2016 Postmortem
Related: About this forumLOL @Gallup
Either they are right or almost every other national and state pollster is wrong. The results are simply irreconcilable. The law of large numbers, the Central Limit Theorem , and Gaussian Distribution suggest they are wrong.
If they aren't wrong from a purely statistical sense they are certainly on the outside of how you would expect polls to perform given the margin of error.
If there are any statisticians here I welcome your input.
Larkspur
(12,804 posts)If so, it will have a 3-7 day lag time.
DemocratSinceBirth
(99,714 posts)But there's lots of odd findings. Such as suggesting there is a eight point deficit between registered voters and likley voters where the president is leading by two among the former and trailing by six among the latter.
Jennicut
(25,415 posts)But that is even crazy considering Obama has big leads in RV's in every other poll that measures that. Gallup says Romney's 48% in RV's adds 4 extra points to his LV's. With Obama, he loses one point from LV's. In the Wash Post/ABC poll RV's were Obama 50 and Romney 43.
DemocratSinceBirth
(99,714 posts)If they are right every other pollster is wrong because the others show the narrowest of leads for both candidates.
Maximumnegro
(1,134 posts)Romney's entire advantage in this poll comes from a massive lead in the South. Now sure, some of that may be Florida, but the state-level polling certainly doesn't show that. So Romney is driving up big margins in Texas, Alabama, Oklahoma, Mississippi and other such presidentially irrelevant states? Good for him! I'm sure that'll be cold comfort as he loses the states that actually matter in the Midwest and West.
demwing
(16,916 posts)If the polling is factual, other pollsters should be picking it up as well
Maximumnegro
(1,134 posts)He's just saying that's what gives Romney the ridiculous +6 in Gallup. No matter what Gallup's LV model has been found wanting. Thay are playing games.
DemocratSinceBirth
(99,714 posts)No need to unskew it.
Either they are right or everybody else is wrong.
DemocratSinceBirth
(99,714 posts)Looking at Gallup's own crosstabs it's highly likely that President Obama would win the Electoral College while losing the popular vote by six or seven percent. Remember, they haven't assigned the undecideds.
There has never been a gap between the popular vote and the Electoral College vote nearly as large in our nation's history.
Conclusion. The entire model is flawed.
Maximumnegro
(1,134 posts)Jennicut
(25,415 posts)And that popular vote was extremely close. 47.9% to 48.4%
Gallup's #'s with Obama winning the electoral college are not mathematically possible. Which means their #'s in the East, Midwest and West are wrong and all state polls are wrong. Which is laughable.
Blue Owl
(50,514 posts)n/t
woolldog
(8,791 posts)I'm worried about the Gallup numbers.
progressivebydesign
(19,458 posts)DemocratSinceBirth
(99,714 posts)I am just flummoxed that as large as institution as they are with the concomitant resources available to it can be so incompetent.
woolldog
(8,791 posts)they did?
Tell me more, link?
DemocratSinceBirth
(99,714 posts)Shouldn't they have just said our super duper poll suggested Ford would win by 1% and he lost by 2% thus indicating we were off by 3% and predicted the wrong winner.
They also had Bush beating Gore in the pop vote 48%-46%
(They weaseled out of that great prediction by saying they picked the right eventual winner. I don't remember their poll claiming to predict the Electoral College winner)
And in 010 they showed the Republicans having a 15% generic ballot advantage. Their actual advantage was 6.4%. That's a 250% difference....