Bad model, no biscuit

| No Comments

I think I would be fine-tuning my model - from the Washington Examiner:

Election model that predicted 2018 blue wave has Trump losing, liberal voters ‘terrified’
A university election model that predicted the blue wave in the House in 2018 almost to the seat is predicting a big loss by President Trump next year due to an explosion of bitter partisanship and Trump hate.

An election forecast model designed by Rachel Bitecofer, assistant director of the Wason Center for Public Policy at Christopher Newport University, predicted that Trump will lose the Electoral College 297-197, with 270 of 538 needed to win.

Three key states that helped push Trump over Hillary Rodham Clinton in 2016, despite her winning the popular vote, Pennsylvania, Michigan, and Wisconsin, will turn back to the Democrats, she said.

“Trump’s 2016 path to the White House was the political equivalent of getting dealt a Royal Flush in poker,” said Bitecofer. “It’s probably not replicable in 2020 with an agitated Democratic electorate.”

I just do not see it. Computer models are great for some things but for politics and climate, they just do not work. Their designers are too partisan. The computer is just a shell to hide their agenda.

Leave a comment

March 2023

Sun Mon Tue Wed Thu Fri Sat
      1 2 3 4
5 6 7 8 9 10 11
12 13 14 15 16 17 18
19 20 21 22 23 24 25
26 27 28 29 30 31  

About this Entry

This page contains a single entry by DaveH published on July 1, 2019 11:53 AM.

Word to the wise was the previous entry in this blog.

Obama's legacy - Iran is the next entry in this blog.

Find recent content on the main index or look in the archives to find all content.

Monthly Archives

Pages

OpenID accepted here Learn more about OpenID
Powered by Movable Type 5.2.9