Welcome to DU! The truly grassroots left-of-center political community where regular people, not algorithms, drive the discussions and set the standards. Join the community: Create a free account Support DU (and get rid of ads!): Become a Star Member Latest Breaking News Editorials & Other Articles General Discussion The DU Lounge All Forums Issue Forums Culture Forums Alliance Forums Region Forums Support Forums Help & Search

BzaDem

(11,142 posts)
25. Polls are often wrong in the same direction. State polls underestimated Obama by 2.5% on average.
Wed Dec 14, 2016, 12:43 AM
Dec 2016

That's precisely why Nate Silver was less wrong than all the other models this year. He took into account that polling error is often correlated between states (in addition to the high number of undecideds and third party voters). This makes sense -- if the polls are less likely to reach a group that ends up voting slightly more heavily in one state, the same is likely to be true in other states.

Exit polls are often off by even more than that. See President Kerry. Most people will not talk to exit pollsters, and differential non-response bias (and non-representative precinct selection) will skew the toplines of the exit polls (which is why even the exit pollsters say not to take the toplines seriously).

This should be the narrative Jean-Jacques Roussea Dec 2016 #1
because it's the truth. nt LaydeeBug Dec 2016 #2
Actually, it is off by quite a bit. Coyotl Dec 2016 #39
your last sentence is so SPOT ON, it's haunting. nt LaydeeBug Dec 2016 #46
How could the vote be hacked MichMary Dec 2016 #3
Central tabulators are. triron Dec 2016 #4
But-- MichMary Dec 2016 #5
OK if you are in such disbelief, read this (brief) triron Dec 2016 #6
No. They did not do hand recounts, they merely used optical scanners. You wouldn't need an "army" synergie Dec 2016 #33
Because there are apparently a lot of people who just can't believe dionysus Dec 2016 #41
If people spend all their time, energy, and money .... Yurovsky Dec 2016 #42
Thats the attitude i like! This "it's the end of the world our lives are ruined1!!1eleventy!" dionysus Dec 2016 #44
Please don't ask awkward questions like that mythology Dec 2016 #7
Also this triron Dec 2016 #8
And this triron Dec 2016 #9
Thank you for these links. bookmarked. nt LaydeeBug Dec 2016 #11
And this triron Dec 2016 #10
do a few searches on the subject, and you will see. it is totally possible, in multiple ways, and TheFrenchRazor Dec 2016 #20
The disks and cards that they are loaded with could be hacked. synergie Dec 2016 #32
there... are... four... lights 0rganism Dec 2016 #12
The polls were MORE WRONG in 2012. Were you complaining then? BzaDem Dec 2016 #13
You are referring only to national polls triron Dec 2016 #14
Same thing with swing state polls. BzaDem Dec 2016 #15
All off in the same direction (or at least almost all)? triron Dec 2016 #16
Polls are often wrong in the same direction. State polls underestimated Obama by 2.5% on average. BzaDem Dec 2016 #25
I was referring triron Dec 2016 #26
Exit polls had Gore winning Alabama and Georgia, and Bill winning Indiana and Texas. BzaDem Dec 2016 #27
You have a link or reference to this data? triron Dec 2016 #28
btw Richard Charnin triron Dec 2016 #30
Here is a link triron Dec 2016 #29
States the exits said Gore won that he lost: AL AR AZ CO FL GA MO NC NV TN TX VA BzaDem Dec 2016 #31
I will grant that some exit polls may be wrong triron Dec 2016 #37
The statistical argument doesn't work without a perfect random sample. BzaDem Dec 2016 #45
They've been doing this incrementally for years now. Brownback in Kansas, even McTurtle LaydeeBug Dec 2016 #36
all the statistical models this year were WAY off Fast Walker 52 Dec 2016 #17
That is a story pushed by MSM as well. triron Dec 2016 #18
um, because it's TRUE Fast Walker 52 Dec 2016 #19
um, no it's not LaydeeBug Dec 2016 #21
you're *joking*, right? LaydeeBug Dec 2016 #22
Are you saying that the polls were less wrong in 2012 than in 2016? BzaDem Dec 2016 #24
No. I am saying the election was rigged this time...why conflate two cycles? LaydeeBug Dec 2016 #38
I think you're missing the point SickOfTheOnePct Dec 2016 #47
Oh it's entirely possible LaydeeBug Dec 2016 #48
LOL SickOfTheOnePct Dec 2016 #49
Funny how that goes. nt LaydeeBug Dec 2016 #50
knr again triron Dec 2016 #23
Wow, looks like I might be psychic. nt LaydeeBug Dec 2016 #34
see post 47 in http://www.democraticunderground.com/12512622637 triron Dec 2016 #35
As for optical scanners and tabulators, wikipedia has this: triron Dec 2016 #40
triron, you are an inspiration on this...THANK YOU!!!!! nt LaydeeBug Dec 2016 #43
Latest Discussions»Retired Forums»2016 Postmortem»Oh, the ones who count th...»Reply #25