01 November 2015

Polling round-up: 16-31.10.2015

This round-up repeats and enhances the data from the special one-week round up that I did HERE.

So, let's look at the October week 3-4 numbers:

GOP nomination - National:

Week 3:
NBC/SurveyMonkey, released 16.10.2015Trump +5 (was at +15 in the previous poll)
IPSOS/Reuters, dated 14.10.2015, released 16.10.2015: Margin =  Trump +13 (RRV + IRV, R-leaning), Trump +14 (RRV only), Trump +7 (IRV, R-leaning only)*
NBC/WSJ, released 19.10.2015Trump +3
CNN / ORC, released 20.10.2015Trump +5
Monmouth University, released 20.10.2015Trump +10
Morning Consult, released 20.10.2015Trump +26
ABC/WAPO, released 20.10.2015Trump +10
IPSOS/Reuters, released 22.10.2015:  Margin =  Trump +15 (RRV + IRV, R-leaning), Trump +13 (RRV only), Trump +12 (IRV, R-leaning only)

We see 8 national polls over 7 days and see a positive jump in Trump's numbers as of polls released on 20.10.2015.  One poll is a repeater, so if you take the first IPSOS/Reuters out of the equation and average the 7 polls, then it is: Trump +10.3, a landslide margin over 2nd place Dr. Ben Carson - nationally.

Week 4:

Morning Consult, released 26.10.2015: Trump +15
CBS/ NYT, released 27.10.2015: Carson +4
Gravis Marketing (R) / OANN, released 27.10.2015: Trump +14
The Economist / YouGov, released 29.10.2015: Trump +14
IPSOS/Reuters, released 29.10.2015: Margin =  Trump +8 (RRV + IRV, R-leaning), Trump +2 (RRV only), Trump +15 (IRV, R-leaning only)
St. Leo University, released 29.10.2015: Trump +0.5 (statistical, near-mathematical tie)
IBD/TIPP, released 30.10.2015: Trump +5

We see 7 national polls all released within 5 days and for the first time, Carson wins a national poll and there is really also a tie (CBS) and a statistical tie in the IPSOS poll among RRV only (+2). The average of these 7 polls is: Trump +6.6. So, in spite of three polls showing Trump +14 or higher, the average for week four is lower than for week three, by 4.7 points. Also, we see a top-four column of candidates emerging, where all four more often than not come in at double digits: Trump, Carson, Rubio and Cruz. Polling from both week 3 and 4 of September clearly shows Fiorina receded back into low single digits. Her boomlet may now be over with. None of these polls yet show the results from the Republican debate: we will see that data likely starting today, November 1st, 2015.



GOP Nomination - State Polling (10 states):
Arizona:
MBQF Consulting, released 30.10.2015: Carson +0.7 (statistical tie, and a big shift away from Trump in this state)

Florida:
St. Leo University, released 29.10.2015: Trump +4.3. Rubio is in 2nd and over 20%

Iowa:
Quinnipiac, released 21.10.2015:  Carson +8
Selzer / Des Moines Register / Bloomberg, released 23.10.2015Carson +9
Average, week 3: Carson +8.5
week 4:
CBS/YouGov, released 25.10.2015: TIE (27/27)
Monmouth University, released 26.10.2015: Carson +14 (some fascinating internals)
Loras College, released 26.10.2015: Carson +12 (some fascinating internals)
Average, week 4: Carson +8.7,
No real change over week 3 (+0.2 shift is just statistical noise). Carson is now in a statistically significant lead in primary battleground Iowa.

Massachusetts:
Emerson College Poll, released 20.10.2015Trump +33.9

New Hampshire:
PPP (D), released 20.10.2015Trump +7
Bloomberg / Saint Anselm / Purple Strategies, released 21.10.2015Trump +7
Average: Trump +7
Week 4:
CBS/YouGov, released 25.10.2015: Trump +26
Contrary to Iowa, CBS is showing a huge shift toward Trump in battleground New Hampshire.

North Carolina:
PPP (D), released 27.10.2015: Trump +8

Pennsylvania:
Franklin & Marshall, released 29.10.2015: Trump +1 (statistical tie)

South Carolina:
CBS / YouGov, released 25.10.2015: Trump +26
Clemson University, released 27.10.2015: Trump +4

It's really not possible to see such values out of the same state within the same time. That's a 22 point difference in margin - statistically impossible. One of those two polls is WAY off and since Trump has consistently led by double digits here since the first SC poll was taken with him in the mix, I tend to believe the CBS poll more. That being said, average: Trump +14

Both Trump and Clinton are cruising for their respective parties in this state. It is not even remotely competitive.

Texas:
Dixie Strategies/KTVT CBS 11, released 27.10.2015: Carson +1 (statistical tie). Cruz is in 3rd place.

Wisconsin:
St. Norbert / WPR / WPT, released 21.10.2015Carson +2

There was not a deluge of GOP nomination state-polling, but the results from Iowa indeed reflect a change in Dr. Ben Carson's favor and the two margins are very similat to each other. Carson is also slightly ahead in Wisconsin, but the margin is well within the MoE and therefore, it is a statistical tie. In Massachusetts, Donald Trump has a blowout-landslide lead over the field.

What remains to be seen is out the strange Trump tweet about Iowan's brains being drug addled if they are favoring Carson - and how that will play in the next round of polling.

All of the toplines, survey group and some internals for all of the GOP data is HERE to be found.


DEM nomination - National:

Week 3:
NBC/SurveyMonkey, released 16.10.2015Clinton +14 (with Biden in the mix)
IPSOS/Reuters, dated 14.10.2015, released 16.10.2015:  Margin = Clinton +17 (DRV +IRV, D-leaning), Clinton +26 (DRV only), Clinton +4 (IRV, D-leaning only- all with Biden in the mix)
CNN/ORC, released 19.10.2015Clinton +16 (with Biden in the mix)
Monmouth University, released 19.10.2015Clinton +27 (with Biden in the mix), Clinton +33 (without Biden)
NBC/WSJ, released 19.10.2015Clinton +20 (with Biden in the mix), Clinton +25 (without Biden)
ABC/ WAPO, released 20.10.2015Clinton +31 (with Biden in the mix)
Morning Consult, released 20.10.2015:  Clinton +32 (WITHOUT Biden in the mix)
IPSOS/Reuters, released 22.10.2015: Margin = Clinton +10 (DRV +IRV, D-leaning), Clinton +16 (DRV only), Sanders +7 (IRV, D-leaning only - all with Biden in the mix)

As with the GOP field, this makes for 8 polls in less than seven days, with one repeater. With both the Monmouth and NBC/ WSJ polls, we can see that Clinton's margin over Sanders rises by 5 points in both polls without Biden in the mix.  Since Biden is definitely not running, for the averages, I am taking the values that can be taken without Biden to average, so the average of 7 polls (excluding the first IPSOS/Reuters) is: Clinton +23.9, a very large landslide margin. And in a number of those polls, Clinton is back over the 50 mark again. As was the case with Donald Trump and the Republican field, Clinton's numbers also dramatically improved as of the 19th of October (it was the 20th for the GOP). There is no statistical doubt that Hillary Clinton is the prohibitive front-runner within the Democratic field.

Week 4:
Morning Consult, released 26.10.2015: Clinton +27
The Economist / YouGov, released 29.10.2015: Clinton +32
IPSOS/Reuters, released 29.10.2015:  Margin = Clinton +15 (DRV +IRV, D-leaning), Clinton +20 (DRV only), Clinton +2 (IRV, D-leaning only)
St. Leo University, released 29.10.2015: Clinton +42.8
NBC / SurveyMonkey, released 30.10.2015: Clinton +20
IBD/TIPP, released 30.10.2015: Clinton +15

That makes for 6 new national polls, released within 5 days of the fourth week. Average: Clinton +31.1.  That is a marked shift upward from the 3rd week and statistical proof (which is also being noted by many pollsters) that would-have-been Biden voters have moved to Clinton, and not to Sanders.



DEM nomination - State polling (9 states):

Florida:
University of N. Florida (UNF), released 21.10.2015Clinton +38.7 (with Biden in the mix)
Week 4:
St. Leo University, released 29.10.2015: Clinton +37.9

The two polls are almost identical to each other. It's a close race for the Republicans in this state, for the Democrats, Clinton is cruising.

Iowa:
Selzer / Des Moines Register / Bloomberg, released 22.10.2015Clinton +5 (with Biden in the mix), Clinton +7 (without Biden)
Quinnipiac, released 23.10.2015Clinton +11 (without Biden)
Average: Clinton +9
Week 4:
Loras College, released 27.10.2015: Clinton +38 (interesting internals)
Monmouth University, released 27.10.2015: Clinton +41 (also an interesting internal at the poll)
Average: Clinton +39.5
That is an enormous shift for Clinton over week three, a shift of 30 points in her favor.

Massachusetts:
Emerson College Poll, released 20.10.2015Clinton +23.8 (without Biden), Clinton at almost 60%.

New Hampshire:
Suffolk University / Boston Globe, released 16.10.2015Clinton +1.4 (with Biden in the mix)
Franklin Pierce / Boston Herald, released 18.10.2015Sanders +8 (with Biden in the mix), Sanders +10 (without Biden)
PPP (D), released 20.10.2015Clinton +8 (with Biden in the mix)
MassINC/WBUR/NPR, released 20.10.2015Clinton +4 (with Biden in the mix)
Bloomberg / Saint Anselm / Purple Strategies, released 21.10.2015: Sanders +5 (with Biden in the mix)
Average: Sanders +0.32
Week 4:

CBS/YouGov, released 25.10.2015: Sanders +15
No average for week 4 is possible, but the CBS poll shows Sanders firmly in the driver's seat in this state.

North Carolina:
PPP (D), released 27.10.2015: Clinton +37
Clinton is cruising among Democrats in the Tarheel State.

Pennsylvania:
Franklin & Marshall, released 29.10.2015: Clinton +34
That is a crushing margin for a state that Clinton won by circa +9 over Obama in the 2008 primaries.

South Carolina:
CBS / YouGov, released 25.10.2015: Clinton +43
Clemson University, released 27.10.2015: Clinton +37
Average: Clinton +40

Both Clinton and Trump are sweeping with their respective parties in this state.

Texas:
Dixie Strategies/KTVT CBS 11, released 27.10.2015: Clinton +49
That is a blowout margin that is statistically impossible to undo. I have never seen a candidate, ever, who was showing these kinds of margins, go on to lose the state in the ensuing primaries or in a GE.


Wisconsin:
St. Norbert / WPR / WPT, released 21.10.2015Clinton +2 (with Biden in the mix), Clinton +5 (without Biden)

The state polling for the DEMS also shows some surprises, namely, that Hillary Clinton is regaining ground in New Hampshire, is cruising in Massachusetts, is now statistically significantly ahead in Iowa, but it is surprisingly close in Wisconsin, a possible new battleground opportunity for Bernie Sanders.

The long-and-short of weeks 3 and 4 of October 2015 on the Democratic side show very clearly at both the national and the state levels that a.) Biden voters are moving toward Clinton and b.) she is cruising everywhere that is being polled except for New Hampshire (and to a certain degree, Wisconsin). If this trend continues, then Bernie Sanders may end up only being a one or two trick pony in the DEM primaries.  Looking at that massive margin that Hillary is enjoying in Pennsylvania, I wish that an Ohio poll or two had come in during this time-frame, for Hillary won both states against Obama during the 2008 primaries by very similar margins (+9.14 in PA, +8.66 in OH). That would have been fascinating to see.

All of the toplines, survey group and some internals for all of the DEM data is HERE to be found.

D vs. R matchups: NATIONAL POLLING

CNN/ORC, released 19.10.2015:  Clinton and Sanders vs. Carson and Trump (4 matchups). Trump loses both. Carson wins both.

NBC/WSJ, released 19.10.2015: Clinton and Sanders vs. Rubio, Carson and Cruz (6 matchups), all D wins.

Morning Consult, released 20.10.2015: Clinton vs. 6 GOPers. She wins all 6 matchups.

Rasmussen Reports, released 20.10.2015; Clinton vs. Trump and Fiorina. Loses to Trump by 2, wins over Fiorina by 6. The topline numbers are ridiculously low, practically impossible for this stage in the game.

Week 4:

Gravis Marketing (R) / OANN, released 27.10.2015: with 7 Hillary vs. GOP contender matchups, Hillary wins 5 and there are two matchups.  Just one month ago, Gravis (R) / OANN was showing Trump with 54% nationally and beating Clinton by 9 points, which brings me to some critical points: apparently, Gravis either finds no undecideds at all, because all of the matchup values between the two major candidates always add up to 100%, or Gravis filters out the undecided vote and recalculates the margins, which is intellectually dishonest, to put it charitably. Gravis (R) releases very few internals, but this time, it did and the internals match up to the demographics of the 2012 election. Gravis does not archiv this stuff anywhere, it just lands on its national polling page, so the only way to save the data for posterity's sake is to screenshot it. OANN is tied to World Net Daily, a birther outlet. Real Clear Politics has, until now, refused to post the national polling results from Gravis (R) / OANN.

St. Leo University, released 29.10.2015: 6 Clinton vs. GOP matchups, she wins all six,and big. Also, 6 Clinton vs. GOP vs. Trump as Independent matchups, and Clinton wins all of them as well.




D vs. R matchups: STATE POLLING (3 states)


New Hampshire:
PPP (D), released 21.10.2015: Clinton vs. 9 GOPers, Sanders and Biden, each vs. 3 GOPers (15 matchups total). The three common denominators are: Rubio, Carson and Trump. Clinton ties Kasich, beats the other 8. Sanders and Biden beat all of their opponents. The New Hampshire poll was released on the day that Vice-President Biden announced that he was not running for president.  These numbers reflect a stark improvement for Hillary Clinton over last month. New Hampshire is now a 5-for-6 Democratic state on the presidential level, having gone for the Democratic candidate in every presidential cycle since 1992, excepting 2000,, but it was a hotly contested battleground state in both 2000 and 2004. Some pollsters claimed that it was a battleground in 2012, but Obama won the state easily by +6, three time the margin with which Romney won North Carolina.

North Carolina:

PPP (D), released 27.10.2015:  9 Clinton vs. GOP matchups, 4 Sanders vs. GOP matchups: Sanders loses all four. Clinton loses 6, wins two, but two GOP wins and both of her wins are withing the MoE. That being said, the GOP margins over her have shrunk decisively since the last PPP poll was taken of NC.

Wisconsin:
St. Norbert / WPR / WPT, released 21.10.2015: Clinton, Sanders and Biden against Carson, Bush and Trump. The DEMS win all 9 matchups. Trump does the worst, with Clinton +11Sanders +18 and Biden +19 over Trump. Obama won Wisconsin in 2008 by +14 and in 2012 by +7. Wisconsin is now a 7-for-7 Democratic state on the presidential level, having gone for the Democratic candidate in every presidential cycle since 1988, but it was a hotly contested battleground state in both 2000 and 2004. Some pollsters claimed that it was a battleground in 2012, but Obama won the state easily by +7, roughly the same margin with which Mitt Romney won Georgia, which was never in play in the 2012 presidential cycle.

It's good that these two states were polled from reputable pollsters, for they are both states within the so-called "Blue Wall", states that become hotly contested battlegrounds in close presdential elections. In a way, both of these states are pretty good canaries in the coal mine.

All of the toplines, survey group and some internals for all of the Presidential D vs. R matchup data is HERE to be found.


GUBERNATORIAL / SENATORIAL:

This LINK has data concerning Gubernatorial/ Senatorial races. Thus far, races have been polled in Kentucky, Louisana, Maryland, New Hampshire, North Carolina and Wisconsin. The Louisana polling was right on the mark, as proven last Saturday. The Republican polling firm Vox Populi is the only one to show the race for Tuesday between Bevin and Conway to be tied, all others show Conway demonstrably ahead. But VP also polled the 2016 senatorial, showing Paul well ahead of Democrat Edelen, probably a message to Paul from VP.

There was also less time for FAV/UNFAV data to be logged, but this one from Texas was interesting:


Dixie Strategies/KTVT CBS 11, released 27.10.2015: Obama 44/50, -6.  I was expecting to see him at -20 in Texas.

FACIT: there was just a bundle of polling in the last two weeks and it clearly shows that Clinton, in the absense of a Biden candidacy, now has a very commanding lead in the race for the Democratic nomination. Insiders, many of them, are already saying that it's all over. I'm not going to be subjective about it. All I can say is that the numbers points to overwhelming margins for Clinton in just too many states, so, at least at this point in time, numerically, I don't see how she can lose the nomination.

On the Republican side, the polling has taken on an entirely different dynamic than it had in the Fall of 2011. In 2001, Mitt Romney was mired at about 23% on the aggregate, sometimes he was on top, but as of August, it was Perry, then Cain, then Gingrich and in the primaries, it was Santorum who gave him a challenge. This time around, Donald Trump has held a statistically significant lead for three solid months on end, and only now has Carson beat him and tied him in a couple of national polls. Even so, on the aggregate, Trump is leading pretty much everywhere. His national average over the GOP field is less than Clinton's over the DEM field, but in many individual states (Massachusetts, South Carolina, for instance), he is also recording crushing margins in polling. I would say that all bets are still off concerning the Republican field, but it looks very much like their are now fout top-tier candidates, and Jeb Bush is not one of them.

No comments:

Post a Comment

Constructive comments and critique are always welcome. Please keep it polite and respectful.