Another objection to my talk: Risk ends up being a lagging indicator and not a forecasting one. Good news: we can make better forecasts if we understand the constraints, and calculating risk this way provides these upper and lower limits for variables.