If a city averages 15 murders per year, then the likelihood that two of those murders would fall on the same date is somewhere around 25%. If the number of murders reaches 23, the probability is 50% .
Springfield is in a situation where the number of homicides which have occurred this year are below the total which had taken place last year at this time, and they are taking place at a pace which would set a record for fewest homicides in a decade…but the news is bad because two have taken place in a recent 24 hour period AND 3 in 10 days.
When only 2 homicides had occurred in 4 months I wrote about my frustration that there was nothing mentioning this fact in the media, even while I stated that it was not statistically significant, because I knew that when the story reversed itself the media would not hesitate to exaggerate the importance of that change.
I wish I could say that prediction required some kind, any kind, of extreme insight. There is so much more to say about this, and I will be saying it soon. For now it should suffice to say that “innumeracy” is widespread and it leads to irrationality.
Pingback: Rational Urbanism | No One Is Ever In Danger In Suburbia