Grim reaper strikes yet again
Posted 09 May 2018 - 09:46
A more mundane (if slightly geeky) explanation is that this an example of statistical clustering, which is to be expected in any random process. I'm reminded of the case of the iPod Shuffle. Initially users complained that it wasn't random enough because it sometimes played two or more consecutive songs by the same artist. So Apple tweaked it to prevent this happening, which was effectively making the selection less random to sound more random. Go figure!
Posted 09 May 2018 - 11:41
The histogram of number of celebrity deaths reported in the BBC link above was produced for a Radio 4 programme called 'More or Less' so in my opinion the data sampling needs to be taken with a large pinch of salt. Still your post, Aeolienne, has given me cause to put brain into gear and to try to recall the little statistics that I have had to use in my working life. Also not geeky at all. Perhaps this thread could develop into 'The Statistics Thread' !