The book's subject-matter is extremely interesting and Netz has an enviable and deep grasp of a wide range of material and associated background. I am happy to note the classicist opinion: the importance of the book lies in the kinds of questions it asks, and it’s demonstration of how we might begin to answer them. This is what makes the book a game changer.

Unfortunately, Netz's quantitative analysis is marred by elementary and avoidable technical errors. The question then is how much this matters. Since he clearly knows what he is talking about, it is quite possible that his conclusions are correct even so. It is also possible that the quantitative data he has collected would also support his conclusions if differently [properly] handled.

But one would really need to start from the beginning, with a technically well-founded specification of the data to collect to address specific and well-posed research questions.

I think my main concern would be not with faulty analysis - which can be fixed, especially if the data still exists - but with faulty study design, so that the data collected may be insufficient in quantity or just of the wrong type to answer the questions ( )