Like what you've read?

On Line Opinion is the only Australian site where you get all sides of the story. We don't
charge, but we need your support. Here�s how you can help.

  • Advertise

    We have a monthly audience of 70,000 and advertising packages from $200 a month.

  • Volunteer

    We always need commissioning editors and sub-editors.

  • Contribute

    Got something to say? Submit an essay.


 The National Forum   Donate   Your Account   On Line Opinion   Forum   Blogs   Polling   About   
On Line Opinion logo ON LINE OPINION - Australia's e-journal of social and political debate

Subscribe!
Subscribe





On Line Opinion is a not-for-profit publication and relies on the generosity of its sponsors, editors and contributors. If you would like to help, contact us.
___________

Syndicate
RSS/XML


RSS 2.0

US Election 2020: are the analytics correct?

By Charles Essery - posted Wednesday, 4 November 2020


2016 was a great year for my "betting-wins" of the political events that year. I am not a betting person, ignore the Melbourne Cup as I am an animal-lover, and in fact wouldn't know how to pick or arrange a bet on a horse race. A friend runs bets for a small group of us who are interested in world politics. In 2016, I was a newbie-maverick by forecasting Brexit, the UK Election and worst of all (in friends eyes) the Trump win.

Other friends/colleagues work within the polling and market analysis business, and they were shocked by my trifecta. The question is, have the pollsters and analysts got their models correctly "tuned" this time unlike 2016? Or like climate change modellers, are they using inappropriate models, with unrealistic assumptions and wishful parametrisation, to get the results they want to tell their clients and the media?

It's on now, and all the media hype, biased reporting and wishful (wilful?) opinion givers are fighting to make their mark on TV, cable, radio, podcasts and of course on Twitter and Facebook. The parallels between climate change cultism (CCC) and Trump Delusion Syndrome (TDS) is incredibly strong in my opinion. Both are driven by democrats/ left-wing/ Green politics and are supported by left leaning media/ pollsters and now it would seem Twitter (Cruz vs Twitter) Both use models that they believe represents their view of the world.

Advertisement

In the case of CCC, their beloved Global Circulation Models are founded on simple energy balance equations that ignore "inconvenient data" and hence only test their results against "cleaned data". Pollsters and political analysts equally choose their own model of the world, but in their case, they are modelling people's behaviour/views. Like CCC modellers, they have specific assumptions, often based on those that suit their paying clients. And like CCC modellers, they are often selective in what data to use, by configuring the sample sets to match client expectations and by the questions asked and how these are framed.

This is an "anonymous sourced graph" (no, not from Miles Taylor!) that is simple, yet shows so much. This is not a New York Times "analytic" like their "Pro-Chaina" propaganda on Covid-19 (eg the outbreak was due to a black Ops attack on Wuhan (NYT 17March20) or that the virus came from the wet markets (NYT, 20Jan20, NYT, 13 Jun20)

Unlike the NYT, I have sourced this graph from a known Online forum/blog (Working paper by Olivier Coibion et al

The two bar charts show % chance of a Trump of each party members (so in case of Republican's, 2.3 % have a 100% confidence that Trump will win, while only 1.6% of Democrats belief he has zero chance of winning). Ironically the Independents (Libertarian) highest believe is that there is a 550:50 chance of Trump winning.

This simple graph provides some interesting insights to me.

Advertisement
  • Independents/Others are balanced with a peak at 50% and low numbers at the extremes. In US terms, that means they are somewhere to park your abstain vote! They will never be in power.
  • Republicans are more confident and committed that Trump will win (2.3% 100%), than democrats are about Biden (0.7% 100%).
  • Democrats' level of confidence is less intense and their distribution curve more balanced and less confident about a Biden win

Pre Covid-19, Trump had survived Clinton's bile on media, the Russian investigation, the Ukraine affair and Impeachment, yet the economy was doing well (all Obama's work of course!). He also had to chorale the residual internal rifts in the Republican camp.

The Biden democrats have been viciously prosecuting the "Covid-19 mismanagement" argument from this "Jackass" Trump who is unfit to rule. But as all the smoke and mirrors stuff fades, the early 10+ Biden lead is shrinking, with some polls even suggesting too close to call. While Trump supporters at his "protest events" shout "FOUR MORE YEARS!", who will utter the words "You're fired"?

So, eyes down for a full house and I am wondering will I continue my "lucky" streak with my analysis. It's going to be the best show in town, and if President Donald Trump's "protest events" are anything to go by, it will be entertaining (unless you are watching CNN, MSN, ABC (US), BBC, or the ABC(AU)). I have declared Trump to win with my friends, and will be happy with a cheap, quaffing chardonnay from ALDI, while I know others in the group who want Biden, will expect a bottle of Grange!

  1. Pages:
  2. 1
  3. All


Discuss in our Forums

See what other readers are saying about this article!

Click here to read & post comments.

17 posts so far.

Share this:
reddit this reddit thisbookmark with del.icio.us Del.icio.usdigg thisseed newsvineSeed NewsvineStumbleUpon StumbleUponsubmit to propellerkwoff it

About the Author

Charles Essery is an independent water consultant, who has been an Australia resident since 1990.

Other articles by this Author

All articles by Charles Essery

Creative Commons LicenseThis work is licensed under a Creative Commons License.

Photo of Charles Essery
Article Tools
Comment 17 comments
Print Printable version
Subscribe Subscribe
Email Email a friend
Advertisement

About Us Search Discuss Feedback Legals Privacy