in on all sides. Ditto McCain: I doubt he will have a massive bounce, either, and for exactly the same reasons.”

 

To be fair, the Obama team also sent out it’s propaganda, saying that it expected the convention to bring no bounce with it, which is just as ridiculous.

 

So, now let’s look at the national polling numbers thus far:

 

Pollster

Date

Obama

McCain

Margin

Margin Comparison to day before

Gallup*

08/24

45

45

0

--

 

08/25

45

45

0

0

 

08/26

44

46

-2

-2

 

08/27

45

44

+1

+3

 

08/28

48

42

+6

+3

 

08/29

49

41

+8

+2

 

08/30

 

 

 

 

 

 

 

 

 

 

Rasmussen**

08/24

48

45

3

--

 

08/25

48

45

3

0

 

08/26

46

46

0

-3

 

08/27

46

47

-1

-1

 

08/28

47

47

0

+1

 

08/29

49

45

+4

+4

 

08/30

 

 

 

 

 

*Gallup does three day tracking. For instance, the figures published on 08/29 were for the time frame from Aug 26-Aug 28, all interviews of which were conducted before Obama’s acceptance speech on 08/28 and most of which were conducted before Bill Clinton’s endorsement and Joe Biden’s acceptance speech on 08/27. Which means that the numbers for 08/30, when they come out, will reflect 08/27-08/29, but the interviews on 08/28 will still be BEFORE Obama’s acceptance speech. First the date from 08/28-08/30 will start to give us a clear picture of an Obama bounce.

 

**Rasmussen’s daily tracking results are collected via telephone surveys of 1,000 likely voters per night and reported on a three-day rolling average basis This is a marked difference to Gallup, which holds it’s polls in the daytime.. The margin of sampling error—for the full sample of 3,000 Likely Voters--is +/- 2 percentage points with a 95% level of confidence. The poll  update from 08/29  was  based upon interviews conducted Tuesday, Wednesday, and Thursday night. Virtually all of the interviews were completed before Obama’s acceptance speech last night.

 

Currently, according to Gallup, Obama has gone from 45% to 49%, a 4% percentage bounce and an 8% margin bounce. According to Rasmussen, Obama has gone from 48% to 49%, a 1% percentage bounce and a 1% margin bounce, but as indicated, both pollsters have not yet produced data that came after Obama’s speech. Gallup attributes much of Obama’s improvement to-date on the return of conservative democrats (read that as: Hillary democrats) to the DEM fold. From one poll update to the next, his catch of the conservative democrats jumped from 63% to 77%, a 14% jump.

 

 

Were I to average the current margins together, then we would have a +6% margin for Obama, which is within my prediction. However, since the public opinion on his acceptance speech and McCain’s pick of Palin has not really been measured yet, the numbers from this weekend will tell us much about the DEM bounce and the coming GOP bounce. I suspect that Obama will jump over a 10 point lead (this is easy, he only need to gain by 2 and McCain only needs to lose 1 point, and we suddenly have an 11 point lead for Obama), which McCain will temper down to a 6 or 7 point lead again, unless of course Palin steals the show among women voters.

 

Which is why John McCain timed his VP pick announcement for Friday at noon EST. He wants to quash as much DEM convention bounce as possible, and I cannot blame him for it.

 

Without a doubt, history has been written this week. On both sides of the aisle. WOW.

26 August 2008

DEM convention - Day 1 - an assessment

I watched the most important moments of DAY 1 of the democratic national convention, including Kennedy's speech and of course, Michelle Obama's speech.

Some personal, non-statistical comments:

First, it looks like a slick, well organized convention, ala Bill Clinton 1992.

Second, I think that Kennedy brought the house down because many think that this will be his last speech. So, love him or hate him, he is a respected icon within the democratic party and a voice that still carries weight.

Third, like her or hate her, Michelle is a sincere person and you can read this immediately in her body language. No doubt she surpassed the expectations of the democratic party. I am certain that the right will be railing on Michelle Obama all day long today, because they are scared that she will now catch on with the public. And I think she will. She hardly looks or talks or gives bodily signals like a "black nationalist", as the right is trying to brand the Obama's, which in my opinion is nothing more than badly cloaked racism. Just remember: the side that is the nastier is usually the losing side. I expect that many in the GOP are going to give their party a very bad name by going over the line about Michelle Obama. All the past indications tell me that the GOP is anything other than a class act this year. Which is ok with me, for the deeper they dig their own hole, the farther down they will fall.

I also doubt that the republicans will think to do something similar with Cindy McCain at the republican national convention next week. First, being the woman who helped John McCain engage in adultery in the late 1970's and early 1980's, she can hardly make case for "moral values" in the GOP. Second, she is stiff as a brick and I doubt that the GOP wants to appear stiff. Third, enough people now know that McCain, in a fit of rage (oh, just imagine that red button in the White House), called her a c**t in front of witnesses in 1992 and I doubt that the McCain campaign wants to remind the public of these unseemly details. I cannot imagine one single woman in the USA wanting to vote for this man.

There are already comments swirling around that the DEMS did not attack the GOP enough on Day 1, but if you look at past precedent, the steam for a convention usually builds on Day 2, so Day 1 was just fine.

What will be interesting to see will be how the Clinton supporters react tonight. I do think that tonight will be, in many ways, the most important night of the campaign for the democrats, for tonight we will not only hear Hillary Clinton speak, but will also see her body language up close. Of course, the right is raging about how the Clinton's are stealing the convention, but I think that both Hillary and Bill speaking is not only appropriate, but is also their prerogative for this party. and it shows how magnanimous Obama is. And then the right brings out these commercials about Obama not picking Hillary as his VP candidate. As if anyone thinks that the GOP would care about the Clintons. If I remember, a McCain supporter called her a "bitch" just months ago and he laughed at the comment. So, I doubt DEMS are buying this line. You watch: tonight, Hillary will give a resounding speech that will bring the house down, and not only among her supporters, but also among the Obama supporters. And Bill will also use his extraordinary powers of persuasion to rally more than just the base. And I just want to remind, I started as a Clinton fan and still like Hillary. And my gut tells me that the vast majority of DEMS think somewhat as I do - with common sense.

For this reason, it is my opinon that all this talk about dis-unity within the democratic party is about to take an end. And then it will be time for the so-called PUMA people to grow up and stop behaving like spoiled brats. As one delegate put it: "how many times must Barack Obama apologize for having won?" People who think they can hijack a party earn no respect from me.

and don't forget: the Paulbots will also be at the republican convention next week. So, the saga is not over with. Republicans who are laughing right now may not be laughing very much next week. Paul and Barr can do immense damage to McCain and the Perot saga of 1992 tells us that this can also happen late in a campaign.

-------------

About Denver: a little trivia. CNN had a small headline about the last DEM convention in Denver, which was in 1908, where William Jennings Bryan was nominated. To note, in spite of losing in a landslide to Taft, Bryan narrowly carried Colorado with a 1.12% margin. Now, Obama is not Bryan. He is better than Bryan, imo. And 1908 is not 2008, for I think Obama will win and he will carry CO - which is why I find the information interesting :)

25 August 2008

15%??

The McCain campaign has already put out it’s take on the expected convention bounce for Barack Obama. They think he will jump 15 points, comparing his bounce with Bill Clinton’s from 1992.

 

This is what the McCain team has published:

 

“Monday marks the beginning of the Democratic National Convention in Denver. A combination of factors makes this particular convention historic on many levels. Democrats have just completed an incredibly compelling primary cycle that has both energized and divided the Democratic Party. Because of the unique nature of the Democratic primary, we believe Obama will receive a significant bump from his convention…This cycle mirrors Bill Clinton’s Democratic convention in 1992: A historic 16-point bump. Barack Obama is more similarly situated to Bill Clinton in 1992 than any other candidate in recent history. Bill Clinton was a new candidate on the national scene; he was running in a “change” oriented election cycle and the economy was voters’ top issue — a dynamic he was able to capitalize on. He received a 16-point bump coming out of his convention. .. Obama’s stadium address on Thursday — the 45th anniversary of Rev. Martin Luther King Jr.’s “I Have A Dream” speech — will result in effusive and overwhelming press coverage. On Thursday, Obama will give a great speech, as has been his trademark. The press will sing his praises and remark on his historic address and Obama’s place in history. For example, The Associated Press today published an article comparing the historic nature of the addresses - a week before Obama’s speech. This coverage will be impenetrable and will undoubtedly impact the polls.”

 

I call this “propaganda” for I am sure that the McCain camp has cited a figure of 15% bounce, knowing that it will probably not be so, so that they will have given themselves political ammunition. Also knowing that if he makes 15%, they won’t look like they have egg on their face. But if he gets 20%, then the McCain team has a real problem on it’s hand.

 

I doubt that Obama will get such a bounce. Statistically speaking, this is highly unlikely.

 

Gallup has put out a study of convention bounces going back in history to 1964, including a chart of the numbers with the democrats and the republicans side by side.

 

Gallup dealt with the famous Clinton bounce of 1992 (16%) and reminded that Ross Perot pulled out of the race during the democratic convention that year, but then gave the caveat that “the 16-point bounce is based on a comparison of Bush vs. Clinton trial heats before and after the Democratic Convention“. Nonetheless, there can be no doubt that Perot made a difference here, for the simple reason that he jumped back in and took 18.91% of the PV on election night 1992, reducing Clinton’s 16 point lead down to less than 3 points. Not only that, they openly admit to comparing trial heats without Perot in the mix BEFORE the convention, but did not publish their 3-way figures (Bush-Clinton-Perot). And the Clinton bounce is the clear outlier in these statistics: it is more than double the DEM bounce average and more then triple the GOP bounce average.

 

So, I decided to first look at the numbers a little more closely. Here the gallup figures, as a quick table (open elections are in italics, the winning candidates are in bold):

 

YEAR

DEM CANDIDATE

BOUNCE

GOP CANDIDATE

BOUNCE

2008

Obama

??

McCain

 

2004

Kerry

-1

Bush, Jr.

2

2000

Gore

8

Bush, Jr.

8

1996

Clinton

5

Dole

3

1992

Clinton

16

Bush, Sr.

5

1988

Dukakis

7

Bush, Sr.

6

1984

Mondale

9

Reagan

4

1980

Carter

10

Reagan

8

1976

Carter

9

Ford

5

1972

McGovern

0

Nixon

7

1968

Humphrey

2

Nixon

5

1964

Johnson

3

Goldwater

5

 

 

 

 

 

Avg. 1964-2004

---

6.18

----

5.27

Avg. 1980-2004

---

7.71

---

5.14

Avg. 1992-2004

---

7.00

---

4.50







 

First, I wish to remind that Gallup is talking about national poll numbers here and we all know that the real battle is fought from state to state, so in my opinion, the best way to measure a bounce is to do in-depth state by state polling immediately following both conventions, but that is generally too costly and we already can assuming that not all 50 states will be polled after the respective conventions. But I plan to do that after both conventions – work with the individual state numbers that are available during the week(s) following both conventions.

 

Second, Gallup is only referring to it’s own national poll numbers. More fascinating would be to see the national polling averages (Gallup plus all of it’s major competitors) for each of these years, but alas, I am pressed for time and cannot dredge up all of those numbers this far back in history. Not only that, I seriously doubt that there were as many polls and pollsters in 1964 as is 2008, so a comparison is hard to make anyway. But if there is anyone out there who already has these numbers compiled, I would be very thankful for his or her input.

 

Third, as far as I can tell, this is the first time in recent history, or perhaps ever, that the two conventions are back to back, which means that the bounce and the counter-bounce will follow immediately after each other. Not all states will be polled just one week after the first convention, so this will cause people to want to interpret things that may not even be there. For my part, I can only suggest caution in interpreting the numbers.

 

Fourth, the bounce itself must be interpreted in light of the general standing of both parties pre-convention. For instance, if party B gets only a 2 point bounce, while party A gets an 11 point bounce, but party A was 25 points behind party B before the convention, then it means that party A is still behind. So, my largest critique of the gallup study is that it presented the bounce numbers in a vacuum of sorts.

 

Fifth, imo, the real proof of the pudding is seeing how much of each bounce remains during the month of September, as many voters will revert to their pre-convention convictions.

 

That being said, we see at first glance very quickly that a higher bounce does not necessarily mean a win in the fall. In 1988, 1984, 1980 and 1964, one party had the higher bounce, but the other party won in the fall. And the only example of a large difference in the two bounces that led to a massive PV/EV landslide for the party with the higher bounce was 1972. And it is statistically true that the democrats have tended more to have the higher bounce and still lose the GE.

 

We also see that in the case of a re-election, the second convention does not necessarily bring a larger bounce. See: Reagan 1988, Bush, Sr. 1992, Clinton 1996, Bush Jr. 2004. Of course, this has no bearing on this year, is interesting nonetheless.

 

We also see that from these 11 cycles before 2008, only 3 were “open” elections (election without an incumbent president running for re-election): 2000, 1988 and 1968, but in all three cases, it the residing vice-president who made a go at the presidency. There is also an election that had the look and feel of an open election (1976, due to Nixon’s resignation in 1974), but was not. So, 2008 will be the first year since 1928 (Hoover vs. Smith) !!!! where neither a sitting president is running for re-election, nor is a sitting vice-president a candidate for president.

 

It is, however, interesting that both parties got identical bounces in 2000, and we all know how that election ended up. With the national poll numbers so tight in this year, we could make the argument that identical convention bounces in 2008, should that be the case, could make for a repeat of 2000. I am not saying that this will happen, but am making a point of comparison.

 

It is also more interesting to see how small the bounces were in 2004, a year that was almost as emotion charged as this year is. Is it possible that the American electorate makes up it’s mind earlier as time goes on, in the age of Internet? That the constant, up to the second flow of information is dampening the afterglow of events like political conventions? I am asking these two questions to see if there are readers out there who would like to respond with some ideas and/or available statistics.

 

So, I have made three sets of averages: from 1964-2004, from 1980-2004 and from 1992-2004. Why? Well, the demographics of the union and party allegiances have shifted radically since 1964. The Reagan revolution of 1980 transformed the south into GOP territory and the Clinton sweep of 1992 redefined the Northeast and West Coast.

 

FACIT: looking that the averages, it would be more logical and fair to assume that an Obama bounce will be between 6-8% and the McCain bounce would be between 4-6%, if statistics are our only guide. However, with race a factor in this election plus a disgruntled core of Clinton supporters, plus the war in Iraq plus the price of gas plus the housing market crisis, plus plus plus…., these statistics may all just fly in the wind of history. If Obama pulls off a 15% jump, then I would be amazed, for I am of the personal opinion that the nation is far too divided among political trench lines for this to happen. Add to that the fact that this is the longest campaign in American political history, there is indeed the possibility that voter fatigue may already be setting in on all sides. Ditto McCain: I doubt he will have a massive bounce, either, and for exactly the same reasons.