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 General Discussion The DU Lounge All Forums Issue Forums Culture Forums Alliance Forums Region Forums Support Forums Help & Search
13 replies = new reply since forum marked as read
Highlight: NoneDon't highlight anything 5 newestHighlight 5 most recent replies
This seems to be another not-very-good night for pollsters (Original Post) DavidDvorkin Feb 2016 OP
I think the last few elections -- Hell Hath No Fury Feb 2016 #1
What happened to Nate Silvers 82% for Hillary? It's more like 50%/50%. nt 99th_Monkey Feb 2016 #2
82% likelyhood of winning. DCBob Feb 2016 #7
Are ALL (100%) of the votes now in? nt 99th_Monkey Feb 2016 #10
Do you now understand what the 82% means? DCBob Feb 2016 #11
I was never in doubt what it meant 99th_Monkey Feb 2016 #12
He could be corect if she loses. 18% chance of her losing. DesMoinesDem Feb 2016 #13
That was chance of winning instead of results. jeff47 Feb 2016 #8
Okay I am for Bernie all the way, but that's not what he said... ljm2002 Feb 2016 #9
Did any come close? JohnnyLib2 Feb 2016 #3
Most of the decent pollsters are within MoE. (nt) jeff47 Feb 2016 #4
I hope they choke. roguevalley Feb 2016 #5
Most of the reliable pollsters got within their MoE. jeff47 Feb 2016 #6
 

Hell Hath No Fury

(16,327 posts)
1. I think the last few elections --
Tue Feb 2, 2016, 12:20 AM
Feb 2016

have shown that pollsters have yet to adapt to the changes technology have brought to the political landscape.

 

99th_Monkey

(19,326 posts)
12. I was never in doubt what it meant
Tue Feb 2, 2016, 12:39 AM
Feb 2016

but that 'probability' number would assume more than a .2% squeaker that could have
gone either way, and STILL may, in case you didn't know .. they are STILL counting the
freaking votes.

jeff47

(26,549 posts)
8. That was chance of winning instead of results.
Tue Feb 2, 2016, 12:29 AM
Feb 2016

He was predicting she had a 82% chance to win, not that she'd get 82% of the vote. So Nate is "right" even if she wins by 0.1%.

And yes, he is giving a statistic that is designed to mislead.

ljm2002

(10,751 posts)
9. Okay I am for Bernie all the way, but that's not what he said...
Tue Feb 2, 2016, 12:30 AM
Feb 2016

...the 82% was what he thought were the odds of her winning, not what he thought her winning margin would be.

He still doesn't come out looking that good, but then neither do a lot of the pollsters. For which we can all be thankful!

jeff47

(26,549 posts)
6. Most of the reliable pollsters got within their MoE.
Tue Feb 2, 2016, 12:28 AM
Feb 2016

Though there were plenty of robodialing pollsters that blew it.

Latest Discussions»Retired Forums»2016 Postmortem»This seems to be another ...