As It Happens·Q&A

'You couldn't be that wrong if you tried,' pollster says of U.S. election forecasts

Democratic pollster Chris Kofinis says he's tired of defending his profession as his colleagues make the same mistakes over and over again.

Pollsters who wrongly predicted a Biden sweep 'clearly should not be in the business,' says Chris Kofinis

As of Wednesday evening, there was still no clear winner in the U.S. presidential election. But Democrat Joe Biden, right, didn't get the sweeping victory over Republican Donald Trump, left, that pollsters had predicted. (Jonathan Ernst, Brian Snyder/Reuters)

Read Story Transcript

Democratic pollster Chris Kofinis says he's tired of defending his profession as his colleagues make the same mistakes over and over again.

The U.S. presidential election, as Wednesday afternoon, was still too close to call, with ballots continuing to be counted in several states. But one thing is clear. The pollsters who predicted Joe Biden would win in a landslide were wrong.

Kofinis is the CEO of Park Street Strategies. And over the last month, as several polls predicted a sweeping victory for Biden, he was warning his fellow Democrats not to get too excited.

Here is part of his conversation Wednesday with As It Happens host Carol Off. 

Chris, the last time we talked [in October], you were feeling anxious, despite polling that should have been should have been comforting to you. How are you feeling now?

I guess I shouldn't say I told you so — but I told you so?

I've done this for a long time. And, unfortunately, I think what we saw was a pretty dramatic collapse of polling and a lot of pollsters who simply failed to do their jobs.

You cannot be this wrong, this consistently wrong, and actually pretend like you know what you're doing.- Chris Kofinis, Park Street Strategies

How did they fail and why?

There was this phenomenon of the silent Trump voter. There has been a lot of debate about it, and I've seen in my research — both the focus groups we've done [and] the polling that we've done — that it was real. The only question I couldn't answer was how many of them were there. 

The other problem was, I think, people were assuming that a certain voter was always going to turn out to vote Democrat. And what we saw, for example, in places like Florida and in other states, was Trump was overperforming with Hispanics. We saw that in our polling. He was overperforming with African-Americans. We saw that in our polling months ago. And I think pollsters just ignored that reality or simply weren't doing their jobs right.

Polling is both an art and science, and it's understandable and it's reasonable to say you're going to be wrong. But I think there's just this fixation of making polling like you're using it as a score sheet for a baseball game. And it doesn't work that way.

And I have to say, as someone who does this for a living, I'm pretty upset about it because I'm tired of having to defend people in this business of mine who clearly should not be in the business. Because you cannot be this wrong, this consistently wrong, and actually pretend like you know what you're doing.

That's certainly what people are saying today — is that we just can't trust the polling. The polling is useless in the United States.

To be honest, if you're if you're going to listen to pollsters who are going to tell you that Biden is up by, you know, 11 [points] nationally, yeah, they're going to be wrong ... If you're going to listen to pollsters who say there's not a silent Trump voter, they're going to be wrong. 

I'll give you two examples: Kentucky Senate race, and the South Carolina Senate race. Between those two races, over $150 million dollars was spent to try to win for the Democrats. In both races, the candidates lost by about 15 points. In both races, polling showed the race is either tied or close. You couldn't be that wrong if you tried.

So that was a wasted money?

A hundred per cent wasted money.

I think we need to step back. I would say people who are in the media and people who do what I do for a living and analysts and political strategists, we need to step back and stop being so arrogant and maybe start listening to people who we may not personally like their views.

When I do focus groups, I don't agree with a lot of the people in the group, but I listen to them. I try to understand them. I try to figure out: why do they think that way? What I don't try to do is change their minds. That is not my job.

And I think what you're seeing is pollsters who are trying to change people's minds, because I cannot explain this simply by the idea that polling was off. No. This is two presidential elections in a row that were a disaster. There's a bigger problem here, and people need to start asking that serious question about what's wrong.

And I'm going to tell you what I think is wrong. What is wrong is there are people in this profession who should not be doing it. And there's media people, major news outlets in major news networks in the United States that are listening to pollsters who have no idea what they're doing.

Trump supporters face off Biden supporters outside of a polling site in Houston, Texas, on election day. (Go Nakamura/Reuters)

Is it also possible, though, that the campaigns themselves are listening to the pollsters and the wrong pollsters and they shouldn't be so much? ... What we've seen is the complaints that Florida was lost because the Biden campaign did not read Florida. They were reading Florida according to what the pollsters were telling them. That's a problem, isn't it?

There's no question that's a problem.

In consulting, we call it pleasing the client, right? Giving the client what they want to hear, because, you know, if you tell the client what they want to hear, they're more likely to keep you.

I don't care whether my client keeps me or not. I'm going to tell them what the reality is as I can best understand it. And if they like it, great. If they don't like it, that's too bad.

Florida, Texas and Ohio — three states that clearly were lost long before election day, where the Biden campaign spent money, resources and time going to, instead of going to places like Pennsylvania, you know, Wisconsin, Michigan.

I think what's happened they tried to create this impression that there was a huge battleground, like Texas was in play. Texas was never in play.... Ohio was never in play. You do not have a state in play, meaning competitive, if you lose it by eight points.

FiveThirtyEight, as you know, that's a one that many people follow online. Nate Silver, who's the founder, says the polling models predicted a Biden win because they showed he could survive errors in polling. So he says ... if Joe Biden wins this election, then it proves that he did survive the errors in polling. So what do you make of that argument?

It's a very convenient rewriting of history. If you look back at literally all the stories before election day, they were pointing to a Democratic tsunami. 

So let's assume for argument's sake that Nate is right. What he's ignoring — and this is the convenient part of his analysis to make himself look better — is that ... Democrats, lost seats in the House. We will probably, most likely, almost certainly not win the Senate. By any measure, this was not a Democratic tsunami.

You're just rejoicing the fact that you were right in the outcome, but your numbers were wrong. It's like me saying, you know, the Toronto Blue Jays won the game, but instead of winning them by 20, they won by one. It's not the same kind of analysis.

Were all these things inevitable? ... Or had they done things differently and responded to the polling differently, that there could have been a larger Biden win, or even a Senate win?

I don't know the answer to that.

With Trump, what's really unique was that people were reluctant to say they supported him primarily because ... people call them racist, xenophobic. You know, think of the negative that has been applied to Trump over his last four years. People didn't want to be associated with that. At least some didn't. And it was enough to create ... what I call the silent Trump voter. Other people call it the shy Trump voter.

I can't tell you four years from now [whether] we're going to have the same kind of dynamic. 

But at a minimum, we got to start asking tougher questions, like tougher questions about what voters think and what voters feel. 

If you think, as a pollster or as an analyst, that you are the smartest person in the room, you could never be wrong, you're infallible — you're going to be wrong.


Written by Sheena Goodyear. Interview produced by Chloe Shantz-Hilkes. Q&A has been edited for length and clarity.

Add some “good” to your morning and evening.

A variety of newsletters you'll love, delivered straight to you.

Sign up now

Comments

To encourage thoughtful and respectful conversations, first and last names will appear with each submission to CBC/Radio-Canada's online communities (except in children and youth-oriented communities). Pseudonyms will no longer be permitted.

By submitting a comment, you accept that CBC has the right to reproduce and publish that comment in whole or in part, in any manner CBC chooses. Please note that CBC does not endorse the opinions expressed in comments. Comments on this story are moderated according to our Submission Guidelines. Comments are welcome while open. We reserve the right to close comments at any time.

now