Wednesday, October 21, 2020

Variance. In One Chart

 The other day I ran down proper usage of terms like "often", "rarely", etc. One of the items was a reference to variance. "give or take". Essentially it is recognizing that though something happens "on average" in nature, very few things, if any, are exactly on the point. For example the earth doesn't circle around the sun in the exact same path each and every year. Sometimes  it's closer at it's closest point and sometimes it's further away at it's furthest point. And sometimes vice-versa, farther at its closest point than "usual" and closer during its farthest point than "usual".  Such events are examples of variance. It's important to understand that when discussing phenomena. And so we have the graph of worldwide COVID deaths.


Note the variance in trajectories. Some countries have high rates and some countries have very low rates. If you looked at the mean it would be somewhere around the 9 per million (eyeballing it). There are many countries far below that number and many countries far above that number. Also looking at the chart there isn't really a discernable pattern to it. While Sweden and New Zealand get attention as outliers in terms of deaths and lockdown/mask policies, places like Gambia, Madagascar and Ghana get no attention but are at the extreme low end of deaths. Anyone like to guess why?


Oh and anyone believing China's numbers can come see me about the sale of a bridge spanning the East River. I mean from 185 per million  to zero? Sure.


This also undescores, for me at least, that a lot of the talk about what South Korea and other Asian nations "did right" and "took seriously" is really about variance and the nature of human bio diversity. Some populations are more susceptible for various reasons where others are less susceptible.