Generational Theory Forum: The Fourth Turning Forum: A message board discussing generations and the Strauss Howe generational theory

Full Version: Donald Trump: polls of approval and favorability
You're currently viewing a stripped down version of our content. View the full version with proper formatting.
Pages: 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40
One of the last polls that I expect to see from 2019 due to the impending Holidays. I apologize for posting it in three different threads, but it is germane to impeachment, the 2020 election, and polling:

:Morning Consult/Politico, Dec. 19-20, 1387 RV (prior poll Dec. 14-15)

This poll was conducted entirely after the House impeachment vote.

Approve 41 (nc)
Disapprove 56 (+1)

Strongly approve 27 (+3)
Strongly disapprove 46 (+2)

Do you approve or disapprove of the House of Representatives impeaching President Trump? (Previous wording: would you approve of...)

Approve 52 (+2)
Disapprove 43 (nc)

And would you approve or disapprove of the Senate removing President Trump from office?

Approve 51 (+1)
Disapprove 42 (-1)
50 state polls from one source... from a pollster secretive about its methods and often out of touch with other pollsters.

Morning Consult, December '19:

[Image: genusmap.php?year=2012&ev_c=1&pv_p=1&ev_...&NE3=0;1;6]

Trump advantage 10% or higher (really dark for 20%+)
Trump advantage 5% to 9%
Trump advantage 1% to 4%
tie -- no advantage (white)
Trump disadvantage 1-4%
Trump disadvantage 5% to 9%
Trump disadvantage 10% or more (really dark for 20%+)

Medium-to-deep red seem to give Trump a loss outside of the margin of error (4%) for 284 electoral votes. 147 are on the bubble (I can in no way see that NE-03 goes any way except for Trump) as they will be decided by 4% or less. That leaves 117 sure-things for him.

Trump isn't winning with these state numbers.
If you wonder why I have not been posting on this thread, then it has been because

Quote:(1) most of the polling has related to a contested Democratic set of primaries, and (2) polling has been remarkably stable.

Then the securities markets tanked (yes, the eleven-year bull market has come to an end), and COVID-19 has started killing in large numbers. I posted this on the Forums (now called "Talk Elections") on what is still Leip's Atlas of Elections. Many of the assumptions that have been reliable since about 2000 may no longer be so -- except for the ideological  polarization of the American electorate.

   Good question. Things are going wild. This virus is a confirmed killer. Assuming close splitting of the electorate, how this virus ravages the population will shape the election. If it should kill poor non-whites selectively then it could put such states as Michigan, Pennsylvania, and Wisconsin  back within the range of Trump victories. If it should selectively kill low-information white voters in rural areas then it favors Joe Biden. Unlike HIV/AIDS (the most infamous "new" lethal infectious disease of the last fifty years before this one that has killed tens of thousands of Americans)  it kills fast, so many people generally considered likely voters might not be voting. One has no ability to cast a ballot from six feet under. Someone with HIV/AIDS could be voting for several years after diagnosis. Someone with COVID-19 could die in a week.

So far people most at risk are those who have done recent traveling, and that tends to skew toward well-heeled, well-educated, urban and suburban voters. But how do we know that it will not spread to rural areas with poor, ill-educated people with bad habits such as smoking, physical inactivity, and alcohol abuse -- maybe some oxy and meth on the side?

Assumptions about the 2020 electorate can be rendered completely absurd! If people in large numbers are undergoing medical treatment then they might not be reachable by pollsters at the time.

Add to this -- the securities markets are in chaos at or after the end of an extended bull market, and Americans are  being laid off in large numbers. To be sure most Americans own no stock even through a 401K plan, but any reduction in consumer demand will result in mass layoffs. If your retail choices get reduced to Dollar General, Family Dollar, Wal*Mart, Aldi, Kroger, Big Lots, Target, Walgreen's, CVS, Rite-Aid, Home Depot, gas station-convenience stores, auto parts stores, and maybe some regional box stores then you have lost much choice of what you can buy. As a group those are not high-end marketers, to put it lightly. So let us suppose that you are in the business of manufacturing high-end audio equipment. People just do not buy that sight-unseen...or sound unheard.

Maybe you buy books, audio recordings, or recorded video sight-unseen from the Big A, who should be doing well enough now. But such people know what they are looking for. (Closure of public libraries might be a boon for the Big A, but that is likely to be short-lived). Much marketing depends upon bored people going into a store and finding something to pique their interests. That is over for a while.  Need I go into a long discourse on how reduced sales lead to reduced production and that higher unemployment leads to a less sympathetic view of the leaders most influential at the time?

I may have expected the long bull market to come to an end, but I could not say when. COVID-19? Nobody predicted that until it happened, and nobody can predict how it will shape the electorate.

I have made assumptions on how approval and especially disapproval ratings either ensure re-election of an incumbent President (Obama 2012 is obvious), make it shaky (Dubya in 2004), or doom any re-election bid (Carter in 1980). COVID-19 is the wild card not supposed to be in the deck  that somehow found its way into the deck and is now in play.    

Contrary to what you may have learned in high-school civics in the 1990's or earlier, electoral politics is not so much about politicians convincing people to vote for them, but instead (in most states) the ratification of the demographics of the time. Thus states similar in economics but not ethnic mix (let us say Arkansas and New Mexico) can vote very differently. COVID-19 can change the demographic mix depending on who it kills. Not since 2001 have so many Americans died of the same cause in one day, and I do not see that ending soon. One month of such deaths can give a death toll higher in one month alone as the total number of combat deaths in the wars in Korea or Vietnam in a much shorter time.

COVID-19 has shattered some of my core assumptions about this Presidential election.  I might hold off on new maps unless some poll is 'juicy'.
OK -- so you might get some groceries at Whole Foods or Trader Joe's, which seem to be a bit more upscale.
NPR: Poll: Americans Don't Trust What They're Hearing From Trump On Coronavirus

I don't usually report on polls, but it seems like the Happy Talk do nothing policy Trump has been running is finally having some effect.
Do not lie.
Do not cheat.
Do not steal.

Do not tolerate lying, cheating, or stealing by others.

Such is in essence the Honor Code of the military academies, institutions that recognize the necessity of enforcing character and making it indelible in a personality. Obviously little else in life is quite combat that requires trust in superiors knowing what they are doing, expressing reality even when such is inconvenient or unsettling, not taking credit from others or fabricating results, and obviously not taking what one is not authorized to take. To be sure one might need to take stuff from a military store to allow completion of an objective, but one had better sign for it so that the quartermaster knows what needs be replenished. Someone else might need much the same things. Liars, cheats, and thieves in combat can do great harm.

In real life, too? Of course. Maybe we must tell little white lies (that dress looks great on you even if it is the wrong size; the awful performance on the violin of a seven-year-old child is at least promising; the painting of an automobile graveyard has unusual meaning as a piece of art; the Detroit Tigers really do have a chance of winning -- if you are selling tickets to their games). But such is commerce and social life, and by a certain age most people know that excessive frankness in judgments is good for a knuckle sandwich or social exclusion.

Donald Trump is definitely a liar.
(04-08-2020, 02:33 AM)pbrower2a Wrote: [ -> ]Donald Trump is definitely a liar.

I know that. You know that. A lot of people know that. Then again, there are those people he is telling what they want to hear...
(04-08-2020, 02:49 AM)Bob Butler 54 Wrote: [ -> ]
(04-08-2020, 02:33 AM)pbrower2a Wrote: [ -> ]Donald Trump is definitely a liar.

I know that.  You know that.  A lot of people know that.  Then again, there are those people he is telling what they want to hear...

There just might not be enough of them left for him to win. From the forums of Leip's Election Atlas (my material in blue and bold)
Quote:Ipsos/Reuters, April 6-7, 1116 adults (1-week change)

Approve 40 (-4)
Disapprove 55 (+4)

They have not (yet) published strongly approve/disapprove or RV numbers this week (they're sometimes oddly inconsistent that way).

Quote:To add to the previous post:

The Economist/YouGov weekly tracker, April 5-7, 1500 adults including 1147 RV


Approve 45 (-2)
Disapprove 51 (+5)

Strongly approve 26 (nc)
Strongly disapprove 38 (-4)


Approve 45 (-2)
Disapprove 53 (+3)

Strongly approve 30 (+1)
Strongly disapprove 43 (+3)

2020 (RV only):

Generic D 47 (+2), Trump 42 (nc)
Biden 48 (+2), Trump 42 (nc)
Sanders 48 (+3), Trump 42 (-2)

GCB (RV only): D 48 (nc), R 39 (nc)

No way does the President win with numbers like those. Note well that time is running out on the 2020 electoral season. Disapproval at 51% makes re-election require significant improvement in the view of the President, and disapproval at 55% suggests a landslide loss. 

55% disapproval means that GOP support erodes in places in which Republicans are usually safe. It would be difficult on my part to predict where that would happen... maybe white Southerners might become less tribal in their voting habits. 

Of course there is no legal difference between winning the White House with 275, 375, 475, or even 525 electoral votes... but at such a point I could even see Doug Jones getting re-elected in Alabama and some surprising losses for Republicans.

COVID-19 has been killing Americans in large numbers (in what epidemiologists call "excess deaths") for perhaps two weeks, and the toll in such a time is beginning to exceed  combat deaths in "small" wars. The raw numbers can easily reach those of American combat deaths in the Korean and Vietnam wars. Americans, and people in other advanced industrial societies, do not ordinarily die in large numbers of respiratory infections except as complications of existing (and usually terminal) cases. Pointless death offends us, which explains why we have laws against drunk driving -- let alone murder. Political leaders involved with pointless death such as wars whose purpose few people understand or understand for reasons inconvenient to those leaders, are in deep trouble. Think of LBJ, who actually achieved some very good things for multitudes of Americans.  
This is not so much approval and disapproval of the President put instead a potential projection of an electoral collapse for the President. When the Pope makes statements more credible to Americans (including non-Catholics) than the President ... and the Queen of England deep into her nineties is far more coherent and rational... then the President has a huge problem.

With several caveats (these polls involve on-line, self-selected polling that may understate populations unlikely to be on the Internet or to not trust polling, such as older, poorer white people of limited education -- and I say this because other ethnic groups don't have such a variance from other voters in their groups), here is one near-50-state poll involving the Internet.


Note that a Trump collapse looks much like this.

[Image: genusmap.php?year=2012&ev_c=1&pv_p=1&ev_...&NE3=2;1;7]

Note that Vermont, the District of Columbia, North Dakota, South Dakota, and Wyoming are not polled. I have educated guesses about the districts. But we need not be stupid about those electoral bailiwicks. Trump will be absolutely crushed in Vermont and DEE CEE (deep red) and will win North Dakota (medium blue), South Dakota (medium blue), and Wyoming firmly (deep blue).

Biden 55% or higher
Biden 50-54%
Biden ahead with 49% or less
ties are in white
Trump ahead with less than 49%
Trump 50-55%
Trump 55% or higher

...I look at this and see Trump losing to Biden, Biden getting absolute majorities of 50% or more in enough states in which to get elected. Biden also wins every state that Hillary Clinton won outright -- except Texas, which appears as a tie. Biden also wins Louisiana and Montana, which to Trump by more than 10% in 2016. Sure, it is only a 1% lead for Biden... but I have my suspicion that Trump is in trouble in Louisiana more than in any other state in the Deep or Mountain South (except Georgia) because Trump may have lost a constituency which swung from D to R about twenty years ago. It is almost exclusively associated with Louisiana: Cajuns. They seem to have Catholicism heavily entrenched in their culture. The Pope may have fewer military divisions than the President, but he certainly has shown far more credibility than the President on COVID-19. COVID-19 has ravaged Louisiana more than many other states.

There are other Catholic constituencies that can be swing voters: Polish- and Italian-Americans in Michigan, Minnesota, Nevada, New Hampshire, Ohio, Pennsylvania, and Wisconsin; Cuban-Americans in Florida, and perhaps Cajuns in Louisiana. People who more trust the Pope than the President are big trouble for the President.

This could be ahead of a current reality... but there is plenty of time for this sort of reality to emerge and little time for the President to undo his damage.
Showing the trend, which really counts, for states of my selection. The gap is between Biden and Trump in the electoral matchup (how people would vote on the day of the poll), R (Republican) suggesting a gap favoring Trump, D (Democrat) suggesting a gap favoring Biden, and "TIE" suggesting no obvious gap.  

I have chosen states with an eye (although I do not always get it) to geographic diversity within groups and, with two exceptions, states with six or more electoral votes.  

Reliably R:

state 3/1 3/15 4/9
AL 16R 19R 12R
IN 10R 10R 5R
KS 16R 15R 12R
WV 12R 9R 6R

States racially-polarized (usually R, but oddities this year):

state 3/1 3/15 4/9
LA 2R 4R 2D
MS 3R 6R 1R
SC 6R 3R 2R

Fringe of contention Trump wins, 2016

state 3/1 3/15 4/9
NC 5D 4D 10D

Obama 2012/Trump 2016 + AZ

state 3/1 3/15 4/9
AZ 1R 1D 5D
FL 6D 6D 11D
MI 8D 8D 11D
OH 1R 2R 5D
PA 7D 11D 12D
WI 4D 5D 8D

Bare Trump losses 2016

state 3/1 3/15 4/9
ME 12D 13D 17D
MN 15D 16D 21D
NH 1D 2D 7D
NV 12D 12D 16D

Fringe of contention against Trump, 2016

state 3/1 3/15 4/9
CO 10D 11D 15D
NM 17D 19D 18D
VA 9D 8D 13D

Reliably D

state 3/1 3/15 4/9
CT 23D 25D 28D
IL 24D 24D 28D
WA 22D 22D 28D

D max-out

state 3/1 3/15 4/9
CA 32D 33D 36D
MD 30D 29D 34D
NY 31D 32D 37D


Do I think this polling completely reliable? Hardly. It rules out people who have no Internet access or do not know how to use it. That likely means older, poorer, less-educated people. If such people are Hispanic or non-white then that means little because poorer people in  such populations vote much like the younger, richer, better-educated people in their groups. The sorts of people that Hillary Clinton understandably yet regrettably called a "basket of deplorable(s)", the "low-information voters" for which Trump expressed love, are likely under-represented. This likely most applies to the states that I have placed in the "racially-polarized" category, where white people are especially likely to at least be poorer and less educated... in states in which usually voting divides neatly on ethnic lines.  

Still, although I see effectively no change between March 1 and March 15, I see an overall swing of about 5% in practically all states. This is consistent with the Quinnipiac poll (+8) from the middle of last week, one for CNN (+11), and the poll by a British pollster with whom I am unfamiliar (+12). I would not be surprised to find polling by news agencies in countries that find, except for COVID-19, the 2020 Presidential election of the United States to the most likely biggest story of the year.  

Collapses happen in Presidential races, and they usually have obvious and evident cause. 2000 deaths per day from COVID-19? That is approaching Civil War levels.
Some state polls from real clear politics:

Florida: Trump vs. Biden UNF Biden 46, Trump 40 Biden +6
Wisconsin: Trump vs. Biden Marquette Biden 48, Trump 45 Biden +3
New York: Trump vs. Biden Siena Biden 58, Trump 33 Biden +25
Michigan: Trump vs. Biden MRG Biden 44, Trump 41 Biden +3
Arizona: Trump vs. Biden Monmouth Biden 46, Trump 43 Biden +3
Arizona: Trump vs. Biden NBC News/Marist Biden 47, Trump 46 Biden +1

National approval:
President Trump Job Approval
Disapprove +6.5
From realclear politics April 24, 2020:

President Trump Job Approval
Disapprove +5.3

General Election: Trump vs. Biden
Biden +5.9

Direction of Country
Right Direction35.1
Wrong Track57.7
Wrong Track +22.6

2020 Generic Congressional Vote
Democrats +7.4

April 24:
Florida: Trump vs. Biden FOX News Biden 46, Trump 43 Biden +3

Thursday, April 23

Race/Topic Poll Results Spread
Pennsylvania: Trump vs. Biden Susquehanna Biden 48, Trump 42 Biden +6

April 22:
Florida: Trump vs. Biden Quinnipiac Biden 46, Trump 42 Biden +4
Pennsylvania: Trump vs. Biden FOX News Biden 50, Trump 42 Biden +8
Michigan: Trump vs. Biden FOX News Biden 49, Trump 41 Biden +8
General Election: Trump vs. Biden Economist/YouGov Biden 48, Trump 42 Biden +6

April 16:
North Carolina Senate - Tillis vs. Cunningham PPP (D) Cunningham 47, Tillis 40 Cunningham +7

April 15:
Arizona Senate - McSally vs. Kelly OH Predictive Insights Kelly 51, McSally 42 Kelly +9
Civiqs polls, all fifty states:

[Image: genusmap.php?year=2012&ev_c=1&pv_p=1&ev_...7&NE=2;2;7]

Trump disapproval  55% or higher
.........................  52% to 54% 50% saturation
.........................  50% or 51% 40% saturation
.........................  under 50% but higher than approval

ties (white)
Trump approval...  under 50% but higher than disapproval
.........................  50% to 54%  
.........................  over 55%

Note that I make no distinction above 55% approval or above 55% disapproval because states in those categories will go 10% or so at the least one way or the other, and the distinction between 56% and 90% is moot in the Electoral College except for ME-02 or NE-02 not shown in this data. I make a distinction between 51% and 52% disapproval because disapproval at 52% suggests irretrievable loss, but 51% can be turned around (Obama did that once). Remember: the States elect the President and the People don't. Hillary Clinton won the popular vote by running up huge margins in states like California, Maryland, Massachusetts, New Jersey, and New York that had no chance of going for Donald Trump. She was going to win 119 electoral votes from those states if she averaged 57% or 77% of the vote in those states.

I haven't seen polling from this source since February. A result along this line would be a victory for Joe Biden on the scale of Clinton in the 1990's or Obama in 2008.

If anyone wishes to see this as evidence of a Trump collapse -- you are welcome. Tip-offs on electoral collapses for an incumbent President include that states that are usually super-safe become less so, even entering the brink of competitiveness. This map shows Montana as a virtual tossup (which it has been only on rare occasions in Presidential elections) and Kansas, Louisiana, and Mississippi on the fringe of competitiveness at this point. Let's put it this way: when Massachusetts was in play for Eisenhower in 1952 or Reagan in 1980, then the Democrat was in big trouble.

The first evidence that I saw of a likely Trump loss in his re-election bid was that states that had been bare losses for him had swung out of contention following a narrow victory in the Electoral College. To be sure, whether Minnesota, Nevada, and New Hampshire all go to Biden 49-48 or 65-34 mean the difference between twenty and twenty electoral votes that Trump can win without... but if something goes wrong for him in three significant states as such happens, then he loses. Such turned his chance of winning re-election about a 75% chance of a repeat to about a 40% chance of a repeat.
At this rate, if AZ continues in the dark red vein, Biden won't need WI, where his lead in current polls is slim, but will still need PA and MI to win, although they aren't in the dark red category on this map.

Minnesota is not listed dark red, but is likely Biden. If you include all the dark red states, plus PA and MI, where Biden has a good lead now in other polls, and MN, that's 269, to which would be added 1 Maine district for 270. But New Mexico is also not in dark red on this map, but is also likely Biden, which makes it 275.

This is the first time I've seen a national poll with Biden having a double digit lead.
General Election: Trump vs. Biden USA Today/Suffolk Biden 50, Trump 40 Biden +10

However, Biden still has coherence problems and a sexual abuse scandal going against him, plus the general resentment among Bernie supporters. And Trump's ability to sell himself and appeal to his base of prejudiced voters remains. It will also depend on how the virus and the economy look by October.
President Trump Job Approval
Disapprove +7.1
Also posted at Leip's Election Atlas, where I composed the maps. This map set should not be confused with other maps that I have posted unless the others most obviously related (those are the same).

You will notice that standing up to the President and contradicting his policies on COVID-19, even if one is in his Party, looks like a very good idea. Such happens when The President fails catastrophically as a leader. Note that Trump's approval ratings even in his 'best' states has come to sink below 60% Trump is in trouble in states in which he has never done well, but more states are becoming troublesome to him. Well worth noting: Republican Governors who take courses different from those of Donald Trump are doing very well.

Approval only of the President's performance on COVID-19 -- not disapproval. Note the difference in the legend. I am treating anything from 46 to 49% approval as effectively a tie. To be sure, COVID-19 is the focus of most Americans' concerns these days as a clear and present danger to ourselves, or at least to people for whom we care. COVID-19 is about as dangerous as any war for which there is conscription, and I assure you -- I am doing everything possible to dodge its draft! This time, draft-dodging is without qualification a noble act!  

[Image: genusmap.php?year=2012&ev_c=1&pv_p=1&ev_...&NE3=0;1;6]

Trump approval:

60% or higher approval

46-49% (white)
41 or 42%
40% or lower

Trump's "best" state is Wyoming, where he gets 59% approval on his handling of COVID-19. As you can see. sixteen states give Trump 40% or lower approval on his handling of COVID-19, which shows Trump in an unflattering light. To be sure, these states have usually been trouble for Trump for other things, but because COVID-19 is the equivalent of war as a clear and present danger, it is the current focus of political life.    

Note that all Governors fare better than Trump in approval in their states (I take that the number for the District of Columbia is for the mayor or the majority of the city council, either Democratic.  

[Image: genusmap.php?year=1964&ev_c=1&pv_p=1&ev_...&NE3=0;1;6]

Approval of State governors

Red  - Democrats;  Blue -  Republicans

The saturation is the tens digit of approval (first digit of an approval number. Thus Kristi Noem (R, SD)  gets a saturation of "4" in the color red for 49% approval and David Ige (D, HI) gets only 39% approval and blue at the saturation level of "3". Those two Governors get the worst approval ratings.    

You will notice that Governors of Maryland, Massachusetts, New Hampshire, and Vermont are Republicans and that Governors of Kansas, Kentucky, Louisiana, Montana, and North Carolina are Democrats, and thus we have the anomalous colors for those states.

CNN reports Lincoln got better press treatment, Trump claims, as he ups pandemic death estimate

I wonder if he considered if Lincoln might have conceivably been the better president?

Not that Lincoln had a field day.  There were partisan papers back then too...
Trump has earned the opprobrium. The media at best try to be objective, but when something or someone is truly horrible the media can't cover for it. There is no sugar-coating a killer earthquake or hurricane. There is no sugar-coating the awfulness of a Trabant or the movie CATS. (I didn't see the movie, but it didn't look promising).

CBS News -- polling of approval and disapproval of the President's handling of COVID-19.
RCP Poll Averages

President Trump Job Approval
Approve 44.9
Disapprove 52.8
Disapprove +7.9

General Election: Trump vs. Biden
Biden 48.4
Trump 43.3
Biden +5.1

Direction of Country
Right Direction 33.0
Wrong Track 59.0
Wrong Track +26.0

2020 Generic Congressional Vote
Democrats 47.7
Republicans 39.7
Democrats +8.0
Pages: 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40