How can we better apprehend all the confusion of the world around us? To answer this question, the American intelligence services organised between 2011 and 2014 a "forecasting tournament" (The Good Judgment Project) which brought together more than 20,000 participants from various backgrounds. The results of this immense competition reveal the good practices and skills that everyone can mobilise to navigate more intelligently through uncertainty.
Aaking decisions inevitably involves a forecasting phase. Whether you work in HR, business development or IT, the ability to formulate accurate forecasts is one of the keys to making informed decisions. And yet it's incredibly difficult, and even more so in an environment of increasing complexity and uncertainty. How can you improve your forecasting capabilities? Superforcasting authors Philip Tetlock and Dan Gardner present the findings of a study they conducted on the winners of a major competition held between 2011 and 2014: a group they called "superforecasters" because of the incredible accuracy of their projections. Why do these superforecasters stand out from the rest? "The answer is not their education or their access to sensitive information. The answer is not their education or their access to sensitive information, nor their opinions - whether they are liberal or conservative, optimistic or pessimistic," the authors explain. The determining factor is their way of thinking.
ADOPT DATA-DRIVEN THINKING
The study found that the judgment of super-predictors is less susceptible to common cognitive biases because of their ability to synthesize a wide range of information and perspectives.
You don't have to be a mathlete...
Are IQ and education level a determining factor in making you a good forecaster? While super forecasters may be well above average, they are not unbelievably good at what they do, and most of them are not even close to being geniuses, which is a tricky concept to say the least, and is often arbitrarily defined as a top 1 %, or an IQ of 135 and above," observe Philip Tetlock and Dan Gardner. If intelligence and knowledge help, they make little difference beyond a certain threshold. The ability to understand and work with numbers is more helpful. "The authors concede, "Superpredictors are people who know how to count. However, being bad at math doesn't necessarily exclude you from the clan: rarely do forecasts force you to swallow numbers. For example, Philip Tetlock and Dan Gardner describe super-predictors as Bayesian reasoners (named after Bayes' theorem that allows you to update estimates of a probability), but note that this does not necessarily imply being able to solve a mathematical equation. What matters above all to super-predictionists is not so much Bayes' theorem as its basic approach: to move progressively closer to the truth by constantly updating one's reasoning in the light of evidence.
...to draw on data-driven thinking...
When faced with a problem, super-predictive analysts begin by analysing the issue to extract its various components. Then, they try to place the problem in a broader scheme of phenomena...
