I Ought to Write Romance; a Statistically Bankrupt Analysis

I used to study and write poetry, a fact that is relevant to this post in two ways. First, it helps to explain why the statistical methodology to follow is so shoddy, so truly terrible that it would make any self-respecting statistician gouge out her own eyes, Oedipus-style. Second, this post helps to explain, in small part, one of the reasons I stopped writing poetry and started with the epic fantasy.

Every year Goodreads hosts the “Readers’ Choice” awards, which the site bills as “the only major book award decided by readers.” That claim depends, of course, on what you consider “major”, but with nearly two million votes cast in 2013 competition, it’s fair to say that Goodreads is running a big competition. There are twenty categories, with fifteen books nominated in each. I’m thrilled to be nominated in two categories: Best Fantasy and Best Goodreads Debut Author. I also thought this was a perfect opportunity for someone who hasn’t done math since high school to do some math.

Specifically, I was curious to see if there was any difference in the average number of reader ratings per book by category. These ratings have nothing to do with the Choice awards – they’re just the number of ratings each book has accrued since its release. Still, it seemed like the nominations offered a sort of snapshot of each genre.

Of course, there are problems. The books are released at different points in the year, for one thing, and obviously books released in January have more ratings than those released two weeks ago. It’s very possible that readers of certain genres might be more likely to use Goodreads than others. Ratings might not translate well to sales. I accept my D- for data analysis.

Still, check this shit out:

Category        Average Number of Ratings per Nominated Book

YA SF/F:                    23,726

Romance:                   14,871

Fantasy:                     9,957

Fiction:                      9,554

SF:                              5,269

Non-Fiction:             2,591

Poetry:                       242

The discrepancy is pretty astounding. It seems to suggest that writers of Young Adult Fantasy and Sci-Fi are just absolutely killing it, sales-wise, compared to anyone writing what we might call (for lack of a better term) Adult-Oriented Sci-Fi and Fantasy. The YA title with the most ratings was Cassandra Clare’s City of Heavenly Fire, with 73,629 ratings. This is half the number of ratings all fifteen nominated adult fantasy titles combined. The only other genre that even comes close to these numbers is Romance.

That said, I think this is good news for those of us writing non-YA fantasy. I, for one, am delighted to see the YA market so robust. I like to think that when these kids get older, they’ll keep reading fantasy.

And then, of course, there’s the plight of non-fiction. After the breakout success of such books as The Perfect Storm, and Longitude, I expected non-fiction to have a much higher readership. Maybe it does in other years. Maybe non-fiction readers just hate goodreads. I dunno. Again, the holes in my method here are massive, but these numbers seem bleak.

Finally, poetry. It wasn’t until I was in graduate school writing poetry that I started to really understand it would be almost impossible to make money in the enterprise. I admit that this was pretty late for what should have been an obvious realization, but I love poetry. I still write the occasional poem. My son and I recite poems before bed every night. If you want to make a living as a writer, however, and I do, it’s not a viable path (unless you couple it with teaching or prostitution or something). Fortunately, my other great literary love, the writing of epic fantasy, offers more possibilities.

I’m delighted about the nomination of The Emperor’s Blades. Both slates (fantasy and debut) include some incredibly stiff competition, but it’s exciting to find my name in the fray alongside the likes of Rothfuss and Sanderson. I hope some of you wander over to the site, check out some of the categories and titles for yourself, and let me know what you make of all this data. Where have I gone horribly and wildly wrong? Or do you think that, warts and all, this is actually a pretty accurate picture of the current publishing world? Finally, please vote, either for The Emperor’s Blades or another writer you’re excited about. I love doing this, I’m incredibly grateful to be making a living at it, and I couldn’t do it – no writer could – without you.

12 thoughts on “I Ought to Write Romance; a Statistically Bankrupt Analysis

  1. I guess one would have to do some guess work (social scientist would frown upon “guess working data analysis”) but I guess (there, I did it), the numbers reflect and tell us much more of the nature and the demographics of the platform (goodreads) than of the actual readership/sales of books.
    I guess someone who actually reads poetry (as in, “let me just grab this Complete Works of W.B.Yeats before I go to bed”) won’t fit the bill as to the typical Goodreads user. On the other hand, YA Fantasy seems to atract the right demographics for the app (“let me tell my friends I’m reading this / I’m gonna share this on Facebook”).
    Either way, this is of no importance when compared to your nomination: voted for you on both categories. Still reading TEB and it’s absolutely killing me. It’s one of those books that make me doublecheck my watch to see if I’ve got reading time available. Great work!
    Best of luck and thank you!

    • I suspect you’re right about the importance of the platform. My wife was making this point last night, when we were chatting about it. And thanks so much for the kind words about TEB. Let me know what you think of the ending when you finish. Also, your timing is good. Just a couple of months until Providence of Fire drops… All the best, Brian

  2. Congratulations on the nominations Brian!

    You’re right in suspecting that goodreads is not quite a good representation of the publishing world, specially on the misrepresentation of older people (those who don’t use the internet much). Maybe I’m being a little judgemental, but I would have guessed that many older people read more non-fiction, as we’re seing in the data.

    The thing is that for new authors (like yourself) the internet is most likely the way you’ll get discovered (I for one know you because of Tor), so even though the data may not be representative of the whole world, it may be representative of the people a new author is likely to get to unless he has a struck of luck (mainstream media attention for some reason). But I digress.

    A common mistake from non statisticians: the mean is not enough. Maybe there are things missing in there that would clarify stuff for you. For example, I see Sanderson in the Fantasy category, isn’t he getting the mean really high just because of his previous popularity?

    Also for a “mouth feeding genre” analysis you would need to take into account the amount of books that got published. Every category has 15 nominees, but I don’t think that there are the same books published in the YA category than in poetry. You made it to the list because lets face it, your book is awesome. But there are a lot of authors that got left out. Maybe in the poetry category every single publication got nominated (I’m exagerating for the sake of the argument) so, in average, authors get less attention in fantasy.

    If you want to play more with the numbers, that’s what I’d do next. You don’t get a D-, because you were interested in even collecting the data and playing with it.

    Good luck and again, congrats!

    PS You could also get the number of REVIEWS instead of ratings and see what happens.

    • It would be really interesting, as you suggested, to look at reviews or even the ratio of reviews to ratings (seems like that might be a proxy for the passion readers feel for the book, negative or positive).

      I also wonder what Goodreads’ selection criteria for the nominations was in the first place. Obviously, some balance of release date, number or ratings, and average rating. Still, it seems like some books (Tower Lord) would have made the list, but did not…

  3. Most book bloggers are teens or young millennials, so it makes sense that the demographics on Goodreads is similar. I’d say 98% of my contacts on Goodreads are either teenagers, writers, or librarians.
    I am 26 and graduated from a writing program, yet MOST of my writing and reading peers hadn’t heard of Goodreads until I told them about it. These are people who are online constantly. So it’s not a stretch to say that 35-year-olds and up are underrepresented in the Readers’ Choice Awards.

    • Thanks for the perspective, Lara. I’ve started asking people, just out of curiosity, if they’ve heard of goodreads, and I’m finding exactly what you are: most people I know, even those who read a LOT, have never heard of the site. When I explain what it is, they don’t seem to be interested…

      • I never got book recommendations as a kid, so I LOVE what Goodreads has to offer. But I read all genres. If I only read romance or mysteries or science fiction, I wouldn’t need Goodreads—I’d just pick the next one on the shelf.

  4. Hey Brian I loved your book the Emperor of Blades. It’s not in my top 5 but I still thought it was pretty good. I’ve read the excerpts from Providence of Fire and they were superb. I’ll be reading it when it comes out. Quick Question – If the Csestriim live in the vaniate their whole lives, what gives them motivation then? Even the need to exist is based partly on the fear of death. If the Csestriim have no fears, loves or sorrows, what moves them to achieve something?

    • Thanks for the kind note and for the question about the Csestriim. This has been a subject of much discussion in the Staveley home over the last couple of years. The crux of the problem lies in the words themselves: need, desire, motivation, etc. When I was teaching Buddhism, the kids would always point out on day one that the Four Noble Truth (“To eliminate suffering, eliminate desire.”) seems to make an impossible demand. After all, if you have no desire for ANYTHING, why eat, why move, why take any action at all, including paying any attention to the noble truths themselves? To resolve the problem, you need to go back to the original Pali, where the word that gets translated as desire is tanha. It means, literally, thirst, but encompasses in this Buddhist usage, all manner of selfish, acquisitive desires. It exists in distinction from chanda, which covers non-selfish desires (helping the poor, following the eight-fold path, etc). Since both words get covered by the single English term, desire, serious confusion arises.

      I imagine a similar thing with the Csestriim. They “want” things — more knowledge, survival, etc — but not in the way that we want them. To put it differently, the levers of their desire are different from the levers of our desire. Fear does not move them (no emotion does), but they can feel a compulsion or impulse to act in certain ways. You could imagine a computer program tackling some difficult problem of chess, for instance — it doesn’t “want” to do it in any way we would understand, but it does it all the same. This is only an analogy. The Csestriim aren’t computers. But I like trying to imagine utterly alien sources of motivation.

      Sorry for the long answer! This is just a topic that I enjoy puzzling over. Thanks again for getting in touch!

      • That’s a fascinating answer. Hinduism and its derivative religions Buddhism and Jainism have always given importance to the separation of oneself from worldly desires. Unlike Abrahamic religions, which foxus on heavenly bliss, Hinduism based religions have always given importance to separation of oneself from his/her emotions and accorded it the highest status. But I would consider that as a bit of a paradox. See, I’ve always viewed a good person as being a person whose self-emotions is invested in others while a sociopath, the opposite end of the spectrum is someone whose emotions are utterly uninvested in others. This is why, for example, people say doing charity makes me happy. An addict is someone whose self-emotions are invested in objects completely. In fact a gambler is said to experience same highs as a cocaine addict. But you see, in my worldview, neither the good person nor the sociopath nor the addict are self-less, their self-emotions are invested in different things. If you think about it, the only truly self-less thing is a computer. The only important thing to a computer is the orders you give it. It does not, of its own, exert a choice either for or against the order. I would even go so far as to say that emotions are the reason why despite being highly intelligent, having a perfect memory etc, computers aren’t sentient. Another question though about the books though, we have seen humans exerting themselves to achieve the vaniate. Will we ever see Csestriim trying to achieve emotion? That would be REALLY fascinating.

  5. Really enjoyed your books, BS. Looking forward to the third. Actually chatted about the size of YA fan/sf recently with a buddy of mine who is a literary agent, he claims the YA category in general is notoriously unreliable. It is stocked with all sorts of books that likely would not make the cut into fantasy or scifi if they were targeted to adults. Thus, the seemingly enormous size, it is more of a catch-all category. Reality is, adult books are more sophisticated in their categorization, YA much less developed. That being said, it is certainly a large category, and I’d love to see the category parsed with a bit more consistency vis-a-vis adult fiction.

    • That’s an interesting observation, Nathan. I’d never really thought about it that way, but now that you mention it, it seems obvious. YA sweeps up everything from fantasy to “literary” fiction to mystery novels — all genres that, in their adult incarnations, get separate shelves in the bookstore. Thanks for passing that along!

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s