Yes, I completely agree. The author should have applied it to his "odds of success vs extent to which you consider second order and subsequent consecuences" graph and saved everybody the time.