Forecasting has been an exciting topic for many. We look up to those who can predict the future. They are promptly elevated to the divine level. Nostradamus is one of them. But the reality is that popular predictions have an extremely low success rate.
Nate Silver & Philip E. Tetlock offers an interesting point of view in understanding the science behind it. While seeking out various personalities, Tetlock asserts that the experts who are the media lovers (press, television, etc.) are on the losing side. Interestingly, the more these personalities are exposed in the media, the worse the predictions they make. On the other side, another subgroup of experts, whose predictions have worked really well.
Tetlock classifies these two group of personalities into Hedgehogs and Foxes. At high level,
“Fox believes in many little ideas, but Hedgehogs believes in one big idea”
Fox personalities have better hit ratios in forecasting than hedgehogs. Hedgehogs have done a great deal of work in one or two areas and have developed a framework. They feel strongly that the world operates within their framework. That "confidence" is what attracts the media, the press to feature it. Conversely, the Fox personalities think that the world is not so simple and ready to change their mind based on the new evidence. In other words, these are Bayesian thinkers. Their opinions are highly probabilistic and change their predicted outcome based on the new streams of evidence that they see. This frequent shifting of opinion is coming across as a ‘not-so-confident and very-volatile’ personalities to the outsiders. They typically remain anonymous unless they are extremely popular (Think: Jeff Bezos)
I have had an experience in dealing with these mindsets. One of my former boss is foxy by design. This nimble style makes him a great decision maker. Those who understand his guiding principle viewed him as a strong, innovative and strategic leader. The people who had the hedgehog spirit found him less strategic as he changes his opinions, decisions quite fast.
Jeff Bezos in one of his interview said,
“The people who I know are right a lot changed their mind a lot.”
Forecasting work well for Bayesian thinkers who constantly update their predictions based on the new evidence that they come across. Their decisions are highly probabilistic in nature.