I'm really curious how you would suggest to measure the "true value" of books.

I think there may be no such thing. Base ball is a clearly defined game with conditions for winning and losing, which gives you objective information on these true values.

But aesthetics (here book appreciation) simply isn't objective to begin with.

> I'm really curious how you would suggest to measure the "true value" of books. ... Aesthetics (here book appreciation) simply isn't objective to begin with.

And this would be the crux of the problem. We have sales numbers (objective) and reviews (subjective). It is compounded that undervalued books will have fewer reviews to work with.

But the hardest part is that people are not good judges of these things. Or maybe they are, but they have different tastes than you do. So, to action!

1) You rate the top 100 books in amazon as high valued or not.

2) A scraper + statistics finds people who agree with you.

3) Looking at the bottom half of books, find those which your subset of people like.

4) Farm it out to netflix, since that is what they do anyway.

You can analyze the "true value" of books relative to whatever particular aims you have in mind. Just because there's nothing intrinsic to the books that specifies what those aims should be does not mean the effect of the books isn't in-principle measurable, and couldn't be judged for effectiveness once you have specified your goals.

You write "true value" and "whatever particular aims" in the same sentence. That seems like a contradiction to me.

I agree that you can define things to measure objectively about books. However, I submit that in the end judgment of quality will always remain a subjective matter.

It's not a contradiction at all. Remember that the original context was "Moneyball" - sussing out the "true value" of baseball players using statistics as opposed to basing decisions on less accurate gut (or social) assessments of value.

This use of "true" merely means "accurate" - corresponding with reality. The aims, in the case of baseball, are implicit, of course - the aim of any manager is to win games. In the case of books, people's aims will differ more. This means there's no single "true value" but for any given aim there will be some amount that book will contribute to it. This is usually hard to determine, but that doesn't make it subjective; it makes it that much more interesting if you are able to get at it.

Value is always subjective. Beane's analytic model doesn't calculate value it defines a model for winning games and analyzes players based on the model.