Skip to content
Link copied to clipboard

Are you a super-forecaster? The Good Judgment Project needs you

Are you an expert at predicting the future? The Good Judgment Project may want you in 2017.

The brainchild of professors Philip Tetlock and Barbara Mellers, a husband-and-wife team teaching at the University of Pennsylvania, the Good Judgment Project is a sort of crowdsourcing machine that allows anyone to make expert forecasts on topics of all kinds. For example:

• Will there be a terrorist attack in the United States with five or more fatalities in 2017? (Perhaps one similar to the recent attack at Berlin's Christmas market?)

• Will our new president, Donald Trump, host Russian President Vladimir Putin on a state visit to the United States, or vice versa?

If you think you know the answers, you might be what they call a "super-forecaster."

"We're trying to civilize the dialogue," said Mellers. "The goal of the Good Judgment Project helped us to determine whether some people are naturally better than others at prediction, and whether prediction performance could be enhanced."

To become a super-forecaster, for instance, "you have to leave your wishes and opinions, your baggage, behind"  in order to make accurate predictions and win, Tetlock explained.  "You have to want to understand sentiment in order to win."

Mellers and Tetlock's effort was one of five academic research teams that competed in an innovative tournament funded by the federal government's Intelligence Advanced Research Projects Activity. Forecasters were challenged to answer geopolitical and economic questions that U.S. intelligence agencies posed to their analysts.

The IARPA initiative ran from 2011 to 2015 and recruited more than 25,000 forecasters who made well over a million predictions -- on topics ranging from whether Greece would exit the eurozone,  to the likelihood of a leadership turnover in Syria, to the risk of a financial panic in China.

The Penn professors' crowdsourced model decisively won the tournament, besting even the intelligence community's own analysts. As a result, Tetlock and Mellers are now working jointly with IARPA, a division of the Office of the Director of National Intelligence, hosting other tournaments.

They also founded a private start-up company in 2015, Good Judgment Inc., that makes their "super-forecasters" available for hire to corporate clients such as sports teams, multinational corporations, even oil-producing countries and foreign governments. Tetlock's book Superforecasting: The Art and Science of Prediction (Random House) was published in paperback in September.

The couple live in Ritttenhouse Square and talked recently about how soon-t0-be-President Trump might change the landscape for the economy and world affairs, and which questions their super-forecasters will tackle in 2017.

After signing up, anyone can take part in the "Trump 100 Days" challenge at GJOpen.com, the Good Judgment Project's public website. Currently being asked at GJOpen: What will President Trump accomplish in his first 100 days in office? Will his administration take action on immigration, health care, or trade? Add your forecast to the crowd, explain your reasoning and be challenged by others, then find out how you stack up.

What do Tetlock and Mellers hope to predict for 2017 and beyond?

They're wondering how Trump might compare to historical figures ranging from Richard Nixon, to former Italian Prime Minister/media mogul Silvio Berlusconi, even to Ronald Reagan.

For instance, Tetlock said, if an analogy between early World War II-era British Prime Minister Neville Chamberlain and Trump is used, "then the next round of questions might include whether the U.S. will cede Ukraine, Estonia, and other Baltic countries to Russian influence. If we use Reagan as an analogy, then Trump may push U.S. GDP growth to 3 percent or higher. Each analogy contains the seeds for questions in our tournaments - and it's better than a bar fight."

Americans make predictions in all areas of life - sports, gambling, investments - and that's similar to the way the Good Judgment Project makes predictions about elections, treaties, even wars.

"They feel like one-off events to us, but they happen over and over in history. There really is nothing unique under the sun," Mellers said.

They are currently pondering the fallout from Brexit, questions such as when Article 50 will be triggered for the United Kingdom to leave the European Union, whether that might prompt another snap election in British government, whether the Bank of England will consider more interest-rate cuts, and whether President Trump will flip on free trade and strike a deal with the U.K.

"Is Trump capable of dramatic shifts of a Nixonian sort, such as when Nixon went to China?" Tetlock asked. "He was famously anti-Communist, Nixon, but then traveled to China and was photographed shaking Mao's hand in 1972. That's what we're wondering about Trump."

Good Judgment also is looking at economic, health-care, and violence questions. What other health pandemics, such as Zika and Ebola, might develop? Will there be another terrorist attack in the U.S. by June 2017? How high will the murder rate be in Chicago at the start of this New Year? What will the price of Brent crude oil hit by 2018?

They are also examining whole industries and the possibility of a fourth Industrial Revolution, led by artificial intelligence. Tetlock and Mellers will travel later in January to the World Economic Forum in Davos, Switzerland, to present on this topic, which encompasses questions such as the self-driving car industry and how it will disrupt truck driving and auto manufacturers.

"The idea is that by 2040, A.I. will displace large fractions of blue-collar and white-collar workers. How fast and how deep will this fourth Industrial Revolution last?" Tetlock asked. "How many accounting jobs might disappear? Will [IBM's] Watson MD outperform the best cancer physicians and their diagnoses in two years?"

Mellers, for her part, has been measuring the change in cognitive styles and open-minded thinking among those who take part in the Good Judgment challenges.

"To be an accurate forecaster, you have to focus on reality," she said. "You can't hold on to ideology or preconceptions. It's an exercise than opens your mind."