Trump approval ratings thread 1.6

[pbrower2a]

Polls after Labor Day should be interesting. That's all that I say as a prediction. 

[pbrower2a]

https://twitter.com/ppppolls/status/1300464455575310337

Georgia in contention, Trump is in trouble.

In view of a recent post in which I got estimates of Trump winning certain states, I don't need to keep any "seat-of-the-pants" estimates of a Trump win. I might give an update around October 1 based upon matchups of the time. I see no good model for two months away from the election... yet.

Trump approval 55% or higher
Trump approval 50-54%
Trump approval positive but under 50%
ties are in white
Trump approval negative but disapproval under 50%
Trump disapproval 50-54%
Trump disapproval 55% or higher

[pbrower2a]

About a month from now, a different line will apply to President Trump's chances of getting re-elected:

Time to election  |1 point|5 points||10 points|20 points|
one day............. |...64%|....95%|.....99.7%|.99.999%|
one week........... |...60%|....89%|.......98%|...99.97%|
one month......... |...57%|....81%|.......95%|.....99.7%|
three months..... |...55%|....72%|.......87%|........98%|
six months..........|...53%|....66%|.......79%|.......93%|
one year.............|....52%|...59%|.......67%|.......81%|

From three months out to one month out, leads in the 5-to-10% range give 8% or 9% increases in the likelihood of a win for a leader. On the other sides, a 1% margin goes from giving a 55% chance of winning to a 57% chance, which of course reminds us of the margin of error; a 20% edge gives the leader a likelihood  going from 98% to nearly 100%. 

[pbrower2a]

39, 41, and 42 approval... the difference is slight. Any one of these is execrable. Trump needed to be at 42% approval ten months ago (that is, a year before the election) to have so much as a 25% chance of winning. I would guess that he needs at least 45% approval to have a chance if everything goes right for him, and the only thing that will go right for him will be that Democrats can run up the vote totals in a few medium-to-giant states (especially California, Illinois, and New York) and get a decisive plurality of the popular vote nationwide and break less-than-even in the Electoral College.

Disapproval at 54 or 55 means that Trump will need to convince people now against him to change their minds. That will be a tougher sell than getting me, who knows the rules of probability, to play craps in a casino. (Craps is an honest game unless someone introduces "Chicago dice" as "Big Julie" did in Guys and Dolls, but the house has an edge and slowly but inevitably cleans out the gamblers of they play long enough).

Polling is remarkably stable, and the bounces between Parties following their conventions seem to have been adequate for Biden but not for Trump.

     

[pbrower2a]

I am ready to redo my seat-of-the-pants estimates of Biden and Trump chances based on match-ups alone.  I would need to do some interpolations, and at this I take the dangers of interpolation (much less dangerous than interpolation. Obviously 50-50 is 50% for both.

State data is from here:


https://www.cbsnews.com/news/joe-biden-donald-trump-opinion-poll-08-16-2020/

August 10.


three months..... |...55%|....72%|.......87%|........98%|

lead  likelihood

0   50    10  87
1   55    11  88
2   59    12  89
3   64    13  90
4   69    14  91
5   72    15  92
6   76    16  93
7   80    17  94
8   83    18  95
9   85    19  96

The interpolation is nearly linear, and that may be inadequate for small leads. This model suggests that even a 3-point edge for Biden at this point (late August) is far from trivial.

Here is a map of the probabilities of a Biden win based upon the edge that one or the other has. Numbers are not electoral votes this time: Data is from August 10, so convention bumps do not appear:

 

Biden likelihood 0 to 9 (saturation 8 )
Biden likelihood 10 to 19 (saturation 6)
Biden likelihood 20 to 29 (saturation 4)
Biden likelihood 30 to 39 (saturation 4)
Biden likelihood 41 to 49 (saturation 2)
white  -- tie, exactly even

Biden likelihood 51 to 59 (saturation 2)
Biden likelihood 60 to 69 (saturation 4)
Biden likelihood 70 to 79 (saturation 5)
Biden likelihood 80 to 89 (saturation 6)
Biden likelihood 90 or higher (saturation 8 )

It is already two weeks obsolete, because the numbers are based on Senators winning elections based upon their leads at three months (I have done linear interpolation). This is also the last polling data that both

(1) comes from all 50 states,
(2) from the same time, and
(3) from the same source.  

Senatorial and gubernatorial elections for statewide contests for electoral votes by Presidential campaigns. This may be far from a perfect model. Biden has an 80% chance of winning Wisconsin, which this map shows as the most likely tipping-point state. He also has a 69% chance of winning North Carolina and Florida (each), a 64% chance of winning Arizona, a 50% chance of winning Ohio, and a 41% chance of winning Texas .  These six states are dissimilar enough that they could as well be considered independent events. Trump has about seven chances in 1000 to win all six states in question. (I am not considering Iowa, as Biden is not winning Iowa while losing Wisconsin).

At this point, Biden is trying to consolidate the states that he needs or might need. Trump is trying to keep his hope alive in states in which he has as little as a 20% chance of winning.

Another way of putting it: Trump has about as much chance of winning Wisconsin as he does of losing Kansas.

[pbrower2a]

Another way of putting it: Trump has about as much chance of winning Wisconsin as Biden hedoes of losing Kansas.

This does not compute.  You are saying that Trump's chances in Wisconsin and Kansas are the same.

Correction noted and made.

[pbrower2a]

Civiqs - thru 9/2

National: 41/56 (-15)

Alaska: 43/55 (-12)
Arizona: 43/53 (-10)
Colorado: 38/59 (-22)
Florida: 45/53 (-8)
Georgia: 44/53 (-9)
Iowa: 46/52 (-6)
Kansas: 52/45 (+7)
Maine: 37/60 (-23)
Michigan: 41/56 (-15)
Minnesota: 39/58 (-19)
Montana: 47/50 (-3)
Nevada: 36/60 (-24)
New Hampshire: 37/58 (-21)
New Mexico: 43/54 (-11)
North Carolina: 45/53 (-8)
Pennsylvania: 42/55 (-13)
South Carolina: 50/48 (+2)
Texas: 48/49 (-1)
Virginia: 38/59 (-21)
Wisconsin: 43/54 (-11)

https://civiqs.com/results/approve_president_trump?uncertainty=true&annotations=true&zoomIn=true

Trump approval 55% or higher
Trump approval 50-54%
Trump approval positive but under 50%
ties are in white
Trump approval negative but disapproval under 50%
Trump disapproval 50-54%
Trump disapproval 55% or higher

[pbrower2a]

https://twitter.com/Politics_Polls/status/1302483498951901185

New Mexico was on the fringe of competitiveness in 2016.

Approval 39/56, so average this with the Civiqs number. Either way... if Trump had any idea that New Mexico would give him its five electoral votes in 2020... well, that chance looks to be as accessible as if it were in a black hole. 

Trump approval 55% or higher
Trump approval 50-54%
Trump approval positive but under 50%
ties are in white
Trump approval negative but disapproval under 50%
Trump disapproval 50-54%
Trump disapproval 55% or higher

[pbrower2a]

CNN in two swing states.

Trump approval
NC: 46/51
WI: 43/55

Trump approval 55% or higher
Trump approval 50-54%
Trump approval positive but under 50%
ties are in white
Trump approval negative but disapproval under 50%
Trump disapproval 50-54%
Trump disapproval 55% or higher

How does Trump win with such horrible numbers on approval?

Delaware, DEE CEE, Illinois, Oregon, and Rhode Island are not shown, but at this point Trump has disapproval numbers (figuring that Delaware, DC, Illinois, Oregon, and Rhode Island) give Trump disapproval numbers of 55% or higher, then Biden has a very strong grasp on 272 electoral votes. This is more electoral failure by Trump than winning by Biden,  but that will not matter in November. 

[pbrower2a]

Defaming America's soldiers and veterans is political suicide for someone on the Left -- legitimately so. What is so bad about Trump is that he exposes a contempt for soldiers that might manifest in a willingness to sacrifice them for his glory.

[pbrower2a]

https://twitter.com/stella2020woof/status/1304541236473344001

Translate the approval into votes for and disapproval as votes against, and one comes close to Mondale 41%, Reagan 58% in 1984. That may be imprecise, but it is closer than anything else in the last forty years. It is relevant. 

[pbrower2a]

At this point, approval and disapproval measures matter far less than do the current match-ups. Trump has had plenty of time in which to improve his image as a leader and has failed. What suggests that he will magically do better in the last one-and-a-half months (roughly fifty days) to win?

The disapproval measures have suggested his current estimates of results in match-ups. Unless disapproval numbers have as their focus events that have become less pressing (which explains why Iowa went from looking like an electoral disaster for during the trade war that hurt farmers a couple years ago to having a Trump edge), disapproval indicates that people have given up on him.  Trump may have his base intact, but that will not be enough. Goldwater and McGovern energized their bases, too.

I see fewer statewide approval and disapproval numbers now -- or perhaps I am not looking as hard now.       

[pbrower2a]

Guess you're not big on the 538 model then...which seems to be charitable to Trump to be honest but I believe in the methodology so I go with it.

What a bunch of hacks.

You tell them they got 2016 wrong and they throw a hissy fit because “that’s not how models work”, they scream. Yet they celebrate when they get it ‘right.’ You can’t have it both ways.

Additionally, they don’t even bother hiding their strong liberal bias - and because the bias isn’t removed, it not only shows up in their written analysis, but also in how they construct their models and statistics.

The guy is so arrogant, he attempts to ‘correct’ and ‘adjust’ polls. Give me a break.

So no, I don’t trust 538 who had H Clinton at 88% chance to win the 2016 election, in mid October 2016 - less than one month from the election. What a joke.

What they basically said after the election: Donald Trump was a long-shot, and long-shots occasionally win. 11-1 chances win about one time in twelve. Some people like betting on such long-shots, and race-track operators offer such bets. Heck, some people bet on 200-1 long-shots.  That is what an 88% chance means.

Polls are snapshots, and they may not reflect reality for long. Events can happen. Maybe this year some prominent pol gets exposed in a picture with him and Jeffrey Epstein... and an underage girl. I'm not saying that such will happen. His chances of getting re-elected suddenly slip from 88% to 8%. Earlier polling would not show that. (Illustration chosen for its blatant qualities and not for likelihood, by the way).

Last time I got a sinking feeling when Donald Trump successfully brought Paula Jones into the debate... a sinking feeling that says, "prepare for the worst four years of your life". I had lost both parents and my life savings (that is how nursing homes operate). I was broke and in grief at the time of the 2016 election. I am no happier. The Trump Presidency is grief in itself.

Make America Great Again? The prospect of spending the rest of my life doing a job that pays minimum wage, has harsh management, and depends that one smile all the time is almost enough to make me want to... 

[pbrower2a]

CNN in two swing states.

Trump approval
NC: 46/51
WI: 43/55

Trump approval 55% or higher
Trump approval 50-54%
Trump approval positive but under 50%
ties are in white
Trump approval negative but disapproval under 50%
Trump disapproval 50-54%
Trump disapproval 55% or higher

How does Trump win with such horrible numbers on approval?

Delaware, DEE CEE, Illinois, Oregon, and Rhode Island are not shown, but at this point Trump has disapproval numbers (figuring that Delaware, DC, Illinois, Oregon, and Rhode Island) give Trump disapproval numbers of 55% or higher, then Biden has a very strong grasp on 272 electoral votes. This is more electoral failure by Trump than winning by Biden,  but that will not matter in November. 

[pbrower2a]

TRUMP JOB APPROVAL

When asked how President Trump is handling his job as president, likely voters had varying opinions across the three states.
In Kentucky: 55 percent approve, 41 percent disapprove.
In South Carolina: 50 percent approve, 47 percent disapprove.
In Maine: 38 percent approve, 60 percent disapprove.
TRUMP CORONAVIRUS RESPONSE

When asked how President Trump is handling the response to the coronavirus, likely voters gave similar responses to his overall job approval.
In Kentucky: 54 percent approve, 43 percent disapprove.
In South Carolina: 49 percent approve, 48 percent disapprove.
In Maine: 37 percent approve, 61 percent disapprove.

(Quinnipiac)

Trump approval 55% or higher
Trump approval 50-54%
Trump approval positive but under 50%
ties are in white
Trump approval negative but disapproval under 50%
Trump disapproval 50-54%
Trump disapproval 55% or higher

How does Trump win with such horrible numbers on approval?

Delaware, DEE CEE, Illinois, Oregon, and Rhode Island are not shown, but at this point Trump has disapproval numbers (figuring that Delaware, DC, Illinois, Oregon, and Rhode Island) give Trump disapproval numbers of 55% or higher, then Biden has a very strong grasp on 272 electoral votes. This is more electoral failure by Trump than winning by Biden,  but that will not matter in November. 

[/quote]

[pbrower2a]

Approval and disapproval ratings got us to the match-up numbers. Elections are match-up events, and not simply referenda on approval and disapproval. I expect approval and disapproval numbers to largely disappear from polling as more concern becomes "who votes".

Very soon, polls will have four categories:

HV (have voted)
LV  (likely voters)
RV  (registered voters)
A (adults)

Adults who have not yet registered to vote may be running out of time in which to register to vote. 

Some votes are already in. Mine, for example, I can't see anything that Donald Trump could do that would convince me that he will cause me to question my vote. (Yes, I am a Party hawk... but this time it is easy to vote for a devout Christian with a conservative lifestyle who supports American troops and promotes rational thought as a necessary feature of any good solution for anything. In many years that was a Republican. Not this year.

HV (Have Voted) is the most solid part of the vote. This will include some "likely voters"... and people who might have seemed "not-so-likely voters". For me the voting part of 2020 is over. I will be involved in other activity.

If you want to see some episodic indicator that things are going badly for Trump, I took a trip to Democratic headquarters in my county and someone came in for some Biden yard signs, That person had said that he had never voted Democratic in his life and was voting Democratic this time. That is one person... 

The voting has begun, and some signs look bad for Republicans.

[pbrower2a]

Civiqs, 50-state approval poll

Trump approval:

55 or higher (saturation 7)
50-54 (saturation 5)
under 50 but positive (saturation 2)
exact tie (white)
45-50 but negative (saturation 2)
40-44 (saturation 5)
under 40 (saturation 7)

New Mexico (46-51)  is something of a surprise, but this poll relies heavily upon Internet access which may be lacking among the large number of very poor people, heavily First Peoples or Mexican-American, who would likely vote D in the state.

Can anyone tell how Donald Trump wins (fraud excluded, and I have no desire to give or receive any unsolicited suggestions except to forward those to the Federal Bureau of Investigation) with numbers like these? I can about as easily see the Detroit Tigers winning the Super Bowl this year (the Detroit Tigers are a baseball team). At this point, matchup numbers are closer to definitive. Matchup numbers are consequences of approval and disapproval numbers.

Addendum: the 42-55 split on approval is tied for the worst for this President. 

[pbrower2a]

Vermont: Trump approval 30-64

http://projects.vpr.org/vpr-vermont-pbs-2020-polls-september

No surprise here. We don't see many polls of Vermont.

Trump approval 55% or higher
Trump approval 50-54%
Trump approval positive but under 50%
ties are in white
Trump approval negative but disapproval under 50%
Trump disapproval 50-54%
Trump disapproval 55% or higher

How does Trump win with such horrible numbers on approval?

Delaware, DEE CEE, Illinois, Oregon, and Rhode Island are not shown, but at this point Trump has disapproval numbers (figuring that Delaware, DC, Illinois, Oregon, and Rhode Island) give Trump disapproval numbers of 55% or higher, then Biden has a very strong grasp on 272 electoral votes. This is more electoral failure by Trump than winning by Biden,  but that will not matter in November. 

[pbrower2a]

Because match-ups are more important, I am paying more attention to those. Approval polls got us to the matchups that we have.

[pbrower2a]

Because match-ups are more important, I am paying more attention to those. Approval polls got us to the matchups that we have.

Approval ratings in Michigan and Wisconsin, Marist for NBC:

MI 43/52 (-9)
WI 42/54 (-12)

Whitmer approval: 56/40 (+16)
Evers approval: 50/43 (+7)

Favorabilities:
Trump: 40/56 in MI, 40/58 in WI
Biden: 47/46 in MI, 50/46 in WI

Result:

MI
Biden 52%, Trump 44% (Among RVs, Biden 52%, Trump 43%)

WI
Biden 54%, Trump 44% (Among RVs, Biden 52%, Trump 44%)

9:05 AM · Sep 27, 2020

The prospective vote for Trump in both states is roughly his approval rating, and the prospective vote for Biden is roughly Trump's disapproval numbers.

[pbrower2a]

Why hasn't pbower2A has been putting up any approval rating maps, he has been quiet as of late. I think it's due to Barrett, the SCOTUS vacancy irregardless what happens will cement a Conservative majority on the Crt for perhaps until 2032, as long as Alito and Thomas are there.

Biden even he gets in, will be hampered by a 6/5 or 6/3 Conservative majority, that was the point I was making thru these threads.

Citizens United probably stays until then too, Unions when RBG died tried to get donations for all these candidates like Espy and McGrath and HEGAR whom have 0 chance and they are obstructing the stimulus bill. Bob Casey, Manchin, and Sinema have broke with Pelosi on 300 instead of 600 in a 7.9 percent unemployment

For much of the Trump era, approval numbers have been portents of the Presidential election. What does matter is the match-ups, and those seem to be close to approval numbers in state and national numbers. Trump is extremely polarizing; people love him or hate him as President. I noticed in two statewide poll that showed approval numbers that Trump's share of the vote is likely to be close to his approval rating and Biden's vote is likely to be close to Trump's disapproval rating.

Most significantly, time is running out for President Trump (at least for winning re-election), hospitalization or not. I'm not going to speculate on his chances of recovery, let alone his ability to wage a spirited (if perverse) campaign. Trump needs a spirited and competent campaign to have a chance of winning, which was just as true eleven months ago as it is now. What has changed is that time is running out. Football teams come back from 17-point deficits in the first quarter all the time at all levels of the game.  NFL teams win (I would guess) about 95% of all teams in which they have 17-point leads at the start of the fourth quarter  (equivalent to early August this year) and win over 99% of the time with five minutes left in the game and a 17-point lead.  Electoral politics, like football, is a timed sport, and even small leads become decisive. As in football there are strategies for teams with 17-point leads in the first quarter to kill opportunities for an opposing team to make a comeback. It's easy to say that two touchdowns and a field goal can cause such a lead to evaporate, and I have seen that happen often. On the other hand, the team with the lead can chance its offense to a more conservative running game that grinds down the clock and put five defensive backs (the "nickel defense") in the backfield to make long pass plays ineffective and risky for the offense. The team behind needs two more touchdowns to win than does the team with the lead... because nobody gets six field goals in a game.  

There is much that I cannot say. I am not in the prophecy business. The closest that I get is in using probabilities lifted and perhaps modified from others with caveats on applicability.      

[pbrower2a]

I posted this on page 1 of this thread as an amendment, and this is the basis for my model for the "short game":

...how important are leads with time? Close to Election Day, electoral leads of even 1% can give the leader nearly 2/3 of a chance of winning the state. Leads that may not look 'that bad' for the nominee behind in polling can go from troubling to ominous to politically lethal over a year even if the lead remains the same.

  I just got my hands on Nate Silver's The Signal and the Noise (why so many predictions but some don't)It relates probability well, and as I have suggested, being up 5% in a binary election a year before means little, being up 5% a month before the election is huge. It is from 2012, and it relates much other than elections (like sports, poker, and even chess). What it says of electoral leads as a campaign approaches its conclusion is telling.

On page 63, Figure 2-4 shows the probability of a Senate candidate winning (1998 to 2008) with a certain lead (1, 5, 10, and 20 points) at one year, six months, three months, one month, one week, and one day. Because statewide races for President are much like statewide races for the Senate -- with the qualification that Presidential nominees do not usually make appearances where they see themselves losing -- unless they really are losing nationwide.

Time to election  |1 point|5 points||10 points|20 points|
one day............. |...64%|....95%|.....99.7%|.99.999%|
one week........... |...60%|....89%|.......98%|...99.97%|
one month......... |...57%|....81%|.......95%|.....99.7%|
three months..... |...55%|....72%|.......87%|........98%|
six months..........|...53%|....66%|.......79%|.......93%|
one year.............|....52%|...59%|.......67%|.......81%|

(I am going to put this back in my "electoral theory" section because it will remain relevant.

So what conclusions can I draw? You might be surprised that a five-point lead one month before Election Day is no less significant than a twenty-point lead one year before the election. Thus one hears things like Democrats saying "We have a chance of winning West Virginia if everything goes right" and Republicans say that they have a chance of winning Massachusetts... yadda, yadda, yadda. Or is it "Yabba, dabba, doo!" Likewise, being one point ahead on the day before the election is worth almost as much as being five points ahead six months before the election or even ten points ahead  a year before the election.

(I expect to put this or something like it in the upcoming "2022 Elections" and "2024 Elections" threads that will be of interest after this election winds down)

Caveats:

1. This is about Senate races. Senate races, like contests for the Electors of any one state, are statewide elections. There are simply more of them than are Presidential elections. I would expect more noise about Presidential races.

2. Presidential campaigns can triage themselves out of races that they see as sure wins or sure losses. Presidential nominees who have big leads can make campaign trips to support Senate candidates in states in which they expect themselves to lose to support a Senate candidate. A Senate nominee cannot triage himself out of a race that he is likely to lose.

3. I am going on individual states for Presidential races.

4. I am rejecting obvious hack polls such as those of TRAshy FALsehood GARbage* that attempt to keep the spirits up of those supporting a nearly-sure loser. I also reject those insider polls that seem advocacy for one side. Insider polls are useful for saying that a state rarely polled isn't going to go for a certain nominee. I'll take an insider poll that tells me what just about anyone can reasonably expect such as that Donald Trump is going to win South Dakota or that Joe Biden is going to win Maryland. 

5. Outliers happen, and they can say things about events that people who do not know the local political climate may see happening that the rest of us don't. Maybe some challenger has an unusually well-run and well-funded campaign against someone beloved elsewhere but struggling in his state. Many liberals could not see Russ Feingold going down to defeat in 2010, but they did not recognize that the Republican Party and front groups had opened the spigots. "Money makes ze vorld go around, ze vorld go around, ze vorld go around" (from the cynical musical and movie Cabaret). 

*TRAshy FALsehood GARbage? Someone tweeted an internal document stating that they could give their clients the results that they want as a special service.

[pbrower2a]

Trump's hideous approval ratings have gotten him here:

We are one month away from Election Day, so the line in boldface applies.

Time to election  |1 point|5 points||10 points|20 points|
one day............. |...64%|....95%|.....99.7%|.99.999%|
one week........... |...60%|....89%|.......98%|...99.97%|
one month......... |...57%|....81%|.......95%|.....99.7%|
three months..... |...55%|....72%|.......87%|........98%|
six months..........|...53%|....66%|.......79%|.......93%|
one year.............|....52%|...59%|.......67%|.......81%|

Things are likely worse for Trump in states in which he has gone behind by 10% or more. But sticking to this model I can interpolate on a assumption that the probability rises in a linear fashion, which may itself be an understatement in much of the range. Arbitrarily I am going to treat any 15% lead one month away as giving either nominee a 99%+ chance of winning, and I am not going to make distinctions above 99%.

Lead	%win
00	50
01	57
02	63
03	69
04	75
05	81
06	85
07	89
08	91
09	93
10	95
11	96
12	97
13	98
14	99
15+	99+

So based solely on the latest credible poll of the last week or so, based on the chance of Biden winning:

AL: Auburn University: Trump +20 (any surprise?)
NYT/Siena: AZ - Biden+8
DE-UDEL: Biden +21 (any surprise?)
NM-New Mexico Political Report/PPP: Biden +14
PPP-NC: Biden +4
ND: DFM Research: Trump +14
PA: Ipsos: Biden +5
UT: Y2 Analytics: Trump +10
WI: Ipsos: Biden +6
CBS: Tied in OH; Biden +7% in PA
MO - MO Scout/Remington: Trump+5 (Biden has an outside chance of winning a state that hasn't been close since 2008)
NJ - DKC Analytics/Brach Eichler: Biden +14%
NYT/Siena: Biden +7% in PA,
NYT/Siena: Biden +5% in FL
AZ - Strategies 360/Smart and Safe Act: Biden +4% (marijuana advocacy)
GA-Redfield & Wilton: Biden +1
Hart Research (D): Biden leads GA (taken, but +2%) and NC (Biden +2) , Trump OH, IA and TX(all Trump +2)
New York (Siena): Biden +32
Redfield & Wilton: Biden +9 in MI (before debate, others taken)
Redfield & Wilton: Biden +5 in WI (before debate)
VA - Daniel Gade internal (R) - Biden +10% (I have seen worse for Trump in Virginia)
NH-Emerson: Biden +7
Data for Progress: Biden -5 in SC (others taken)
OR-Civiqs/Daily Kos: Biden +17
KS-Civiqs/Daily Kos: Trump +10
CA: Survey USA: Biden +27
IL-Victory Research: Biden +13
ND-DFM Research: Trump +19
MPR News/Star Tribune/KARE-11 Minnesota Poll: Biden 48 Trump 42

The states below rarely get polled, so some of the polls may be old.

MD-OpinionWorks: Biden +32
OK-Amber Integrated: Trump +22
AK-Harstad Strategic Research/Independent Alaska (I/D): Trump +1 (that's Alaska)
Fox News: Biden+11 in NV
VT - Braun Research: Biden +23%
WA (Strategies 360): Biden +22
ME: Suffolk - Biden +12
NYTimes/Siena - Montana: Trump+7
NE-02 - Global Strategy Group (D): Biden +6%
UT-Hinckley Institute/Scott Rasmussen: Trump +18
CO-Global Strategy Group/Progress Now Colorado (D): Biden +11

Chance:
95% or higher for the leader saturation 8
85-94% for the leader saturation 6
70-84% for the leader saturation 5
60-69% for the leader saturation 4
51-59% for the leader saturation 2

exact tie white

[pbrower2a]

There are caveats to this election that can still go Prez Trump's way, 7.9 percent unemployment, as 661K workers go back to work, and a 1.6T stimulus package, if it's passed Trump can get credit, if it's not passed Pelosi can be viewed as an obstructionist, Bob Casey, Manchin, Sinema and blue dogs in the House have broken with Pelosi on her 600, unemployment extension.  Also, the Rs are gonna confirm Barrett by election day which can help his polls with women that Pence doesn't appeal to

Trump is he underdog, but he isn't at 39 percent or 41 percent, Sabato said middle class voters swing back and forth as they did in 2016, and Trump can still win, it's called the silent majority

A large part of the American electorate was willing to make great sacrifices to defeat COVID-19.  They could tolerate having their favorite bars, restaurants, and libraries closed for a while. Had the whole of America done as  Michigan did, then we would probably be opening up by now. Masks and social distancing did far more to stall COVID-19 than did medicine.

At this point I can't even say that President Trump will serve a full term. He did get COVID-19, and he is an obvious candidate for death from it. This election could easily decide between Biden and Pence instead of between Biden and Trump. 

[pbrower2a]

Pbower 2A has barely been on recently, he still can post in the threads and why are you answering for him?

Why isn't pbower2A posting that much and his approval maps.

During the long gap between elections, approval ratings are a fairly good substitute for head-to-head polls.  But now that we're close to the election, we have the real thing; there are plenty of head-to-head polls, which are better than using approval ratings.

Because I am predictable.

Typically approval and disapproval numbers give a fairly good estimate of the upcoming vote. They were stable over a surprisingly-long time (basically between the impeachment effort and the Party conventions).

Trump has his base, and he never was especially far from the sort of vote characteristic of a large-scale loser. Trump got just under 46% of the popular vote; he simply got the "right" votes and may not be so charmed this year. Hillary Clinton ran up popular vote totals in a few states predictably D (like CA, WA, NY, MA, NJ, and MD) but the vote was close in a bunch of states in which Trump got an edge.

I thought well before the election that the best predictor of a vote total in any one state was 100-DIS because Trump is the incumbent this time. I interpreted "disapproval" as meaning "I will vote for someone else who is adequately reasonable". If Trump disapproval in a state was 55%. then he had practically no chance of getting much more than 45% of the vote in that state. In a binary election, such is losing. That is how things worked for Dubya in 2004 and Obama in 2008, and I did not see any reason to see any other dynamic in operation in 2020.

"Disapproval" can mean anything from "I will not vote for him because he is inadequate" to "I'd like to watch him experience the same fate as Saddam Hussein (hanging)". I'm not going to make a distinction in that range because the electoral process makes no such distinction.