First decent poll since the end of the Dem convention, basically shows the conventions cancelling each other out. Which is how it usually is.
Clinton saw an uptick in her favorable rating to 45-51 -a significant improvement.
Clinton leads the two was 50-45, and the three way 46-41
I have longed believed this is a 10-12 point race in the end. I got into a pretty good argument with Ed Goes at the realclearpolitics forum in Philadelphia about it.
There are two groups that haven’t come home: non-trump voting evangelicals and Sanders supporting young people. Ed says these groups are about the same size, and NEITHER is likely to come home. He was nice about it – and I don’t think he is voting for Trump.
He has better access to data than I have.
The gap to focus on is under 30. Note:
Obama 67-31
Clinton 51-34, 15% undecided
In all other age groups, the undecided is below 5. What is out is young, and would vote for Obama. The convention did not close the deal with that group.
Here is PPP on this:
“It’s also important to note that most
of the remaining undecided pool is very
Democratic leaning. They give Barack Obama a 55/33 approval rating, and they’d
rather have him as President than Trum
p by a 59/10 spread. If they ended up
voting for Clinton and Trump by those prop
ortions, it would push Clinton’s lead
up from 5 points to 8. But they don’t li
ke Clinton (a 4/83 favorability) or Trump
Phone: 888 621-6988
Web: www.publicpolicypolling.com
Email: information@publicpolicypolling.com
(a 2/89 favorability). A lot of these fo
lks are disaffected Bernie Sanders voters,
and even after the successful convention this
week they’re still not sold on Clinton
yet. She and her surrogates will have to
keep working to try to win those folks
over and if they can the election enters landslide territory”
http://www.publicpolicypolling.com/pdf/2015/PPP_Release_National_7302016.pdf
The Green Party is already on the ballot in Ohio, Florida, Wisconsin, Michigan, and Colorado.
The RCP average for Ohio has Stein at 3 points, with Clinton leading Trump by 1.3.
In Florida, Stein is at 2.5, with Trump leading Clinton by 0.2.
Left-leaning blogs may be insisting that you can’t vote for Stein, but clearly not everyone agrees. So I don’t see how anyone serious about handicapping the race can afford not to insist on 4-way polls.
Historically third party candidates fade at the line. This was certainly true in 2000, but also true in 1980 and 1968. The exception was Perot – but Perot was in the debates.
I think Stein may well hold her vote in deep blue states. Polling in Vermont shows that for example.
In deep blue states, Stein is a plus. Hillary will win those states anyway, and Stein who will bring in voters who will vote for Democrats downticket because the Greens frequently don’t even bother to run candidates.
So you’re saying that all convention bounces are equal and all fade quickly? Irrelevant factors:
Contested or Kumbaya convention
Open seat or incumbent running
Open seat VP running
Leading or trailing poll numbers going into the convention
Time gap between the DNC/RNC conventions (1 wk to 5 wks)
Days frm the end of the convention to election day (61 to 110 days)
IOW – convention bounces are as real as the “commies under my bed.”
No – they aren’t all equal. A history is here:
http://www.bleedingheartland.com/2016/07/20/in-11-days-we-will-probably-know-who-the-next-president-
is/
Here is a summary. They are real – often the race doesn’t change much from the last convention to the first debate.
Contested conventions can mean trouble – see the data on ’68
On its own, it still appears to be meaningless.
What definition of “contested” are you using? Seems to me that “contested” should mean that the leading candidate arrives at the convention short of the delegates required to secure the nomination on the first ballot and another candidate with a significant number of delegates is working to get more and has a plausible chance to get the nomination.
By that definition, only the 1976 Republican nomination was contested, but Ford wrapped it up before the first ballot as has every nominee since like forever. In 1972 McGovern arrived forty delegates short but he didn’t have a prime leading opponent. Kennedy contested in 1980, but Carter had more than enough delegates. Rockefeller contested the ’68 nomination at the convention, but Nixon had it sewn up before then.
Did McGovern really get a convention dip? Less rancor at that convention then many others when the nominee got a large bounce. Suspect that number reflects what happened a few days after the convention.
But again, what does a convention bounce even mean? HHH got a mere 4 points and Nixon got 14 points, but come November Nixon won by 0.7% (popular vote which is the same criteria used by pollsters in gathering convention bounce numbers.) That winning margin was less than half what Carter got in ’76 when he and Ford got the same convention bounce.
We can’t even say if an incumbent president get a weak bounce, that he’ll lose. GHWB’s was better than Clinton’s and Obama’s. Not saying it might not have some relevance in analyzing election outcomes, but as a single data point, it’s not any better than flipping a coin.
I made no real attempt to be predictive.
I started some analysis that attempted a linear regression based on % of opposition delegates. I kind of thought maybe when there is real opposition, maybe there is more drama.
But it really doesn’t work. All you can say is conventions move opinion quite a bit, in a unique way.
On ’72 what may have been forgotten is McGovern gave his acceptance speech well outside of prime time. The numbers I have are before the Eagleton revelation. Also forgotten is the busing debate on the floor.
If you read the link there is more analysis. No one has ever won if the trailed after their convention, for example.
The data set is way too small for any stat analysis to be valid.
McGovern gave his acceptance speech well outside of prime time. Very bad stage management. OTOH, iirc the media convention coverage that year wasn’t even-handed (using a generous definition of even-handed; Trump got a fairer deal than McGovern did) and too many powerful people in the Democratic party were undermining McGovern.
Your focus on this matter has been worthy for me because it prompted many questions for which I didn’t know the answer. For example, I never new that the order of the RNC and DNC conventions was static from 1864 through 1952; the RNC always held theirs first (with one exception). Since then, the out-party selects their convention dates first and the in-party choice is limited to dates following the other party. Even that is a bit of a crap shoot for parties because convention date decisions precede knowledge of the nominee and/or opponent.
Until 2008 there was a pattern to the convention date selections. A long general election campaign (from end of convention to election day) offers a slight advantage for a challenger (either party or incumbent). Shorter GE campaigns has been the incumbent preference. One does need a bit of a crystal ball to figure out which one would work best in an upcoming election cycle.
For example: In ’68 LBJ had planned to run and thus, the DNC convention was set for the end of August. But he didn’t run and that gave Nixon three weeks to hammer HHH and the Democratic Party. The mess in Chicago then hobbled HHH. One or two more weeks to recover and he might have won.
The 2008 DNC decision was novel and clearly well thought out in using all the variables that could have been known before the decision was made. Principally that GWB/GOP were on the ropes and federal funding of general elections was too small to run a long general election campaign. Also, deny the GOP little space to exploit 9/11 again. Wouldn’t have been as advantageous if GWB/GOP had recovered before July-Aug and the GOP and not the DNC had been in a position to pass on federal funding and the GOP ticket been stronger.
A risk for the candidate of the out-party in finalizing the nomination two or more weeks before his/her opponent is formalized is that it gives the media more time to explore and expose authentic shortcomings of the candidate without the distraction of balancing acts between two candidates. Four to six weeks can practically turn god into punchline joke. Just another one of those variables that DNC/RNC planners have incorporate and balance with all the other variables. Both parties have gone long this year. Not a good decision on the part of the DNC.
As I reviewed the data it became clear that the parties were pushing the conventions later and later, until they ended after labor day. They moved in this cycle because of finance issues.
I think it was mistaken. The later the better.
I remember when I first did this realizing that the order didn’t matter much, nor did incumbent versus non-incumbent. I was really surprised by that.
I have not done it, but I am curious what a look at favorable ratings might reveal.
But I have a day job.
Not seeing any particular advantage/disadvantage for a candidate/party to hold a convention later or earlier.
Of course financial considerations are embedded in the historical information. Far more so during the period from 1976 through 2008 with public financing. (Mostly neglected about 1972 was Nixon’s campaign funding operation.)
Using the measure of days available on the campaign trail for the two major party nominees, 2016 will be the longest general election since 1952. but 1960 was only a few days shorter. 2008 and 2012 were the shortest. Until 2008, there was always a period of two to six weeks between the two conventions. The DNC forced the RNC into a back-to-back schedule in 2008 and expect them to continue doing that until it doesn’t work.
heh- Trump flapping his gums for two to three weeks before the DNC convention would probably have been more advantageous, but a nominee like Trump wouldn’t have been predicted.
One of the bigger surprises I remember was the ’80 DNC – which I assumed was a disaster.
It wasn’t in terms of polling.
1980 broke extremely late in the election. That’s difficult for those that weren’t there to appreciate when their impressions/opinions are formed by the actual election results.
Reagan’s numbers didn’t move in the last week in his own polling. There is a long and complicated argument why the polling just missed and if it missed.
Looking back it still appears to me that it was too-close-to-call. Based on what the pollsters were looking for/at. A majority of the public had was sour but hadn’t completely soured on Carter. But to win, he had to catch a break, even if if was of his own making.
Doubt the pollsters looked carefully enough at Anderson and a large number of Democrats fell for the “it’s not fair” Anderson campaign, but that broke late as well. Democratic partisans loathe Nader who is probably honest enough, and I loathe Anderson because there was nothing honest about his campaign. All it had to do was thwart Carter in a few eastern seaboard states, and the election was over. Poaching Kennedy Democratic voters. Which he did in MA, NY, CT — and the momentum carried from there.
This is not to say that absent Anderson, Carter would have won. That can’t be known, and based on his performance the last year and a half in office, he deserved to lose. Just not to what, at best, would be worse. (Objectively, Carter ran one of the worst campaigns. One repeated by other Dem campaigns — run tough and hard against primary opponents and then put on kid gloves for the general.)
Offhand, not much of a correlation between the convention bounce and the winner. The exception might be that all three times one party had a 10 point advantage, it won. That doesn’t look to be the case this time.