We know the models are imperfect. Isn't that why the Monte Carlo method exists? Also, I have the strong impression Adams is aware of that and chooses not discuss it much with his readers.