I don't disagree with the authors point that expecting massive hours without corresponding extra compensation is wrong. But I can't believe the blanket asserti

I don't disagree with the authors point that expecting massive hours without corresponding extra compensation is wrong. But I can't believe the blanket assertions I've seen that working 40 hours a week is the proven output maximization limit. If I work 40 hours from monday to friday, and I have nothing to do on Saturday, and a strong desire to implement a an important customer request, how is working 4 or 5 hours on my day off going to result in negative output?