It's always worth checking the distribution. A 3 star rating with a normal-ish curve is very different than a 3 star rating with a lot of 5s and 1s -- and in that case the reviews will tell you clearly whose preconceived worldview the content fits.

That's a good point -- I wish we could see Sparklines-like histograms of reviews.

On a related note, for binary (good/bad) reviews, we can the temper the results by assuming a beta(1,1)-distributed prior and then updating [1]. The expectation (which can thought of as an "adjusted average rating") ends up being:

E[x] = (good + 1)/(good + bad + 2)

This adjusts for situations where the number of reviews is small, in which a good average rating can be misleading.

[1] https://www.johndcook.com/blog/2011/09/27/bayesian-amazon/

See Tweet by Paul Graham (https://twitter.com/paulg/status/1085484183466713088):

"If Newton's Principia were published today, it would have 4 stars on Amazon. There would be one cluster of 5 star reviews by people saying it had revolutionized their thinking, and another cluster of 1 star reviews by people complaining it was pointless and hard to read."

and then someone linking to actual reviews on Amazon (avg. rating 4.1).

Especially since one of the books he mentions, Steven Pinker's How the Mind Works, has a 3.97 - which just barely misses the criteria, and yet has a big fat red X. He gave rated it 2 stars, so this seems to be confirmation bias more than anything.

The rule reminds me of the hiring mentality of "it's worse to make a bad hire than to filter out a good hire."

There are so many books available to me that I do use Goodreads ratings as one heuristic for choosing what to read. It would take a lot for me to read a 2 star book (like a very strong recommendation from someone I trust), not because I am sure it wouldn't be worth my time, but because there are plenty of more highly rated books that are more likely to be worth it.

The author even qualifies the rule: "Unfortunately, Goodreads ratings are often not unbiased.."

Quality assessment is not necessarily the same as popularity, surely.

Most people rate books not based on rather they are good books, but rather if they enjoyed the books and believed in what it was saying. Furthermore, there is a certain demographic of people who uses and rates nonfiction books on goodreads, which will skew the rating.

In an online forum like goodreads (ditto imdb, etc) it most defnitely is.

> Reading only very popular books is a recipe to read only books that fit into your own preconceived worldview.

... under the assumption that everybody who rated the book shares your preconceived worldview. Assume that not to be the case on Amazon.com (or similarly large review platforms).

I think it's more laughable for fiction books and less laughable for nonfiction books. Nonfiction books are like tools to me: they either perform or they don't.