Fortunately, we saw in the last chapter that the number of homicides each day act as if they are random observations drawn with a Poisson distribution from a metaphorical population of alternative possible histories. This in turn means the total over the whole year can be considered as a single observation from a Poisson distribution with mean m equal to the (rather hypothetical) ‘true’ underlying annual rate. Our interest is whether m changes from year to year. The standard deviation of this Poisson distribution is the square root of m,
