How Nate Silver Missed Donald Trump (user search)
       |           

Welcome, Guest. Please login or register.
Did you miss your activation email?
May 16, 2024, 11:32:00 AM
News: Election Simulator 2.0 Released. Senate/Gubernatorial maps, proportional electoral votes, and more - Read more

  Talk Elections
  Election Archive
  Election Archive
  2016 U.S. Presidential Election
  How Nate Silver Missed Donald Trump (search mode)
Pages: [1]
Author Topic: How Nate Silver Missed Donald Trump  (Read 3555 times)
emailking
Atlas Icon
*****
Posts: 14,580
« on: January 26, 2016, 10:40:17 AM »

Yes, I know, Lief is going to get really hot and bothered about this article. So be it. The problem is that Silver is using multiple ingredients for his prognostications, rather than just data, just like ordinary mortals do without black boxes. And he seems quite likely to have been wrong. The establishment has proved surprisingly toothless, at least so far. Sure, in the end, events may salvage Silver, but for the moment, he's just been wrong, just like to pick somebody at random, a forumite here who goes by the screen name "Torie."  Even the best of us can be wrong, so be merciful please, with malice towards none, and charity for all. Thank you.

Nate has never said Trump won't be the nominee. He has said to be cautious about simply believing early polls at face value. His model runs have predicted non-trivial chances that Trump will be the nominee. So what do you mean when you say he's "wrong"? That he thought something was unlikely which now appears more likely than it once was? Because anybody could have told you that would happen. There were what 17 some candidates in the race? They were all unlikely statistically according to his models. Obviously some of them would become individually more likely as people dropped out of the race. Moreover, the "frontrunner" if there is one, becomes more likely to win as the primaries draw nearer and the polls become more accurate historically.

Yes he's "wrong" in that someone who was once unlikely is going to be the nominee. If that was Fiorina or Kasich or something you probably wouldn't care. But because he had to talk about Trump so much it gives the appearance, I guess, that he didn't know what he was talking about. :shrugs:

Sure, he may get a lot of unfair criticism, but on the flipside: When he's right, he is lauded as a genius, even when his statistical modeling barely differs from a simple polling average.

It's irrelevant though. He doesn't ask to be lauded and doesn't seem to really care as far as I can tell. Criticism should be dolled out on the merits, not to balance the scales.

Glad that people are finally calling out this discredited fraud. Saw him on MSNBC last week talking about TRUMP. He was sweating LIKE A DOG.

I saw that MSNBC appearance too and he literally was not sweating. He was calm and collected the entire time and even admitted it now seems like things may be swinging towards Trump since the establishment is (apparently) not going to fight him, opting to fight Cruz instead. I have never seen anyone call him a "fraud" except you.
Logged
emailking
Atlas Icon
*****
Posts: 14,580
« Reply #1 on: January 26, 2016, 01:57:37 PM »

It's not that he's not just looking at polling data, it's that his non-polls data this year is garbage. "Endorsement points" is an absurd premise worse than most pbrower stuff.

I think you might be looking at that the wrong way. If a candidate gets a whole bunch of endorsements, that tells you who the establishment is willing throw their weight behind,  which probably does tend to affect the result I hope we can agree? He's not saying necessarily that endorsements make voters to change their minds. Correlations can (emphasis can) be predictive even if there are not causal links.
Logged
emailking
Atlas Icon
*****
Posts: 14,580
« Reply #2 on: January 27, 2016, 08:58:53 AM »

It's not that he's not just looking at polling data, it's that his non-polls data this year is garbage. "Endorsement points" is an absurd premise worse than most pbrower stuff.

I think you might be looking at that the wrong way. If a candidate gets a whole bunch of endorsements, that tells you who the establishment is willing throw their weight behind,  which probably does tend to affect the result I hope we can agree? He's not saying necessarily that endorsements make voters to change their minds. Correlations can (emphasis can) be predictive even if there are not causal links.

And look at all the good that establishment support did for Jeb!...

The point is that having endorsements makes your chances more likely that you will win. It does not have to guarantee that you will win for it to be a logical addition to a statistical analysis on this matter. It does not even have to make a lot more likely that you will win, just that it makes it significantly more likely. It's not an absurd premise because you have an example where somebody got endorsements and didn't win. To show that it's an absurd premise, you would actually have to crunch the numbers to show that there isn't correlation or that the correlation is negative. Now yeah, if it's really clear that every single year it doesn't hold then you can skip that step and just call it absurd. But just offhand I can think of several years where it did, e.g. 2012 on the Republican side. You can't just trivially dismiss it like that.
Logged
Pages: [1]  
Jump to:  


Login with username, password and session length

Terms of Service - DMCA Agent and Policy - Privacy Policy and Cookies

Powered by SMF 1.1.21 | SMF © 2015, Simple Machines

Page created in 0.025 seconds with 12 queries.