All Prediction is Speculation

Four years ago, a large percentage of Americans, and spectators around the world, were shocked by the “unexpected” outcome of the U.S. election.

The outcome was “unexpected” only for those who were locked in the echo chamber of their social feeds and blindly believing the media coverage and the pollsters.

Four years later, it seems we haven’t learned the lessons from 2016. The media once again is hitting heavy with polls, claiming that this time it’s different.

It’s not different.

Why We Are Tempted By Polls

The desire to believe in polls is understandable: uncertainty makes us freak out. We are wired to need certainty; we want — need — to know what is going to happen. So we’ll cling to anything that seems reasonable.

Especially this year, as the pandemic has shattered our previous illusions of certainty.

In this environment, it’s even more tempting to believe in the polls, especially if they’re telling you what you want to hear — but even if they’re not.

Our desire for certainty is so great that we’ll cling to certainty even if the “certainty” we get is the outcome we don’t want.

We just want to know. We want to have it settled, so we can focus on other things. So it’s tempting to put stock in the news reports and the polls.

Please do not do this.

5 Reasons Why You Can’t Believe the Polls

Here are some things to keep in mind about polls, and why you might want to temper your belief in them:

Framing is Critical. The way you frame a question can influence the answer.

Polls Don’t Tell You Who Votes. There’s no guarantee that the voting sample looks like the polled sample.

People Lie. Not just to the pollsters, but to themselves. Polls are measures of intent, and we are terrible at predicting our actions. This isn’t always intentional lying. Keep in mind that approximately 90% of our actions are governed by our unconscious.

The Past Doesn’t Predict the Future. To extrapolate results, pollsters use underlying models that are based on past election outcomes. There’s an assumption of continuity and consistency in patterns that belies the fact that every situation is new. The past does not predict the future.

You Can’t Poll For Unknown Unknowns. What impact will the pandemic have on voting? Will it impact a voter’s willingness to wait on long lines in the rain to vote, or will that voter fear getting sick and leave? What impact will the unprecedented early voting turnout have on the outcome? The models cannot build in the impact of new or unexpected factors.

Build The Muscle of Sitting With Uncertainty

Ultimately, all prediction is speculation.

Nobody is an expert on the future.

As much as we would like to know how it’s going to play out, we won’t know until the votes are counted. And even that’s going to take some time.

In our culture of instant gratification, it’s quite likely that the election results will take longer than an Amazon Prime delivery.

It’s worth noting that this is not unusual. It’s typical for ballots to continued to be counted for days after the election. What needs to change is the media’s approach: they must not call an election prematurely.

And what needs to change is our approach: we must not demand answers prematurely.

Just like everything else this year, this election is going to force us to build the muscle of sitting in the discomfort of uncertainty.

And, obviously, VOTE.