Opinion: Rash pundit predictions skew voter perception and are bad for our politics

Political commentary can be useful, but those claiming to know the future aren’t worth the trouble.



Ben Shapiro, host of his online political podcast The Ben Shapiro Show, at the Conservative Political Action Conference (CPAC) sponsored by the American Conservative Union held at the Gaylord National Resort & Convention Center in Oxon Hill, MD on February 22, 2018.

Jason O'Day, Columnist

Political commentator Ben Shapiro awarded the title of “prohibitive frontrunner” to Sen. Elizabeth Warren, D-Mass, last fall. At the time, she had a narrow lead over former Vice President Joe Biden in most polls.

Then on Feb. 20 he repeated the blunder, crowning Sen. Bernie Sanders, I-Vt., with the same title after he performed well in the first three nomination contests. Shapiro also declared Biden electorally dead, “an actual human corpse,” and engaged in more hyperbole about how Biden had to be scraped off the debate stage floor.

Yet, the Biden campaign was more alive following South Carolina, and even more so on Super Tuesday, when he won 10 out of 15 contests.

My point is that the political news cycle is chaotic, far too chaotic for an ordinary observer to follow. Even the “professional” pundits can’t get it right. There needs to be a better way to consume and analyze news.

Those who make their living commentating on politics have a natural incentive to make confident predictions. If they turn out to be wrong, most people move on and no one really notices. But if a commentator is proven right he can stand up on their desk, wave the results around and declare themselves a genius.

That’s essentially what Shapiro did late Tuesday night on his Daily Wire podcast, smugly noting which states he accurately predicted Biden would win.

It would be nice if commentators spent more time talking about substantive issues and policy positions than fleeting probabilities.

Shapiro is an intelligent conservative voice and I respect him, but he should be more responsible before making capricious forecasts about elections. He’s not the only one; scores of other pundits have spent decades barfing up irresponsible predictions about elections. These wild predictions are more than annoying, tiresome, and hackneyed. They disincentivize voters from showing up at the polls.

This extends to other punditry realms. Most polling aggregators, such as FiveThirtyEight and The Upshot, gave former Secretary of State Hillary Clinton odds of victory ranging between 70 and 99 percent. In fairness to FiveThirtyEight editor-in-chief Nate Silver, he saliently pointed out that most other prediction models were ignoring the possibility of Trump having narrow victories in key swing states like Pennsylvania.

A study by the University of Chicago found that such forecasting inclined many voters to stay home or cast a protest vote for Bernie Sanders. I worked as an election official in Davenport in 2016, and about 50 of the write-in ballots we had to count by hand were for Sanders.

Polls are an important tool to gauge a candidate’s viability. Students wouldn’t want to waste their time working for a dead-end presidential campaign.

However, polls change constantly. It would be nice if commentators spent more time talking about substantive issues and policy positions than fleeting probabilities.

All speculation of this variety must be taken with a grain of salt. Primaries and elections are like weather patterns and Super Bowls because no one, not even experts, have any idea what the end result will actually be.

I could bother predicting a winner of Democratic nomination, but it’s not worth sharing because I’m not a wizard and I lack access to some ultra-exclusive information that half these pundits apparently have.

By June, probably much sooner, the Democrats will have figured it out. Until then I’m content with waiting prudently to find out who that is.

Columns reflect the opinions of the authors and are not necessarily those of the Editorial Board, The Daily Iowan, or other organizations in which the author may be involved.