by Karl Denninger
Market-Ticker.org
Let’s say you have a bad thing, call it “Mozzy”, that happens once in a while.
Maybe Mozzy is an insect that bites people and gives them a terrible disease. Maybe Mozzy is a snake. Maybe Mozzy is a horrible crime.
So you decide you’re going to try to stop Mozzy. You first measure Mozzy, and find that among the 30,000 people in your town there are 300 events of Mozzy in some period of time. Let’s say that this is a town and it’s a physical event, so you take the town, draw a line down the middle and you do something you think will make Mozzy go away in as an experiment. You’re careful to first map where all the 300 Mozzy events happened over the last three months and draw the line down the middle, allegedly at least, as best you can determine, getting half on each side.
So three more months go by and there are 150 Mozzy events in the part where you did nothing; the rate is the same, or 1%.
In the other half there are 15 Mozzy events.