Goodreads Ratings Then and Now

[Yapyap]

Originally Posted by Geralt

I think you can trust Goodreads's ratings when there are 15000 ratings and the average is 4.5. You can be pretty sure that's a good book. Compared to 4.8 rating out of 50 ratings.

Same thing with amazon. If there's 1500 reviews and 4 stars, you can believe them. But if there's 5 reviews and 5 stars, there's a big chance that author has sockpuppet accounts or his mommy and her friends wrote nice things about his book

Agreed, definitely! The more ratings and reviews there are, the more you can "trust" the average ratings - while obviously keeping in mind that not every book people in general have loved will be to your particular taste.

But on Goodreads in general and for newer / not yet released books in particular, it really is helpful to keep in mind that there is no "enforced" system of ratings and while most people probably stick to the GR recommended rating system, it's not at all guaranteed - and that GR staff has shown they have no problems with, say, people giving five stars to a book not released for another six months that they're enthusiastically waiting for.

[NickyWithNook]

Originally Posted by Geralt

I think you can trust Goodreads's ratings when there are 15000 ratings and the average is 4.5. You can be pretty sure that's a good book. Compared to 4.8 rating out of 50 ratings.

Same thing with amazon. If there's 1500 reviews and 4 stars, you can believe them. But if there's 5 reviews and 5 stars, there's a big chance that author has sockpuppet accounts or his mommy and her friends wrote nice things about his book

I don't know. It shows the reviews aren't by sockpuppets, but it doesn't show that it's a good book. Just that it's popular.

Examples:

https://www.goodreads.com/book/show/...om_search=true

https://www.goodreads.com/book/show/...om_search=true

[Fbone]

Originally Posted by NickyWithNook

I don't know. It shows the reviews aren't by sockpuppets, but it doesn't show that it's a good book. Just that it's popular.

Examples:

https://www.goodreads.com/book/show/...om_search=true

https://www.goodreads.com/book/show/...om_search=true

Ratings on those title are 3.56 and 3.67. Neither I'd consider highly rated.

[HomeInMyShoes]

All of this makes me glad that my reading generally has me about a decade behind the curve. Plenty of time for the real reaction to the book to appear.

[issybird]

Originally Posted by Geralt

I think you can trust Goodreads's ratings when there are 15000 ratings and the average is 4.5. You can be pretty sure that's a good book.

Heh. There's a book with 284,000 ratings and an average of 4.1 stars, but I rated it 1 star, because it's godawful.

You can't go only by ratings, even a kajillion of them.

ETA: And another with over a million ratings and an average of 4.44 stars, to which I similarly gave one star and for the same reason.

[HomeInMyShoes]

^Inquiry minds want to know which ones.

[JSWolf]

Originally Posted by Yapyap

Goodreads staff has maintained from the beginning that ratings are for people to use as they see fit. It's a personal cataloging site as much as a social interaction site as much as a book/review database, and everyone can use it for their personal purposes in ways they prefer - which includes rating books by whatever personal criteria one has for their own book catalog.

But, if they see fit to not use rating for what they were meant to be used for, then they are useless. Useless ratings are not what we want.

[Melon Soda]

I have seen some books with an average rating of 4 but with quite a number of 1 or 2 stars with even one commenting it to be the worst book ever read (or something along that line). While I find an average rating more credible the more it has been reviewed, I try to keep it mind that ratings are all about personal preference. It helps when a reviewer writes in details about why he or she gave it a low rating ("unbelieveable Mary Sue character", etc) instead of just writing generic comments like "it sucks" or "awesome book!". That's why instead of just relying on the average ratings, I will also read a few of the more detailed reviews from the mid-range ratings (2-4 stars) for a more balanced view!

[BuyMore]

I always look for the 1 - 3 stars reviews when I'm trying to see if a new author is worth checking out. Sometimes the issues that bother them are a huge turn-off for me, too. In that case, I'll give up the idea of getting it.

Originally Posted by issybird

Heh. There's a book with 284,000 ratings and an average of 4.1 stars, but I rated it 1 star, because it's godawful.

You can't go only by ratings, even a kajillion of them.

ETA: And another with over a million ratings and an average of 4.44 stars, to which I similarly gave one star and for the same reason.

Wonder if it's the same book I've read...

A book that burned me deeply has a pretty good rating on GR :

Spoiler:

Beautiful DISASTER. Seriously, totally disastrous.

Originally Posted by HomeInMyShoes

^Inquiry minds want to know which ones.

You can use the spoiler tag just like I did.

[GeoffR]

The average rating for a book with a large number of ratings would only be useful to me if I had average tastes. I think a rating from just one person with tastes similar to mine would be much more useful.

What I would like to be able to do is submit a list of books I loved, along with a list of books I hated, and get a list of books that were rated highly by people who gave high ratings to the books I loved and low ratings to the books I hated.

[Geralt]

Originally Posted by issybird

Heh. There's a book with 284,000 ratings and an average of 4.1 stars, but I rated it 1 star, because it's godawful.

You can't go only by ratings, even a kajillion of them.

ETA: And another with over a million ratings and an average of 4.44 stars, to which I similarly gave one star and for the same reason.

Because your 1 star somehow cancels out 1.000.000 other ratings.
Yeah, we are all special snowflakes aren't we.

You do realize opinions are just opinions, but when 1 million people rates something with an average of 4.4, statistics come into play then, and what that tells you is that majority of those 1 million thinks this book is good, so there's a big chance that majority of other 1 million people who haven't read the book yet will like it.
It doesn't mean there won't be people who dislike it. That's not how statistics works.

Your personal rating of 1 star is totally irrelevant to 1 million ratings of 4.44
Of course to you it's very important, but for statistics sake, who cares.

[Geralt]

Originally Posted by GeoffR

The average rating for a book with a large number of ratings would only be useful to me if I had average tastes. I think a rating from just one person with tastes similar to mine would be much more useful.

We all have average taste, don't put yourself on a high pedestal.
The only question is average taste in what genre? That's the big thing.
A book can have great ratings by many, many people in say....romance genre and I wouldn't touch it with a stick, let alone read it. I wouldn't enjoy it.
But if a book has 1.000.000 ratings with 4.8 average and it's a sci-fi...well then, you got a deal. There's a big, big chance I will love that book, same as 1.000.000 sci-fi aficionados before me.

[GeoffR]

Originally Posted by Geralt

We all have average taste, don't put yourself on a high pedestal.
The only question is average taste in what genre? That's the big thing.
A book can have great ratings by many, many people in say....romance genre and I wouldn't touch it with a stick, let alone read it. I wouldn't enjoy it.
But if a book has 1.000.000 ratings with 4.8 average and it's a sci-fi...well then, you got a deal. There's a big, big chance I will love that book, same as 1.000.000 sci-fi aficionados before me.

No one has average tastes, the average rating is a combination of different ratings, just as the average number of children in a family might be 1.5 although no family has 1.5 children.

If the average rating of a large number of people reflected my tastes then I would agree with the average rating given to most books, when in fact I usually disagree.

[Yapyap]

Originally Posted by JSWolf

But, if they see fit to not use rating for what they were meant to be used for, then they are useless. Useless ratings are not what we want.

They're not useless for the users who rate books according to their own personal system. If they treat Goodreads as a cataloguing site, for their own personal book catalogue (and that is a valid use of the site), and only look at their own books and ratings, they're perfectly useful.

They're also useful for the person's friends and followers who are familiar with that person's ratings system.

It's only the accumulated average ratings that people's personally different rating systems aren't useful for - but as said above, with a sufficiently large number of ratings, it's not going to really influence a book's average ratings that much all in all if one person goes by a "one star is the best, five stars means it sucks" system or "one star means I am really looking forward to reading it" or "this was an awesome book and I enjoyed it, but it's no Nobel prize winner, so I'll give it two stars" (and I've seen reviews saying pretty much that).

Anyway, it's easy enough for me to not have issues with this as the value of GR isn't in the average ratings for me - it's in me being able to keep a good book catalogue, to read reviews from others, and to interact with my friends. For just ratings with no reviews attached to them, it's only my friends' ratings I pay attention to as I pretty much know their tastes by now.

[treadlightly]

Originally Posted by GeoffR

What I would like to be able to do is submit a list of books I loved, along with a list of books I hated, and get a list of books that were rated highly by people who gave high ratings to the books I loved and low ratings to the books I hated.

This! And Goodreads already has all of this information. Updating their recommendations should be high on their priority list. After all, it would likely sell more books, which I'm assuming Amazon (as one of their major investors/owners?) wants to do.