back to article Meet the Loughborough 'emo' boffins who predicted Trump's victory

This has not been a good year for opinion pollsters, most of whom failed to predict either Britain’s vote to leave the European Union or the election of Donald Trump as US president. To be fair, both were close races. Wise pollsters offer probabilities rather than certainties: FiveThirtyEight’s Nate Silver, who successfully …

  1. Fruit and Nutcase Silver badge

    Recount in Wisconsin?

    In a post on Medium on Wednesday, Mr Halderman repeated concerns he has voiced in the past over the vulnerabilities of paperless voting machines.

    http://www.bbc.co.uk/news/world-us-canada-38090185

    1. Mephistro
      FAIL

      Re: Recount in Wisconsin?

      From the link you provided:

      "There is a deadline for any candidates to demand a recount, and they need to pay fees to file a request."

      Holy shit! So the system was designed from the start to never have to do a recount. Two millions and a half per recount per state? With a deadline? It's nuts.

      The icon is for whoever made/approved this law. For you, a thumbs up for providing the link.

    2. Bruce Ordway

      Re: Recount in Wisconsin?

      I still haven't heard which paperless voting machines were used.

      I didn't think Diebold was being used any more due to the ongoing concerns with their systems?

  2. GeezaGaz

    Why didn't they...

    ...put their money where there mouth is?

    Can't have that much faith in their system.

    I mean if I developed something that was super whizz bang and could predict stuff I would have shoved some cash on it!

    Complete B.S.IMHO

  3. Anonymous Coward
    Anonymous Coward

    Using twitter for political predictions seems bizarre to me.

    Too many tweets make a twat, and twats vote to the left of the rest of society.

    1. Anonymous Coward
      Anonymous Coward

      "Too many tweets make a twat, and twats vote to the left of the rest of society."

      Plenty of twats voted for Brexit and Trump.

      Those are pretty right wing.

      1. bombastic bob Silver badge
        Trollface

        "Plenty of twats voted for Brexit and Trump."

        I couldn't vote for both Brexit _and_ Trump [but would have if I could have]

    2. fraunthall

      re too many tweets make a twat

      Great comment. It really broke me up. Same sense of humour I have. I refer to the 'twitterati' as twits. I don't know where that term originated, but I have been using it for 60 years. How appropriate.

      1. Anonymous Coward
        Anonymous Coward

        Re: re too many tweets make a twat

        "Too many tweets make a twat" is a quote from David Cameron ( yes, really ).

  4. This post has been deleted by its author

  5. HieronymusBloggs

    'emo' boffins

    Mine's the black labcoat.

  6. fraunthall

    They are not alone

    Because you folks are from the UK you may not know about Scott Adams, the Dilbert cartoonist, persuasion expert, blogger, etc. He predicted Trump's victory slightly over a year prior to the election day, and posted countless blogs about the progress of the campaign right up to and after election day.

    He gave his detailed reasons for his prediction, allowed that there could be a number of things, events, or happenings along the way that could change the result, but he dealt with all of those as the campaign progressed. He is not a computer modeller/statistician, just a man who was able to recognize reality and see through the haze of wishful thinking and cognitive dissonance that befuddled almost all of the pundits, academics, reporters and other media types, and certainly a lot of the general public and the politicians. Check out his blog. You will be amazed. Google Scott Adams' Blog.

    1. bombastic bob Silver badge
      Devil

      Re: They are not alone

      "wishful thinking" indeed.

      in the US, polls are often used to MAKE news. questions are asked in leading ways to deliberately skew the results to say what the pollster wants. It has been observed in the past that when you get close to an actual election, the accuracy improves, because the effective use of 'weapons grade' polling "that close" to an election isn't very good, AND the polling companies don't want to make themselves look like COMPLETE fools when their poll results are way way way off (not simply 'way off' or 'barely within the margin of error')

      but yeah, I'd say the media and pollsters had a lot of 'wishful thinking' skewing their results towards Mrs. Clinton. No surprise.

      The Dilbert author got it right. Interesting.

    2. Craigyb

      Re: They are not alone

      Thanks for that, it's a very interesting read. I am from the UK and love Adam's Dilbert cartoons, hilarious yet scarily accurate for the most part.

  7. bombastic bob Silver badge
    Pirate

    focusing on how people 'feel'

    this in and of itself is an indicator of an even BIGGER problem - emotion-based voting. I often say that I feel with my fingers [fingers turned upward and wiggling, with cupped hands] and think with my BRAIN, because *ANY* decision based on emotion is likely to have a similar result as BEER GOGGLES.

    but 'feel' is how people are usually manipulated. it's so easy to stir up emotional unthinking responses.

    And yet the Trump phenomenon was NOT emotion-based, for the most part. At least, not from where I saw it.

    So the premise that seeing how people FEEL via some kind of twitter-scan-bot is going to reveal election results, they should stop at the emotion part, except perhaps for judging liberals and liberalism, because all *THEY* *DO* is "feel".

    Trump's campaign and election was a mass "fed up with it" reaction to politicians in general, who have failed to correctly respond to the electorate on important issues. Brexit is similar, as in when EU fails to properly respond to the UKs needs, you get "that reaction".

    Not a lot of FEEL there, except maybe "frustration".

    1. Bruce Ordway

      Re: focusing on how people 'feel'

      >>And yet the Trump phenomenon was NOT emotion-based

      As I understand it, people voting for Trump are hoping to break the a cycle of non-responsive government. This last election reminds me of an earlier race in Minnesota where I live. Here the frustrated population elected Jesse Ventura as governor. It may not be "purely" emotional but how well thought out was frustrated voting? In the case of Jess, not very. TDB how it works out with Trump (and Brexit). I just have to hope for the best for all people involved.

    2. fajensen Silver badge
      Facepalm

      Re: focusing on how people 'feel'

      The bigger problem has already been happening for some time now: Actual legislation about and actual police enforcement of imagined crimes.

      We police images, words or acts which cause no harm to the property, life and limbs of the "victim" except for "offense", which is an internal state of the mind, cannot be measured and therefore is always assumed.

      Every body can be offended over every thing - so of course one will need to turn the country into an open prison in order to police every one and every thing so no-one is offending anyone.

  8. This post has been deleted by its author

  9. Moosh
    Big Brother

    Drug the masses with false reality

    The issue is that a LOT of US polls were WILFULLY incorrect. For example, many polls on realclearpolitics listed democrat samples much higher than the republican samples; and it was mostly in these polls that Hillary was ahead. Looking closer at the polls, you could tell that if a heavily weighted Dem poll still only had Hillary up by 1 or 2 points, she was in serious trouble.

    Call me a tinfoil hat wearer, but I do not for a moment question the possibility of the US media trying to influence the outcome of the election because Hillary represented the safety of the status quo for domestic policy (despite the danger of a third world war with international policy). I cannot think of another reasonable explanation for consistently oversampling democrats in most of their polls other than an attempt to demoralise the opposition or convince readers of what the "right" opinion to have was.

  10. garbidge

    In what sense are they correct?

    Did their twitter augury predict Clinton winning the popular vote with Trump winning the EC? Because, if not, they got it wrong.

    The emotional heatmap geolocation would have to match polling district locations pretty tightly for this kind of predictive analysis to work. Because, as we know, the EC outcome is incredibly sensitive to vote location. <a href="https://www.washingtonpost.com/news/wonk/wp/2016/11/29/how-the-electoral-college-gerrymanders-the-presidential-vote/">Shifting as few as 4 counties across a state border would have changed the outcome of this election:</a>

POST COMMENT House rules

Not a member of The Register? Create a new account here.

  • Enter your comment

  • Add an icon

Anonymous cowards cannot choose their icon

Biting the hand that feeds IT © 1998–2021