quarta-feira, julho 08, 2020

Comentadores e prognósticos

Uma destas noites estava a jantar e por azar a televisão estava ligada na RTP1. Então, aparece um comentador que eu julgava já estar reformado, Nicolau Santos.

Foi dele que me lembrei esta manhã ao ler:
"Few readers will be surprised that Tetlock learnt from his initial work that the forecasters in his sample were not very good; little better than a chimpanzee throwing darts. What is, perhaps, most surprising is that he found that the principal factor differentiating the good from the bad was how well known the forecaster was. The more prominent the individual concerned, the more often the forecaster is reported by the media, the more frequently consulted by politicians and business leaders, the less credence should be placed on that individual’s prognostications.
...
We both have the experience of dealing with researchers for radio and television programmes: if you profess an opinion that is unambiguous and – for preference – extreme, a car will be on its way to take you to the studio; if you suggest that the issue is complicated, they will thank you for your advice and offer to ring you back. They rarely do. People understandably like clear opinions but the truth is that many issues inescapably involve saying ‘on the one hand, but on the other’."
Trechos retirados de “Radical Uncertainty” de John Kay e Mervyn King 

Sem comentários: