Warning: This post was published more than 5 years ago.
I keep old posts on the site because sometimes it's interesting to read old content. Not everything that is old is bad. Also, I think people might be interested to track how my views have changed over time: for example, how my strident teenage views have mellowed and matured!
But given the age of this post, please bear in mind:
- My views might have changed in the 5 years since I wrote this post.
- This post might use language in ways which I would now consider inappropriate or offensive.
- Factual information might be outdated.
- Links might be broken; embedded material might not appear properly.
Many thanks for your understanding.
The news reader announces a 30% chance of rain tomorrow. Thirty per cent of what? Most people in Berlin think that it will rain tomorrow 30% of the time. Others believe that it will rain tomorrow in 30% of the region. In New York the majority believe that it will rain on 30% of the days for which the prediction was made.
This is the (edited) opening of BMJ 2012;344:e245, a fascinating paper about single event probabilities by Gigerenza and Galesic. It’s only short, quite fun, and may convince you to change your practice. Well worth a read.