Think Tank discusses polls, election night outlook and more

Should you believe the polls? How election night 2020 will be unlike any other. What to look for on election night for an early indication of the outcome.

I have been doing this show for about 7½ years. It has been a real labor of love and something quite different for me. I don’t pretend to be a “radio guy.”

I see people here who can fill the airwaves with their perfect voices, never stumble over a word and do it all effortlessly. That’s not me.

I am here as a policy wonk, someone with the luxury to read several hours a day on matters of policy. But even this is really a hobby.

More significantly, I have spent the last 40 years doing survey research (“polling” if you must). My training in this involves years of study at some of the leading academic survey research centers in the world.

My relevant expertise is both real and extensive.
Ten days before the election, I thought it would be worthwhile to share some of this perspective and address:

• Can we trust the polls after 2016?
• Should we believe them?

Very fair questions.

Why important? If the polls are at all accurate, Joe Biden has this election all wrapped up. But that is a very big “if.”

Most people think professional pollsters will be the first to defend the infallibility of polls. Completely wrong. I know many of the serious professional pollsters in this country personally. Many are friends.

I have listened to them present at professional meetings and I’ve spent innumerable hours discussing methodological issues with them. True professional pollsters know the limitations on and fragilities of their work.

In this hour I will shed some light on the question “can you believe the polls?”

• There are some genuine and enduring problems with polling that have evolved over the last 20 years or so that concern me greatly. And I’ve been talking about these publicly in speeches for all of those years.

• But there are also reasons to believe that the polls will be more accurate in 2020 than they were in 2016.

• And most of the theories you have probably heard about what happened in 2016 turn out to be totally wrong. I will tell you about these as well.

To do this I will walk you through the history of polling over the last 100 years.

Not as a sterile historical exercise, but because some of our current problems have echoes in prior debacles (1936 Literary Digest, 1948 “Dewey beats Truman,” and 2016, among others).

Many of polling’s current limitations have their roots in the problems of decades ago.

We will then consider:

• What happened in 2016 polling (systematic research has been done on this and many of the popular assumptions are simply wrong)

• What is different (and the same) in 2020

• Which pollsters should you trust and why.

• Why election night 2020 will be unlike any other.

• What are the most likely paths to victory for each candidate?

There is a lot riding on the answers to these questions. I hope to provide some real and useful insight.

The Think Tank airs on KTAR 92.3 FM on Saturday 3-4 p.m. and Sunday 9-10 p.m.

Podcasts are available after broadcast.