In the past, I’ve relied on Locus magazine’s list of SF/F published over the past year as the basis for digging into Goodreads ratings of current SF.  But pulling data just once per year isn’t that great, so here’s my latest solution.

Goodreads allows you to have multiple accounts, so I now have this:

• http://www.goodreads.com/sffselect

Here’s the profile description:

• I’m selectively tracking most SF/F novels as they come out, creating sortable shelves of new releases such that their stats can be easily compared.
• As a matter of personal preference, graphic novels, game-related fiction, non-fiction, reprints, urban fantasy / paranormal romance, and young adult novels tend to be excluded (often by consulting the shelves other users have put them on).
• However, I include them occasionally if I have a subjective impression that the book has crossover appeal or if I just kind of feel like it. But that’s pretty rare.
• This account is not itself used to rate books.

Here’s how to use it:

• Click on the 2012 bookshelf.
• Along the top, hit shelf settings and select Author, Avg Rating, Cover, Date Pub, Num Pages, Num Ratings, and Title.  Unselect everything else.
• At the bottom of the page, select Infinite Scroll and sort by Avg Rating or Num Ratings, descending.

Advice on using it:

• Obviously, you need to show the “Num Ratings” column to interpret the ratings at all.  Ignore the rating when few users have contributed to it.
• Heavily discount the avg. rating of sequels.  Their ratings are demonstrably very inflated relative to standalone or initial novels in a series.
• Also discount the avg. rating of fat / epic fantasy.  Their ratings are demonstrably very inflated relative to the overall values of neutral shelves such as fiction or 2012, possibly because they tend to have many sequels.
• You’d have to discount urban fantasy / young adult heavily too, if there were much here.  Their ratings are demonstrably very inflated relative to the overall values of neutral shelves such as fiction or 2012, possibly because they tend to have many sequels.
• In my experience, the mean rating for a large selection of theoretically popular fiction on Goodreads hovers around 3.75 with a median noticeably higher.  So 3.75 and below is probably a reasonable point at which to say a book was only well-received by a small audience.
• In my own ratings, 25% of the books I really liked had ratings below 3.75, which is plenty of reason to investigate books below that threshold.  However, the numbers I’ve given to fiction in the past suggest that, in my case, I should spend about twice as much time investigating books rated over 3.75.

N.b.: I realize that audiences have characteristics of their own that strongly influence the significance of genre labels and the practice of assessing a title with a rating.  Furthermore, I’d be the first to acknowledge the limited utility of Goodreads ratings.  But if someone were to try using Goodreads to assess urban fantasy, paranormal romance, and young adult novels, I think they’d be wise to exclude sequels completely, because they dominate those genres, and the first in the series is probably the most indicative of what the widest audience thought.  Looking at them from an annual point of view is probably not useful to very many people–even those who tend to like work in the genre.