Why you should care
Because even a leader with a good gut needs nourishment.
At a town hall meeting with U.S. troops at Fort Meade last fall, President Barack Obama returned to an observation he makes a lot regarding the challenges of his particular line of work. “[I]f it’s an easy question, it doesn’t get to my desk,” he observed. “And my job is to make a decision based on sometimes imperfect information, and you’re working on the percentages.” So what were the chances that Osama bin Laden was in that house in Pakistan? A 50-50 proposition, according to Obama.
Obama’s candid assessment raises some rather uncomfortable truths about the hazards of presidential decision-making. Our nation’s “Decider,” as President George W. Bush referred to himself, is often working from “imperfect information” that can also be poorly communicated, and which, until recently, was not even couched in terms of “the percentages.” Regarding a Bay of Pigs invasion in 1961, the Joint Chiefs of Staff informed President John F. Kennedy that there was a “fair chance” of success — something he took as far more positive than the 3-to-1 odds against success that they apparently meant to suggest.
The political judgment and predictions of our “best and brightest” experts, pundits and leaders are actually rather dismal.
When the shit hits the fan in the Situation Room, though, don’t we want a decider-in-chief who is not only well-briefed in the available intelligence, but also fully prepared to interpret that intelligence, including “the percentages,” according to the most sophisticated analytical methods we have available? We should be teaching a president-elect how to be better at forecasting statistics and using the decision tree analysis that many business and law school students learn today. Of course, someone like Donald Trump may be unwilling to accept such an arrangement, but all president-elects should be required to take a crash course in decision-making and analysis.
The political judgment and predictions of our “best and brightest” experts, pundits and leaders are actually rather dismal. In the largest and longest-running study of political judgment, Wharton School professor Philip E. Tetlock concluded that the average expert’s political predictions were about as accurate as a “dart-throwing chimpanzee.” Part of the reason for this is that our media, government, corporate and academic cultures reward those whom Tetlock labels “hedgehogs,” after the phrase attributed to the Greek poet Archilochus: “The fox knows many things, but the hedgehog knows one big thing.” Hedgehogs typically have one grand theory about the world (from Marxism to neoconservatism) and express their opinions with great confidence, which makes for good television and occasionally great leaders, such as Winston Churchill, whose focused worldview proved correct when it came to the threat posed by Adolf Hitler.
Foxes, on the other hand, though far less sexy, are much better at making accurate political judgments: They seek out a number of sources, they aggregate and synthesize information and reevaluate prior assessments. Some presidents are more foxlike than others, former White House Counsel John W. Dean writes in FindLaw. There’s Obama and Bill Clinton, for example, while others such as George W. Bush are like hedgehogs. But every president can improve their approach to decision-making, including whom they choose to listen to. Right now presidents generally base their judgments in that area on status or connections, ideological affinity and media prominence — “all proven to be weak predictors of accuracy,” Tetlock tells OZY.
There are ways to improve political judgment, however. It turns out the everyday men and women who emerged from Tetlock’s “Good Judgment Project” are more capable prognosticators than many working in the intelligence services, as Tetlock and co-author Dan Gardner explore in their best-selling book Superforecasting: The Art and Science of Prediction. Indeed, part of improving our government’s political judgment will include getting such agencies to adapt forecasting techniques. But we should also be making sure that our top-dog decider is embracing such knowledge as well; after all, we can’t assume that the person capable of winning a presidential campaign will automatically be a good decision maker. Need an example? Look no further than Canadian prime minister Justin Trudeau, who read Superforecasting prior to his election win. Gardner has also served as an occasional consultant to Trudeau’s office.
Justin Trudeau reading Superforecasting days before his election win. Post hoc ergo propter hoc, right? QED. pic.twitter.com/k2vusMsPMe— Dan Gardner (@dgardner) December 4, 2015
To be sure, the “president need not be a great forecaster” himself, as Tetlock points out, but he “should have mechanisms for sorting out who has better track records than whom, in which policy domains.” And such mechanisms should be in place early. “It’s critical to think carefully about how decisions are made, and which processes can best support those decisions before the decision-making begins,” says Gardner.
Will Trump be a hedgehog, and what are the odds that he and his administration will channel their inner superforecaster — and employ an army of them — to help make better decisions? Hopefully better than a “fair chance.”