This article reminds me of Nate Silver's book [1] which has a far more scientific and in depth look on the failure/success of these types of predictive statistics.
The chapters on the many attempts (or many failed hopes) in predicting earthquakes was particularly interesting, including many times the media has bought into hyped up new charlatans who say they finally figured it out but which ultimately failed to survive under basic statistical scrutiny.
It also has a useful soft introduction to Bayesian statistics and other useful concepts from the field of prediction that I hope more journalists read about. As this seems to be a very common theme in reporting.
Even this journalist couldn't help themselves with this line (combined with some scary looking charts described with an alarmist tone farther down):
> Was there some miscalculation of how frequently these massive flooding events occur? Or, most alarmingly, is something else happening that suggests these catastrophic weather events are becoming much more common?
The failure to mention the effects of El Nino/El Nina seems like a big oversight in this article, especially when we're just coming out of a particularly strong one. Climate stats are an easy one to get wrong - or to shape into any narrative - especially when timeframes and location are easy things to be viewed too narrowly.
The chapters on the many attempts (or many failed hopes) in predicting earthquakes was particularly interesting, including many times the media has bought into hyped up new charlatans who say they finally figured it out but which ultimately failed to survive under basic statistical scrutiny.
It also has a useful soft introduction to Bayesian statistics and other useful concepts from the field of prediction that I hope more journalists read about. As this seems to be a very common theme in reporting.
Even this journalist couldn't help themselves with this line (combined with some scary looking charts described with an alarmist tone farther down):
> Was there some miscalculation of how frequently these massive flooding events occur? Or, most alarmingly, is something else happening that suggests these catastrophic weather events are becoming much more common?
The failure to mention the effects of El Nino/El Nina seems like a big oversight in this article, especially when we're just coming out of a particularly strong one. Climate stats are an easy one to get wrong - or to shape into any narrative - especially when timeframes and location are easy things to be viewed too narrowly.
[1] https://www.amazon.com/Signal-Noise-Many-Predictions-Fail-bu...