News for hackers. Stuff that matters.
Re:Nate Silver’s Lessons for Big Data from the Unpredicted T... by ZeroGravitas on Friday January 20, 2017 @03:43PM (#13446833)
No it didn't, it overestimated the popular vote too much in his direction and was off by about 5 points. Most polls were better than that.

http://thehill.com/blogs/pundits-blog/campaign/315145-one-la...

Since the 538 stuff is mostly a summary of all of those other, better polls, they got the popular vote pretty much bang on. Their modal of how that translated to the electoral college was better than any other model too.

Except for the "Donald Trump will win the popular vote, therefore he'll definitely win the electoral college, oops he didn't win the popular vote, but let's just pretend we're polling the electoral college instead" model.

Re:Nate Silver’s Lessons for Big Data from the Unpredicted T... by NumberSix on Friday January 20, 2017 @04:57PM (#13447297)
USC Dornsife picked the winner. Other polls did not.

Trump got 46.1 percent of the popular vote:

http://www.cnn.com/2016/12/21/politics/donald-trump-hillary-...

That is almost the same as the 46.8 percent of the popular vote for Trump that USC Dornsife predicted. Pretty close to spot on.

The other polls, including 538, underestimated the size of Trump's support.

Re:Nate Silver’s Lessons for Big Data from the Unpredicted T... by ZeroGravitas on Friday January 20, 2017 @07:17PM (#13448027)
It had Clinton a) losing the popular vote, and b) at 43.6%, a prediction of -3.2 points behind Trump versus a reality of +2.1 to Clinton, giving a total miss of 5.3% when other polls got much closer and got the popular vote winner correct.

538 did not conduct polls, only aggregate other people's. And as I said, those polls, and 538's aggregate of them correctly predicted Trump's support, including the fact that it was less than Clinton's.

Re:Nate Silver’s Lessons for Big Data from the Unpredicted T... by jakeogh on Saturday January 21, 2017 @12:44AM (#13449006)
"only aggregate other people"... and apply a "weight" to each. Not that he shouldn't, most of the polls were push polls. I got a real kick (and a nice check from predictit) out of the AZ polls showing H ahead by using a ~20% dem over-sample.
Re:Nate Silver’s Lessons for Big Data from the Unpredicted T... by ZeroGravitas on Saturday January 21, 2017 @08:08AM (#13449887)
Yes, they weight the polls, and in doing so got a very accurate results for the popular vote.

Individual state polls had more variance than the national polls, but those were wrong in both directions, both over and understating Trump's support, mostly depending on the number of less educated white people in the state, who were apparently hard to poll, or to estimate their likelihood to actually vote.

Both you and the other poster seem to be trying to create a narrative that Trump's popular support was underestimated, when really it's just the interaction of where his support was and the electoral college system that fooled most models, though notably not the 538 model, which gave Trump a decent shot at winning. Btw, you almost certainly don't know what an "oversample" is, you should read up on that, as you use the phrase as if it is some devious way to skew the polls in one direction. Actually, while I'm on the topic, you seem to have got the wrong end of the stick about what a push poll is too. Those are devious, but do the really apply here. Just a Wikipedia summary should be enough to set you on the right path.

Re:Nate Silver’s Lessons for Big Data from the Unpredicted T... by jakeogh on Saturday January 21, 2017 @09:10AM (#13450077)
"accurate results for the popular vote" is irrelevant for good reasons.
Re:Nate Silver’s Lessons for Big Data from the Unpredicted T... by ZeroGravitas on Saturday January 21, 2017 @10:33AM (#13450467)
I think you're in the wrong thread, this one is about who "did a significantly better job than Nate Silver and 538 in estimating the level of Donald Trump support", not who would win the electoral college.
Re:Nate Silver’s Lessons for Big Data from the Unpredicted T... by jakeogh on Saturday January 21, 2017 @10:37AM (#13450484)
Funny, because that's the only level of support that matters. 538 got it wrong where it counts.
Re:Nate Silver’s Lessons for Big Data from the Unpredicted T... by jakeogh on Friday January 20, 2017 @06:10PM (#13447714)
Yep. Nate also had Bernie at 1% for Dem MI primary.