Pablo comments on Contrarianism and reference class forecasting