Campaign buttons for Libertarian presidential candidate Gary Johnson and vice presidential candidate Bill Weld at the National Libertarian Party Convention, May 27, in Orlando. (John Raoux/AP)

Every election year brings out a fresh crop of Bold Predictions of Imminent Third-Party Success. The political scientist Brendan Nyhan has compiled years of these predictions. They mainly conform to the usual tropes of punditry: a vague sense that “something is in the air,” enough weasel words to maintain plausible deniability when — surprise! — No Labels, Americans Elect, or Mark Zuckerberg doesn’t end up in the White House.

Our partnership with Good Judgment offers a better answer. For several weeks now, Good Judgment has asked its forecasters a specific question: “Will an independent or third-party candidate win at least 5% of the popular vote in the U.S. presidential election?” And forecasters have had to provide a specific percent chance that this would happen. They can’t hide behind timid phrases about what “could happen” or “might happen.”

What’s fascinating, then, is to see the trend that’s developed:


In the past three weeks, the estimated chance went up from close to 0 percent to 25 percent. This is to say, the forecasters see a stronger possibility that a third party or independent candidate could reach a relatively rare threshold in the popular vote. “Stronger possibility” does not mean “very likely,” of course, because the current odds are only 1-in-4. Nevertheless, the trend is noteworthy. It could reflect the ongoing debate within the Republican Party about Donald Trump, the fact that the Libertarian Party nominated two unusually qualified candidates or other factors.

Why should we believe that this crowd-sourced forecast is valid? As I noted when we announced our election forecasting tournament, Good Judgment has found that ordinary people can be excellent forecasters — even “Super Forecasters.” Moreover, they have found that “crowd-sourcing” among thousands of interested observers can also produce forecasts that are more accurate than any other indicator. I certainly wouldn’t presume to have some better forecast myself.

If you’re interested in joining the tournament and making your own forecasts, sign up here. In the meantime, watch this trend.