Twenty Twenty

A lot of maths used for the management of risk is problematic. There is a desire to make it look pretty, and come up with models and numbers to give the illusion of understanding. If you can measure it, you can manage it, the theory goes. The real value in models is simply a device for thinking through something. A tool to help us compare and communicate. The danger is that when things are complicated, you often see whatever it is you are looking for. If a bunch of investors are looking for the best companies, and they define that as something that gives a return of 15-20%, you can be sure most of their models will spit out 15-20%. You don’t really understand risk if you then use that to rank various different analysts’ work. Your 20% is not my 20%. My 20% is not even my 20%. Add a couple of decimal places, and you realise that 36.79% of numbers are made up. Having a summary number doesn’t give you a full understanding of risk. Doing the work gives you a clue. Getting things wrong gives you a clue. 

your Twenty my Twenty