By Matt Johnson

Ben Kamisar of The Hill posted an article on November 25th. And right out of the gate, his lack of statistical understanding is glaring.

As I explained in *The mathematical ignorance of the mainstream media*, it was never the case that it was impossible for Donald J. Trump to become the 45th President of the United States, 80 percent does not equal 100 percent, and just because a coin lands on heads one-hundred times in a row doesn’t mean it will land on heads on the one-hundred-and-first flip. But that’s exactly what Kamisar believes.

As his article starts off,

Pollsters who missed Donald Trump’s surprise electoral victory are headed back to the drawing board.

He continues from there by illustrating in great detail how the polls heavily favored Clinton. He pointed out, and rightfully so, that the vast majority of polls favored the former Secretary of State all across the country. For example, he noted

Clinton swept all 19 polls conducted [in Wisconsin] since June, according to RealClearPolitics.

Remember what I said about the one-hundred-and-first coin flip? Yeah. About that.

But instead of trying to understand basic probability, Kamisar unfortunately continues on the path of mathematical ignorance that many of his peers have done before him by putting forth a complicated expose of causal explanation. But the polls and the projections were never causal. They were probable, and the explanation was rather simple.

As Kamisar illustrates in his piece, polling methods could have been improved. But again, most of the polls favored Clinton. And in probability theory if multiple samples drawn from a population are approximating a similar result, then the odds are in favor of that result. In this case, that result was tending towards Clinton. But again, this never meant Trump didn’t have a chance to win. That was the illusion.

Unfortunately, many in journalism and the mainstream media believe that polls are a causal prediction. Polls and forecasts are probable predictions. But apparently, the term “prediction” is the problem, not the mathematical ignorance of journalists and the mainstream media.

Look, the article is a mess. The ignorance of probability is glaring and at no point did Kamisar interview a statistician or mathematical scientist. Instead, he stated incorrectly

FiveThirtyEight pollster Nate Silver, who preached a healthy skepticism about the models that predicted a certain Clinton election, had Clinton with a 71 percent chance of winning.

This sentence doesn’t make any statistical sense. First, he seemed to have missed the part that Nate Silver isn’t a pollster. He is a mathematical scientist who is using Bayesian statistics to make political predictions with information that was changing on a daily basis. In contrast, the pollsters were conducting polls for who would have voted for Clinton or Trump on that day, not on November 8th.

In addition, Silver’s models never “predicted a certain Clinton election.” Again, his predictions were probable. But as the quote clearly illustrates, Kamisar believes 80 equals 100,

Nate Silver, who preached a healthy skepticism about the models that predicted a certain Clinton election, had Clinton with a 71 percent chance of winning.

One final thought, there is an argument to be made for critiquing how the polls were conducted. But that’s not this author’s main problem, nor does it change the fact of a probable result from flipping a coin or rolling a die. His problem is that he just doesn’t understand basic probability.

I’ll address bias and error in a future article. But just know immediately that those writing in the mainstream media and journalism in general more than likely don’t understand these two probability concepts either. Why? I haven’t read anything thus far to lead me to believe they do.

*Matt Johnson is a writer for The Systems Scientist, and a mathematical scientist. You can connect with him directly in the comments section, and follow him on Twitter or on Facebook. *

*You can also follow The Systems Scientist on Twitter or Facebook as well. *

*Photo credit: Thamyris71*

Copyright ©2016 – The Systems Scientist

Sent from my iPhone

LikeLike