The Official Obama Approval Ratings Thread

[pbrower2a]

Rasmussen Obama (National)

Approve 51, u (typo fixed)

Disapprove 48%, u.

"Strongly Approve" is at 26%, u.  "Strongly Disapprove" is at 34%, +1.

That is very close to the results of the Presidential election of 2008. 51% approval from Rasmussen may be stronger than 54% from someone else.

PPP is polling Virginia this weekend, and results should be... interesting. The Senate race is a tossup (not a PPP poll, but one for the Washington Post), which suggests that political reality in 2012 may be more like that of 2006 than like 2010.  In 2006 the Democrats won Senate seats despite a lack of available open seats -- all  Democratic Senate gains coming through the defeat of incumbents. To be sure, 2012 will be a Presidential year. 
 

[pbrower2a]

Virginia, before the assassination of Osama bin Laden. Washington Post (49-46). President Obama was going to win Virginia. This was probably after the GOP mucked up on Medicare, Medicaid, and the Ryan Budget (see how I interpret Arizona):
 

Key:


<40% with Disapproval Higher: 40% Orange (50% if 60%-69% or higher disapproval); 90% red if >70%
40-42% with Disapproval Higher: 50% Yellow  
43% to 45% with Disapproval Higher: 40% Yellow  
46-49% with Disapproval Higher: 30% Yellow  
<50% with Approval Equal: 10% Yellow (really white)
<50%  Approval greater: 20% Green
50-55%: 40% Green
56-59%: 60% Green
60%+: 80% Green


Months (All polls are from 2010 or 2011):

A -  January     G -  July
B -  February   H -  August
C -  March        I -  September
D -  April          J  -  October
E -  May           K -  November
F -   June         L -   December

 

S - suspect poll (examples for such a qualification: strange crosstabs, likely inversion of the report (for inversions, only for polls above 55% or below 45%...  let's say Vermont 35% approval or Oklahoma 65% approval), or more than 10% undecided. Anyone who suggests that a poll is suspect must explain why it is suspect.

Partisan polls and polls for special interests (trade associations, labor unions, ethnic associations) are excluded.

Z- no recent poll

Or here:

MY CURRENT PREDICTION OF THE 2012 PRESIDENTIAL ELECTION

(before any campaigning begins in earnest)

assuming no significant changes before early 2012 -- snicker, snicker!

           
deep red                  Obama 10% margin or greater 118
medium red              Obama, 5-9.9% margin   123
pale red                   Obama, margin under 5% 36
white                        too close to call (margin 1% or less) 40
pale blue                  Republican  under 5% 40
medium blue             Republican  5-9.9% margin 32
deep blue                 Republican over 10%   54



44% approval is roughly the break-even  point (50/50) for an incumbent's win.  I add 6% for approval between 40% and 45%, 5% at 46% or 47%, 4% between 48% and 50%, 3% for 51%, 2% for 52% or 53%, 1% for 54% and nothing above 55% or below 40% for an estimate of the vote.

This model applies only to incumbents, who have plenty of advantages but not enough to rescue an unqualified failure.


But --

I have added a yellow category for states in which President Obama defeats all recognized major GOP nominees (so far Huckabee, Romney, Gingrich, Palin, and where available, Thune, Daniels, Christie, and Pawlenty). This will be a yellow category supplanting those in pale blue or and white.

I am also adding a green category for those states that would otherwise be in white, pale pink, or pale blue -- maybe medium blue, as I have seen only one state in that category -- in which who the nominee is matters. This can be rescinded as one of the potential nominees drops out formally or is rendered irrelevant in primaries. I am also adding a deep green color for states in which  only the 'right' nominee has a chance. So far I will label that as "H" for Huckabee or else Obama, "R" for Romney or else Obama, or other initials as appropriate for  anyone else (Gingrich? Daniels? Thune?) should such cases emerge. A tan color is used for a tie.

             
deep red                  Obama 10% margin or greater 118
medium red              Obama, 5-9.9% margin   125
pale red                   Obama, margin under 5% 36
white                        too close to call (margin 1% or less) 3
yellow                        close, but Obama wins against any major Republican candidate  10
Obama wins against all but Huckabee but ties Huckabee 6
Obama wins against all but  Romney 72
close, but Obama wins against a 'blunder' of a nominee 59
pale blue                  Republican  under 5% 6
medium blue             Republican  5-9.9% margin 3
deep blue                 Republican over 10%  54  

[pbrower2a]

Virginia, before after the assassination of Osama bin Laden. Washington Post (49-46) 57-40. President Obama was going to win Virginia even before Osama bin Laden got the opportunity to meet the Great Satan more intimately and permanently.

Because Virginia was one of the states to have felt the worst effects of 9/11, it is easy to figure that the polling effects are stronger there than they would be in such a state as Ohio which had no originating flights no flights scheduled by the airlines to reach it, and no targets hit.  One of the commandeered jetliners flew from Virginia and smashed into the Pentagon, which is in Virginia.  With the arguable exception of New Hampshire (might have had passengers on one of the flights), Virginia is the only State to have ever voted for George W. Bush that felt 9/11 at the personal level.   

A 57% approval for President Obama in Virginia suggests a national landslide -- at least 55% of the popular vote nationwide, and about 400 electoral votes. Projections that show the President behind in Pennsylvania and New Hampshire are likely obsolete.
 

Key:


<40% with Disapproval Higher: 40% Orange (50% if 60%-69% or higher disapproval); 90% red if >70%
40-42% with Disapproval Higher: 50% Yellow  
43% to 45% with Disapproval Higher: 40% Yellow  
46-49% with Disapproval Higher: 30% Yellow  
<50% with Approval Equal: 10% Yellow (really white)
<50%  Approval greater: 20% Green
50-55%: 40% Green
56-59%: 60% Green
60%+: 80% Green


Months (All polls are from 2010 or 2011):

A -  January     G -  July
B -  February   H -  August
C -  March        I -  September
D -  April          J  -  October
E -  May           K -  November
F -   June         L -   December

 

S - suspect poll (examples for such a qualification: strange crosstabs, likely inversion of the report (for inversions, only for polls above 55% or below 45%...  let's say Vermont 35% approval or Oklahoma 65% approval), or more than 10% undecided. Anyone who suggests that a poll is suspect must explain why it is suspect.

Partisan polls and polls for special interests (trade associations, labor unions, ethnic associations) are excluded.

Z- no recent poll

Or here:

MY CURRENT PREDICTION OF THE 2012 PRESIDENTIAL ELECTION

(before any campaigning begins in earnest)

assuming no significant changes before early 2012 -- snicker, snicker!

           
deep red                  Obama 10% margin or greater 131
medium red              Obama, 5-9.9% margin   110
pale red                   Obama, margin under 5% 36
white                        too close to call (margin 1% or less) 40
pale blue                  Republican  under 5% 40
medium blue             Republican  5-9.9% margin 32
deep blue                 Republican over 10%   54



44% approval is roughly the break-even  point (50/50) for an incumbent's win.  I add 6% for approval between 40% and 45%, 5% at 46% or 47%, 4% between 48% and 50%, 3% for 51%, 2% for 52% or 53%, 1% for 54% and nothing above 55% or below 40% for an estimate of the vote.

This model applies only to incumbents, who have plenty of advantages but not enough to rescue an unqualified failure.


But --

I have added a yellow category for states in which President Obama defeats all recognized major GOP nominees (so far Huckabee, Romney, Gingrich, Palin, and where available, Thune, Daniels, Christie, and Pawlenty). This will be a yellow category supplanting those in pale blue or and white.

I am also adding a green category for those states that would otherwise be in white, pale pink, or pale blue -- maybe medium blue, as I have seen only one state in that category -- in which who the nominee is matters. This can be rescinded as one of the potential nominees drops out formally or is rendered irrelevant in primaries. I am also adding a deep green color for states in which  only the 'right' nominee has a chance. So far I will label that as "H" for Huckabee or else Obama, "R" for Romney or else Obama, or other initials as appropriate for  anyone else (Gingrich? Daniels? Thune?) should such cases emerge. A tan color is used for a tie.

             
deep red                  Obama 10% margin or greater 118
medium red              Obama, 5-9.9% margin   125
pale red                   Obama, margin under 5% 36
white                        too close to call (margin 1% or less) 3
yellow                        close, but Obama wins against any major Republican candidate  10
Obama wins against all but Huckabee but ties Huckabee 6
Obama wins against all but  Romney 72
close, but Obama wins against a 'blunder' of a nominee 59
pale blue                  Republican  under 5% 6
medium blue             Republican  5-9.9% margin 3
deep blue                 Republican over 10%  54  

[pbrower2a]

Pbrower, no one in America even knows or cares about the Ryan budget. Please get out of your political bubble for once and look at things objectively.

Maybe not in the Presidential race. For Congressional races -- maybe. Democratic challengers are going to run against it, and in many districts, Republicans are going to try to run from it. ordinarily a politician runs on his record and wins or runs from his record and loses.

In the Presidential election, we all know what event has more immediate influence. The WaPo poll after May 1, 2011 suggests that President Obama would win Virginia by something close to a 57-40 margin (and that is charitable to Republicans). Virginia is close to the national average in voting.

Guess which President was last to win 57% of the popular vote as an incumbent! Reagan won 59%, Nixon 60%, and LBJ 61%.

Dwight Eisenhower, 1956.  1956 was not a pretty year for Democrats. I cannot predict how a 57-43 split of the popular vote would manifest itself in electoral votes, so I can't give a map of such an event. 55-45 likely solidifies Indiana and North Carolina and flips at the least the following states from 2008:

Arizona
Georgia
Missouri
Montana

Ike won 457 electoral votes in 1956 and would have probably won both Alaska and Hawaii if those two states had been voting. Maybe even DC because the Democrats that year had the segregationist vote tied up.
  

[pbrower2a]

Pbrower, no one in America even knows or cares about the Ryan budget. Please get out of your political bubble for once and look at things objectively.

Maybe not in the Presidential race. For Congressional races -- maybe. Democratic challengers are going to run against it, and in many districts, Republicans are going to try to run from it. ordinarily a politician runs on his record and wins or runs from his record and loses.

In the Presidential election, we all know what event has more immediate influence. The WaPo poll after May 1, 2011 suggests that President Obama would win Virginia by something close to a 57-40 margin (and that is charitable to Republicans). Virginia is close to the national average in voting.

Guess which President was last to win 57% of the popular vote as an incumbent! Reagan won 59%, Nixon 60%, and LBJ 61%.

Dwight Eisenhower, 1956.  1956 was not a pretty year for Democrats. I cannot predict how a 57-43 split of the popular vote would manifest itself in electoral votes, so I can't give a map of such an event. 55-45 likely solidifies Indiana and North Carolina and flips at the least the following states from 2008:

Arizona
Georgia
Missouri
Montana

Ike won 457 electoral votes in 1956 and would have probably won both Alaska and Hawaii if those two states had been voting. Maybe even DC because the Democrats that year had the segregationist vote tied up.
  

1956 was a pretty good year for Democrats downballot.

Definately, 1956 was in no way an embrace of the GOP at all. On wikipedia (take it for what its worth) its says that it was primarily a vote for the status quo. They kept the Dems in control of the House and the Senate (which they had gained in 1954) while reelecting Ike in a landslide.

Ike's performance seems to me to be the last hurrah of the ancient Civil War Political map with a lot of his voters coming from legacy voting by people who really were Democrats and voted such on the rest of the ballot.

OK -- it wasn't that bad for down-ticket Democrats. But it was a horrible performance by Adlai Stevenson. Maybe no Democrat had a chance against Ike.

Should approval ratings for the President stay around 55%, then the political map of the 2012 Presidential election  could look almost like an inverse of 1952 or 1956. Tellingly, Democrat Barack Obama won 365 electoral votes -- but won only one state, that barely (North Carolina) that Dwight Eisenhower won in neither 1952 or 1956.   

[pbrower2a]

The new NBC/Wall Street Journal poll backs up Gallup and Quinnipiac:

52% Approve
41% Disapprove

Favorable Rating:

54% Somewhat/Very Positive
31% Somewhat/Very Negative
14% Neutral

If President Obama runs for re-election in the year 2012, do you think you will probably vote for President Obama or probably vote for the Republican candidate?

45% Probably vote for President Obama
30% Probably vote for Republican candidate
  4% Vote for other party
16% Depends on who opponent is

http://msnbcmedia.msn.com/i/MSNBC/Sections/NEWS/A_Politics/_Today_Stories_Teases/11192_MAY_NBC_Poll.pdf

A little math suggests a possible interpretation. Assuming no political or personal scandal, then the absolute worst that President Obama does in 2012 is

45-46-4 (really  47.4 - 48.4 - 4.2), which still might win because the President wins more states with giant EV totals by larger margins while the Republican wins fewer EVs with larger margins.

Such assumes that the GOP has a wonderful candidate capable of getting a convincing message across. Question: who?

At the other side is 61-30-4 (really  64.2 - 31.6 - 4.2), which would be without precedent on the other side. FDR in 1936, LBJ in 1964, and Nixon in 1972 maxed out with about 61% of the vote. It would take a loony, incompetent politician to get trounced that badly.

[pbrower2a]

A rather interesting new poll by EPIC-MRA for Michigan:

Obama: 38% Excellent/Good, 61% Fair/Poor

Snyder: 32% Excellent/Good, 60% Fair/Poor

The poll of 600 voters was conducted April 27-May 3 and has a margin of error of plus or minus 4 percentage points.

http://www.freep.com/article/20110510/COL05/110510020/Poll-60-Michigan-voters-rate-Snyder-s-performance-poor-just-fair

Just a reminder --

"Enhanced Green Fluorescent Protein" polls are not the same as approval polls, so they aren't useful. 

[pbrower2a]

PPP, Virginia.

http://www.publicpolicypolling.com/pdf/PPP_Release_VA_0510424.pdf

You must be logged in to read this quote.

Way down from the euphoric level immediately after the death of Osama bin Laden. Euphoria can disappear fast, especially about someone that many thought 'old news'.  

I'm not going to average this one. All in all, it's still very good. This state can think well of Republicans, including the Governor:

You must be logged in to read this quote.

But Virginians seem to consider the prospective Republican nominees a weak lot:

You must be logged in to read this quote.

Probably not relevant to 2012, but here's a good prospect for 2016:

You must be logged in to read this quote.

Here I discuss someone who might  win against a Democratic nominee much weaker than President Obama -- in 2016. Bob McDonnell is one of the most popular Governors in America irrespective of party. Just look at how much better he fares than do Republican Governors of Iowa, Wisconsin, Michigan, Ohio, Pennsylvania, New Jersey, Maine, Florida, Georgia, and Arizona.
 

Key:


<40% with Disapproval Higher: 40% Orange (50% if 60%-69% or higher disapproval); 90% red if >70%
40-42% with Disapproval Higher: 50% Yellow  
43% to 45% with Disapproval Higher: 40% Yellow  
46-49% with Disapproval Higher: 30% Yellow  
<50% with Approval Equal: 10% Yellow (really white)
<50%  Approval greater: 20% Green
50-55%: 40% Green
56-59%: 60% Green
60%+: 80% Green


Months (All polls are from 2010 or 2011):

A -  January     G -  July
B -  February   H -  August
C -  March        I -  September
D -  April          J  -  October
E -  May           K -  November
F -   June         L -   December

 

S - suspect poll (examples for such a qualification: strange crosstabs, likely inversion of the report (for inversions, only for polls above 55% or below 45%...  let's say Vermont 35% approval or Oklahoma 65% approval), or more than 10% undecided. Anyone who suggests that a poll is suspect must explain why it is suspect.

Partisan polls and polls for special interests (trade associations, labor unions, ethnic associations) are excluded.

Z- no recent poll

Or here:

MY CURRENT PREDICTION OF THE 2012 PRESIDENTIAL ELECTION

(before any campaigning begins in earnest)

assuming no significant changes before early 2012 -- snicker, snicker!

           
deep red                  Obama 10% margin or greater 118
medium red              Obama, 5-9.9% margin   123
pale red                   Obama, margin under 5% 36
white                        too close to call (margin 1% or less) 40
pale blue                  Republican  under 5% 40
medium blue             Republican  5-9.9% margin 32
deep blue                 Republican over 10%   54



44% approval is roughly the break-even  point (50/50) for an incumbent's win.  I add 6% for approval between 40% and 45%, 5% at 46% or 47%, 4% between 48% and 50%, 3% for 51%, 2% for 52% or 53%, 1% for 54% and nothing above 55% or below 40% for an estimate of the vote.

This model applies only to incumbents, who have plenty of advantages but not enough to rescue an unqualified failure.


But --

I have added a yellow category for states in which President Obama defeats all recognized major GOP nominees (so far Huckabee, Romney, Gingrich, Palin, and where available, Thune, Daniels, Christie, and Pawlenty). This will be a yellow category supplanting those in pale blue or and white.

I am also adding a green category for those states that would otherwise be in white, pale pink, or pale blue -- maybe medium blue, as I have seen only one state in that category -- in which who the nominee is matters. This can be rescinded as one of the potential nominees drops out formally or is rendered irrelevant in primaries. I am also adding a deep green color for states in which  only the 'right' nominee has a chance. So far I will label that as "H" for Huckabee or else Obama, "R" for Romney or else Obama, or other initials as appropriate for  anyone else (Gingrich? Daniels? Thune?) should such cases emerge. A tan color is used for a tie.

             
deep red                  Obama 10% margin or greater 118
medium red              Obama, 5-9.9% margin   123
pale red                   Obama, margin under 5% 36
white                        too close to call (margin 1% or less) 3
yellow                        close, but Obama wins against any major Republican candidate  10
Obama wins against all but Huckabee but ties Huckabee 6
Obama wins against all but  Romney 72
close, but Obama wins against a 'blunder' of a nominee 59
pale blue                  Republican  under 5% 6
medium blue             Republican  5-9.9% margin 3
deep blue                 Republican over 10%  54  

[pbrower2a]

Gallup's Weekly total is up again.  Basically a 7-Point increase versus last week, with basically every group gaining.

The Political subgroups are interesting though, He's still below 50% with Indies, mostly because the bump has come mostly from Republicans (10% to 21%).  Liberals and Democrats showed only a 2 and 3-point gain respectively, but this reflect an actual bounce tempered by a handful of "OMG AMERICAN EMPIRE" Liberals going the other way.

Also us Youngens saw a 13-Point approval jump, though this might be due to an usually low total for Obama last week (46%).  All other age groups went up 5 or 6.

Geographically, Obama went up 9 in the East and West, 7 in the South, and 5 in the Midwest.

There's not really any racial differences, though Obama's still off his highs with black voters (though they still overwhelmingly approve).

http://www.gallup.com/poll/124922/Presidential-Approval-Center.aspx

Though Gallup's other poll (Obama vs Generic R) Poll shows statistically insignificant movement in Obama's direction, going from tied to Obama + 3, suggesting the bulk of Obama's approval increase has come from people who probably wouldn't vote for him anyway.

http://www.gallup.com/poll/147500/Obama-Approval-Bump-Hasnt-Transferred-2012-Prospects.aspx

I checked the pattern for approval of Presidents Reagan, Clinton, and Obama... and they look so similar that the difference looks like random noise.

Republicans who approve of the assassination of Osama bin Laden are not likely to maintain approval of the President. 

[pbrower2a]

TX (University of Texas)

Approve: 46%
Disapprove: 45%
Not sure: 9%

http://www.laits.utexas.edu/txp_media/html/poll/files/201105-summary.pdf

OMGZ ! Enter pbrower in 3,2,1: Texas will be an important swing state in 2012 !!!

I can't find the poll, and the report comes from an infamous prankster. 

[pbrower2a]

Missouri,  PPP

You must be logged in to read this quote.

Key:


<40% with Disapproval Higher: 40% Orange (50% if 60%-69% or higher disapproval); 90% red if >70%
40-42% with Disapproval Higher: 50% Yellow  
43% to 45% with Disapproval Higher: 40% Yellow  
46-49% with Disapproval Higher: 30% Yellow  
<50% with Approval Equal: 10% Yellow (really white)
<50%  Approval greater: 20% Green
50-55%: 40% Green
56-59%: 60% Green
60%+: 80% Green


Months (All polls are from 2010 or 2011):

A -  January     G -  July
B -  February   H -  August
C -  March        I -  September
D -  April          J  -  October
E -  May           K -  November
F -   June         L -   December

 

S - suspect poll (examples for such a qualification: strange crosstabs, likely inversion of the report (for inversions, only for polls above 55% or below 45%...  let's say Vermont 35% approval or Oklahoma 65% approval), or more than 10% undecided. Anyone who suggests that a poll is suspect must explain why it is suspect.

Partisan polls and polls for special interests (trade associations, labor unions, ethnic associations) are excluded.

Z- no recent poll

Or here:

MY CURRENT PREDICTION OF THE 2012 PRESIDENTIAL ELECTION

(before any campaigning begins in earnest)

assuming no significant changes before early 2012 -- snicker, snicker!

           
deep red                  Obama 10% margin or greater 118
medium red              Obama, 5-9.9% margin   123
pale red                   Obama, margin under 5% 36
white                        too close to call (margin 1% or less) 40
pale blue                  Republican  under 5% 40
medium blue             Republican  5-9.9% margin 32
deep blue                 Republican over 10%   54



44% approval is roughly the break-even  point (50/50) for an incumbent's win.  I add 6% for approval between 40% and 45%, 5% at 46% or 47%, 4% between 48% and 50%, 3% for 51%, 2% for 52% or 53%, 1% for 54% and nothing above 55% or below 40% for an estimate of the vote.

This model applies only to incumbents, who have plenty of advantages but not enough to rescue an unqualified failure.


But --

I have added a yellow category for states in which President Obama defeats all recognized major GOP nominees (so far Huckabee, Romney, Gingrich, Palin, and where available, Thune, Daniels, Christie, and Pawlenty). This will be a yellow category supplanting those in pale blue or and white.

I am also adding a green category for those states that would otherwise be in white, pale pink, or pale blue -- maybe medium blue, as I have seen only one state in that category -- in which who the nominee is matters. This can be rescinded as one of the potential nominees drops out formally or is rendered irrelevant in primaries. I am also adding a deep green color for states in which  only the 'right' nominee has a chance. So far I will label that as "H" for Huckabee or else Obama, "R" for Romney or else Obama, or other initials as appropriate for  anyone else (Gingrich? Daniels? Thune?) should such cases emerge. A tan color is used for a tie.

             
deep red                  Obama 10% margin or greater 118
medium red              Obama, 5-9.9% margin   123
pale red                   Obama, margin under 5% 36
white                        too close to call (margin 1% or less) 3
yellow                        close, but Obama wins against any major Republican candidate  10
Obama wins against all but Huckabee but ties Huckabee 6
Obama wins against all but  Romney 72
close, but Obama wins against a 'blunder' of a nominee 59
pale blue                  Republican  under 5% 6
medium blue             Republican  5-9.9% margin 3
deep blue                 Republican over 10%  54  

[pbrower2a]

Please do not include the Zogby poll, pbrower.

It's a favorable poll, not an approval poll.

Removed.

[pbrower2a]

With the announcement by Mike Huckabee that he will not be running for President, I can make a subtle change for one state (Iowa). Some subtle changes in wording also become possible, as anyone other than Romney now seems a blunder:
 

Key:


<40% with Disapproval Higher: 40% Orange (50% if 60%-69% or higher disapproval); 90% red if >70%
40-42% with Disapproval Higher: 50% Yellow  
43% to 45% with Disapproval Higher: 40% Yellow  
46-49% with Disapproval Higher: 30% Yellow  
<50% with Approval Equal: 10% Yellow (really white)
<50%  Approval greater: 20% Green
50-55%: 40% Green
56-59%: 60% Green
60%+: 80% Green


Months (All polls are from 2010 or 2011):

A -  January     G -  July
B -  February   H -  August
C -  March        I -  September
D -  April          J  -  October
E -  May           K -  November
F -   June         L -   December

 

S - suspect poll (examples for such a qualification: strange crosstabs, likely inversion of the report (for inversions, only for polls above 55% or below 45%...  let's say Vermont 35% approval or Oklahoma 65% approval), or more than 10% undecided. Anyone who suggests that a poll is suspect must explain why it is suspect.

Partisan polls and polls for special interests (trade associations, labor unions, ethnic associations) are excluded.

Z- no recent poll

Or here:

MY CURRENT PREDICTION OF THE 2012 PRESIDENTIAL ELECTION

(before any campaigning begins in earnest)

assuming no significant changes before early 2012 -- snicker, snicker!

           
deep red                  Obama 10% margin or greater 118
medium red              Obama, 5-9.9% margin   123
pale red                   Obama, margin under 5% 36
white                        too close to call (margin 1% or less) 40
pale blue                  Republican  under 5% 40
medium blue             Republican  5-9.9% margin 32
deep blue                 Republican over 10%   54



44% approval is roughly the break-even  point (50/50) for an incumbent's win.  I add 6% for approval between 40% and 45%, 5% at 46% or 47%, 4% between 48% and 50%, 3% for 51%, 2% for 52% or 53%, 1% for 54% and nothing above 55% or below 40% for an estimate of the vote.

This model applies only to incumbents, who have plenty of advantages but not enough to rescue an unqualified failure.


But --

I have added a yellow category for states in which President Obama defeats all recognized major GOP nominees (so far Huckabee, Romney, Gingrich, Palin, and where available, Thune, Daniels, Christie, and Pawlenty). This will be a yellow category supplanting those in pale blue or and white.

I am also adding a green category for those states that would otherwise be in white, pale pink, or pale blue -- maybe medium blue, as I have seen only one state in that category -- in which who the nominee is matters. This can be rescinded as one of the potential nominees drops out formally or is rendered irrelevant in primaries. I am also adding a deep green color for states in which  only the 'right' nominee has a chance. So far I will label that as "H" for Huckabee or else Obama, "R" for Romney or else Obama, or other initials as appropriate for  anyone else (Gingrich? Daniels? Thune?) should such cases emerge. A tan color is used for a tie.

             
deep red                  Obama 10% margin or greater 118
medium red              Obama, 5-9.9% margin   123
pale red                   Obama, margin under 5% 42
white                        too close to call (margin 1% or less) 3
yellow                        close, but Obama wins against any major Republican candidate  10
Obama wins against all but Huckabee but ties Huckabee 6
Obama wins against all but  Romney 72
close, but Obama wins against a 'blunder' of a nominee anyone other than Romney 59
pale blue                  Republican  under 5% 6
medium blue             Republican  5-9.9% margin 3
deep blue                 Republican over 10%  54  

[pbrower2a]

Gallup's at 48-45. Down 2 from Friday, likely a negative sample. We'll see if it goes back up later today.

The euphoria over the demise of Osama bin Laden was short-lived. Good call, but it's the economy now.

[pbrower2a]

Hawaii checks in once again:



(Gov. Neil Abercrombie)

50% Approve
36% Disapprove

The poll of 614 registered voters was conducted by Ward Research for Hawaii News Now and the Honolulu Star-Advertiser.

http://www.hawaiinewsnow.com/story/14649236/hawaii-poll-half-of-island-voters-approve-of-governor-in-first-six-months-of-term

It changes nothing in my Presidential projection, so I have no new map.

[pbrower2a]

Michigan (EPIC-MRA)

53% (+9) Favorable
40%  (-9) Unfavorable

49% (+11) Excellent/Good
50%  (-11) Fair/Poor

The poll of 600 likely 2012 voters from across the state was taken from May 9-11. It was a margin of error of 4 percentage points.

http://www.freep.com/article/20110516/NEWS15/110516056/Obama-approval-up-Michigan-after-bin-Laden-s-death

Again, EGFP  and favorable/unfavorable both fail to fit the parameters of this thread.

[pbrower2a]

Michigan (EPIC-MRA)

53% (+9) Favorable
40%  (-9) Unfavorable

49% (+11) Excellent/Good
50%  (-11) Fair/Poor

The poll of 600 likely 2012 voters from across the state was taken from May 9-11. It was a margin of error of 4 percentage points.

http://www.freep.com/article/20110516/NEWS15/110516056/Obama-approval-up-Michigan-after-bin-Laden-s-death

Again, EGFP  and favorable/unfavorable both fail to fit the parameters of this thread.

And your repeated posting of general election matchup polling and 5-colored maps modeling Obama's reelection chances does "fit the parameters of this thread"?

I got sharp criticism for using EGFP and "favorability" polls.

If you don't like my five-color maps... tough! 

[pbrower2a]

Well, I agree that you shouldn't put those on your maps.  But this thread isn't just about your maps.

It seems rather silly of you to criticize Tender's posting of E/G/F/P polls in this thread, considering the amount of space you devote to items here that are only tenuously connected to Obama approval ratings.

1. I don't pretend to be neutral on contemporary politics -- including the near-future.

2. I am trying to predict the future upon objective evidence -- and that means reliable polls.

3. Tender Branson himself has told me to remove "favorability" and EGFP polls. I remind people why I don't include them in case we have a newbie who doesn't go through hundreds of posts.

"Favorability" is essentially that one likes the politician even if one thinks him a bumbler, as in "He's a nice guy, but his economic policies are hurting us". "EGFP" includes "fair", which is ambiguous. "He's doing a fair job" can suggest that one is performing adequately, but not remarkably well, as in "not bad for what I expected".  

4.  Anyone can post a poll even if it is blatantly biased. Someone even posted a fabricated poll within the last couple of weeks (it was ostensibly by a University of Texas entity and showed a ridiculously-high level of approval for the President).  I have stated in my methodology (which needs to be repeated because it isn't self-evident) that polls by organizations with blatant bias (political parties and their fronts, political campaigns, unions, trade organizations, ethnic associations, extremist  groups) don't figure in my model.  If I use a poll by the NAACP (which does much good), do I also need to include polls collected by a Klan group?

I checked the link for what proved a bogus poll, and I removed a post in which I had used the data. "Trust but verify" is good practice.  

Likewise I saw a news release of polls supposedly by Marist -- ordinarily a good poster, but one that has been in hibernation in recent months in issuing statewide polling results. The news source was the once-respected UPI that has now fallen into the control of the Unification Church, the cult of Reverend Sun Myung Moon. The poster of that news release said that he wanted to see it from Marist itself. The official release by Marist never came, so maybe the news story was a fake.
 

[pbrower2a]

Interesting poll by Rasmussen:

http://www.rasmussenreports.com/public_content/lifestyle/general_lifestyle/may_2011/50_give_government_positive_marks_for_response_to_weather_disasters

 

You must be logged in to read this quote.

Wording of questions:

You must be logged in to read this quote.

This may as much be a response to state governors, largely in the South, as to the President and the Federal government. It looks better than the response to Hurricane Katrina, though.

[pbrower2a]

pbrower, you misunderstand me.  You're right to exclude EGFP polls from your maps, but the thread isn't about your maps.  "EGFP  and favorable/unfavorable both fail to fit the parameters of this thread" is a silly comment.  They fail to fit the parameters of your *maps*, but they do fit the parameters of this thread.

One is free to post any poll, no matter how questionable the source or  shaky its reliability.  I haven't jumped on obvious typos, which so far seem most of the causes for question.

Sure, this thread is not about my maps. My maps, I hope, serve as a device for memory and analysis.  A year from now I might not post them for the simple reason that someone else will have a more professional approach. Here is that  person's website:

http://electoral-vote.com/

It has been dormant since December 2010, when the last House seat was decided. It will surely be back in operation perhaps as early as January 2012, and when it shows Presidential matchups I will abandon  this effort quickly.

[pbrower2a]

Maine (Critical Insights):

52% Favorable
33% Unfavorable

(Gov. LePage)

31% Approve
54% Disapprove

For the current wave of the study, Critical Insights completed a total of 600 telephone interviews with randomly selected voters across the state between May 5-9, 2011.

http://www.criticalinsights.com/assets/CriticalInsightsTrackingSurveySpring2011.pdf

I am using the gubernatorial poll in my "Governors' approval" map, but I can;t use the favorability poll on my Presidential approval map. Hint: the Governor of Maine is political poison, and I question whether Senator Olympia Snowe can win both re0nomination and re-election as a Republican.

[pbrower2a]

Ohio (Quinnipiac)

49% Approve
45% Disapprove

Looking ahead to the 2012 election for President... If the 2012 election for President were being held today, do you think you would vote for Barack Obama the Democratic candidate, or for the Republican candidate?

41% Obama
39% Republican

Looking ahead to the 2012 election for President... Do you feel that Barack Obama deserves to be reelected, or do you feel that he does not deserve to be reelected?

47% Yes
47% No

From May 10 - 16, Quinnipiac University surveyed 1,379 registered voters with a margin of error of +/- 2.6 percentage points. Live interviewers call land lines and cell phones.

http://www.quinnipiac.edu/x1322.xml?ReleaseID=1602

FOX News Poll:

55% Approve
41% Disapprove

The Fox News Poll is conducted under the joint direction of Anderson Robbins Research (D) and Shaw & Company Research (R). The poll is based on live telephone interviews with a national sample of 910 registered voters, and was conducted May 15-17, 2011 in the evenings.

http://www.foxnews.com/projects/pdf/051811_bin_Laden_web.pdf

New Jersey (Monmouth)

Adults:

60% Approve, 35% Disapprove

Registered Voters:

60% Approve, 36% Disapprove

The Monmouth University/NJ Press Media Poll was conducted by telephone with 807 New Jersey adults from May 12 to 16, 2011. This sample has a margin of error of + 3.5 percent. The poll was conducted by the Monmouth University Polling Institute and originally published by the NJ Press Media newspaper group (Asbury Park Press, Courier-Post, Courier News, Daily Journal, Daily Record, and Home News Tribune).

http://www.monmouth.edu/polling/admin/polls/MUP39_2.pdf

North Carolina (Civitas)

51% Approve
45% Disapprove

(Gov. Perdue)

46% Approve
41% Disapprove

This poll of 600 registered general election voters in North Carolina was conducted May 10-11, 2011 by National Research, Inc. of Holmdel, NJ.  All respondents were part of a fully representative sample of registered voters in North Carolina.  For purposes of this study, voters interviewed had to have voted in two of the past four general elections or were newly registered to vote since 2008.

http://www.nccivitas.org/2011/civitas-poll-perdue-job-approval-remains-under-50-percent

New Jersey, North Carolina, and Ohio updates.


 

Key:


<40% with Disapproval Higher: 40% Orange (50% if 60%-69% or higher disapproval); 90% red if >70%
40-42% with Disapproval Higher: 50% Yellow  
43% to 45% with Disapproval Higher: 40% Yellow  
46-49% with Disapproval Higher: 30% Yellow  
<50% with Approval Equal: 10% Yellow (really white)
<50%  Approval greater: 20% Green
50-55%: 40% Green
56-59%: 60% Green
60%+: 80% Green


Months (All polls are from 2010 or 2011):

A -  January     G -  July
B -  February   H -  August
C -  March        I -  September
D -  April          J  -  October
E -  May           K -  November
F -   June         L -   December

 

S - suspect poll (examples for such a qualification: strange crosstabs, likely inversion of the report (for inversions, only for polls above 55% or below 45%...  let's say Vermont 35% approval or Oklahoma 65% approval), or more than 10% undecided. Anyone who suggests that a poll is suspect must explain why it is suspect.

Partisan polls and polls for special interests (trade associations, labor unions, ethnic associations) are excluded.

Z- no recent poll

Or here:

MY CURRENT PREDICTION OF THE 2012 PRESIDENTIAL ELECTION

(before any campaigning begins in earnest)

assuming no significant changes before early 2012 -- snicker, snicker!

           
deep red                  Obama 10% margin or greater 132
medium red              Obama, 5-9.9% margin   109
pale red                   Obama, margin under 5% 36
white                        too close to call (margin 1% or less) 40
pale blue                  Republican  under 5% 40
medium blue             Republican  5-9.9% margin 32
deep blue                 Republican over 10%   54



44% approval is roughly the break-even  point (50/50) for an incumbent's win.  I add 6% for approval between 40% and 45%, 5% at 46% or 47%, 4% between 48% and 50%, 3% for 51%, 2% for 52% or 53%, 1% for 54% and nothing above 55% or below 40% for an estimate of the vote.

This model applies only to incumbents, who have plenty of advantages but not enough to rescue an unqualified failure.


But --

I have added a yellow category for states in which President Obama defeats all recognized major GOP nominees (so far Huckabee, Romney, Gingrich, Palin, and where available, Thune, Daniels, Christie, and Pawlenty). This will be a yellow category supplanting those in pale blue or and white.

I am also adding a green category for those states that would otherwise be in white, pale pink, or pale blue -- maybe medium blue, as I have seen only one state in that category -- in which who the nominee is matters. This can be rescinded as one of the potential nominees drops out formally or is rendered irrelevant in primaries. I am also adding a deep green color for states in which  only the 'right' nominee has a chance. So far I will label that as "H" for Huckabee or else Obama, "R" for Romney or else Obama, or other initials as appropriate for  anyone else (Gingrich? Daniels? Thune?) should such cases emerge. A tan color is used for a tie.

             
deep red                  Obama 10% margin or greater 132
medium red              Obama, 5-9.9% margin   109
pale red                   Obama, margin under 5% 42
white                        too close to call (margin 1% or less) 3
yellow                        close, but Obama wins against any major Republican candidate  10
Obama wins against all but  Romney 72
close, but Obama wins against anyone other than Romney 59
pale blue                  Republican  under 5% 6
medium blue             Republican  5-9.9% margin 3
deep blue                 Republican over 10%  54  

[pbrower2a]

Pbrower, light green and dark green mean the same thing on your map. 

You are right on the bottom map, and thanks for catching it. It is conceivable, though, that someone might bring some unforeseen strengths as a possible Republican nominee. Daniels? Huntsman? Pawlenty?

At the least, Florida and Pennsylvania seem to get polled very often. Light green seems to mean that Palin automatically loses and that perhaps Gingrich does. Trump didn't figure in with any matches against President Obama.

[pbrower2a]

Now for the upper left corner (WA, PPP), in case Republicans have hope that the state might be drifting their way.

You must be logged in to read this quote.

Naw.


 

Key:


<40% with Disapproval Higher: 40% Orange (50% if 60%-69% or higher disapproval); 90% red if >70%
40-42% with Disapproval Higher: 50% Yellow  
43% to 45% with Disapproval Higher: 40% Yellow  
46-49% with Disapproval Higher: 30% Yellow  
<50% with Approval Equal: 10% Yellow (really white)
<50%  Approval greater: 20% Green
50-55%: 40% Green
56-59%: 60% Green
60%+: 80% Green


Months (All polls are from 2010 or 2011):

A -  January     G -  July
B -  February   H -  August
C -  March        I -  September
D -  April          J  -  October
E -  May           K -  November
F -   June         L -   December

 

S - suspect poll (examples for such a qualification: strange crosstabs, likely inversion of the report (for inversions, only for polls above 55% or below 45%...  let's say Vermont 35% approval or Oklahoma 65% approval), or more than 10% undecided. Anyone who suggests that a poll is suspect must explain why it is suspect.

Partisan polls and polls for special interests (trade associations, labor unions, ethnic associations) are excluded.

Z- no recent poll

Or here:

MY CURRENT PREDICTION OF THE 2012 PRESIDENTIAL ELECTION

(before any campaigning begins in earnest)

assuming no significant changes before early 2012 -- snicker, snicker!

           
deep red                  Obama 10% margin or greater 144
medium red              Obama, 5-9.9% margin   97
pale red                   Obama, margin under 5% 36
white                        too close to call (margin 1% or less) 40
pale blue                  Republican  under 5% 40
medium blue             Republican  5-9.9% margin 32
deep blue                 Republican over 10%   54



44% approval is roughly the break-even  point (50/50) for an incumbent's win.  I add 6% for approval between 40% and 45%, 5% at 46% or 47%, 4% between 48% and 50%, 3% for 51%, 2% for 52% or 53%, 1% for 54% and nothing above 55% or below 40% for an estimate of the vote.

This model applies only to incumbents, who have plenty of advantages but not enough to rescue an unqualified failure.


But --

I have added a yellow category for states in which President Obama defeats all recognized major GOP nominees (so far Huckabee, Romney, Gingrich, Palin, and where available, Thune, Daniels, Christie, and Pawlenty). This will be a yellow category supplanting those in pale blue or and white.

I am also adding a green category for those states that would otherwise be in white, pale pink, or pale blue -- maybe medium blue, as I have seen only one state in that category -- in which who the nominee is matters. This can be rescinded as one of the potential nominees drops out formally or is rendered irrelevant in primaries. I am also adding a deep green color for states in which  only the 'right' nominee has a chance. So far I will label that as "H" for Huckabee or else Obama, "R" for Romney or else Obama, or other initials as appropriate for  anyone else (Gingrich? Daniels? Thune?) should such cases emerge. A tan color is used for a tie.

             
deep red                  Obama 10% margin or greater 144
medium red              Obama, 5-9.9% margin   97
pale red                   Obama, margin under 5% 42
white                        too close to call (margin 1% or less) 3
yellow                        close, but Obama wins against any major Republican candidate  10
Obama wins against all but  Romney 72
close, but Obama wins against someone other than Romney 59
pale blue                  Republican  under 5% 6
medium blue             Republican  5-9.9% margin 3
deep blue                 Republican over 10%  54  

[pbrower2a]

SurveyUSA May Polls:

California: 56% Approve, 40% Disapprove (+10, -10)

http://www.surveyusa.com/client/PollTrack.aspx?g=c717d146-99e7-4de8-8217-9fb792fc2fb8

Kansas: 42% Approve, 54% Disapprove (+6, -7)

http://www.surveyusa.com/client/PollTrack.aspx?g=c9cb9d95-0451-467c-8613-efb0bf5d4a7e

Oregon: 46% Approve, 49% Disapprove (-1, nc)

http://www.surveyusa.com/client/PollTrack.aspx?g=87dabdb7-5435-402e-891e-11b3bf475127

Washington: 50% Approve, 47% Disapprove (+1, nc)

http://www.surveyusa.com/client/PollTrack.aspx?g=c34262fe-7152-4e56-be8d-dd7e32b0f6dd

Yeah, Obama has a 33-58 approval among Oregon Independents ... LOL !

SurveyUSA polls can be interesting in their own way. They are the only polls that I ever see of Kansas... and the May poll of that state is fascinating. For so conservative a state as Kansas , a 42% approval of the President looks very good. President Obama lost the state by 15%, but not since 1992 has any Democratic nominee been as close as 5% to the Republican nominee. Of course, I figure that Ross Perot was picking off lots of conservative voters in 1992. Gore lost the state by 20% and Kerry by 26% in close elections.

I can't see President Obama winning Kansas except against a lunatic (he'd probably get at most 48% of the vote against someone like Romney or Pawlenty now that Huckabee is out), but I can imagine a House seat or two shifting there with a somewhat-sane Republican as a Presidential nominee. Two here, two there, one here, three there -- it all adds up in the House.

Because nobody else polls Kansas I might be tempted to post the SurveyUSA poll for Kansas on my map -- but that is it. The rest look like either "me-too" efforts or otherwise spurious. 

The last poll that I had for Kansas was SurveyUSA poll in December, and if i thought that one reasonable, I see no cause to believe this one unreasonable.



 

Key:


<40% with Disapproval Higher: 40% Orange (50% if 60%-69% or higher disapproval); 90% red if >70%
40-42% with Disapproval Higher: 50% Yellow  
43% to 45% with Disapproval Higher: 40% Yellow  
46-49% with Disapproval Higher: 30% Yellow  
<50% with Approval Equal: 10% Yellow (really white)
<50%  Approval greater: 20% Green
50-55%: 40% Green
56-59%: 60% Green
60%+: 80% Green


Months (All polls are from 2010 or 2011):

A -  January     G -  July
B -  February   H -  August
C -  March        I -  September
D -  April          J  -  October
E -  May           K -  November
F -   June         L -   December

 

S - suspect poll (examples for such a qualification: strange crosstabs, likely inversion of the report (for inversions, only for polls above 55% or below 45%...  let's say Vermont 35% approval or Oklahoma 65% approval), or more than 10% undecided. Anyone who suggests that a poll is suspect must explain why it is suspect.

Partisan polls and polls for special interests (trade associations, labor unions, ethnic associations) are excluded.

Z- no recent poll

Or here:

MY CURRENT PREDICTION OF THE 2012 PRESIDENTIAL ELECTION

(before any campaigning begins in earnest)

assuming no significant changes before early 2012 -- snicker, snicker!

           
deep red                  Obama 10% margin or greater 144
medium red              Obama, 5-9.9% margin   97
pale red                   Obama, margin under 5% 36
white                        too close to call (margin 1% or less) 40
pale blue                  Republican  under 5% 46
medium blue             Republican  5-9.9% margin 32
deep blue                 Republican over 10%   48



44% approval is roughly the break-even  point (50/50) for an incumbent's win.  I add 6% for approval between 40% and 45%, 5% at 46% or 47%, 4% between 48% and 50%, 3% for 51%, 2% for 52% or 53%, 1% for 54% and nothing above 55% or below 40% for an estimate of the vote.

This model applies only to incumbents, who have plenty of advantages but not enough to rescue an unqualified failure.


But --

I have added a yellow category for states in which President Obama defeats all recognized major GOP nominees (so far Huckabee, Romney, Gingrich, Palin, and where available, Thune, Daniels, Christie, and Pawlenty). This will be a yellow category supplanting those in pale blue or and white.

I am also adding a green category for those states that would otherwise be in white, pale pink, or pale blue -- maybe medium blue, as I have seen only one state in that category -- in which who the nominee is matters. This can be rescinded as one of the potential nominees drops out formally or is rendered irrelevant in primaries. I am also adding a deep green color for states in which  only the 'right' nominee has a chance. So far I will label that as "H" for Huckabee or else Obama, "R" for Romney or else Obama, or other initials as appropriate for  anyone else (Gingrich? Daniels? Thune?) should such cases emerge. A tan color is used for a tie.

             
deep red                  Obama 10% margin or greater 144
medium red              Obama, 5-9.9% margin   97
pale red                   Obama, margin under 5% 42
white                        too close to call (margin 1% or less) 3
yellow                        close, but Obama wins against any major Republican candidate  10
Obama wins against all but  Romney 72
close, but Obama wins against someone other than Romney 59
pale blue                  Republican  under 5% 12
medium blue             Republican  5-9.9% margin 3
deep blue                 Republican over 10%  48