Saturday, 7 September 2013

Visualizing The Wheel of Time: Reader Sentiment for an Epic Fantasy Series

In the following blog post, I explore reader sentiment for the Epic Fantasy book series The Wheel of Time, as expressed in user-submitted ratings on Amazon and GoodReads.

If you're a data scientist (or similar), you'll probably be interested in the data analysis which includes some interesting observations about the usefulness - or otherwise - of Amazon reviews. If you're a Epic Fantasy Series reader, you'll be interested in the outcome of my analysis: I've decided to go ahead and read all fourteen books.

Note - This is a spoiler free zone
Updated 17th Sept: charts now more accurate, not quite as pretty

*

Recently, I was looking for a good book to read, and a friend recommended The Wheel of Time series by Robert Jordan. But I'd heard from a few sources that the later volumes were harder going than the earlier volumes. Having struggled with later volumes of the Game of Thrones series (sorry George) I was wary of starting a mammoth fourteen volume series unless I was confident I could make it to the end.

So I did some research. First, I checked the Amazon reader-submitted ratings for the books, which are on a scale of 1 to 5 stars. Here's what they look like for the whole series:


Uh oh, that doesn't look good. Books eight to eleven get some pretty poor scores, though the later books seem to pick up again. It looks like the middle of the series could be hard work.

What's going on?

Next, I checked the GoodReads ratings for the series. GoodReads is a site designed "to help people find and share books they love... [and] to improve the process of reading and learning throughout the world." Here's how the GoodReads ratings (which are also submitted by readers and go from 1 to 5) stack up against the Amazon ratings:



So that's a little different. There's still a hump in the middle, but it's nowhere near as pronounced ... in fact the lowest aggregate rating is 3.84, far higher than the 1.8 for the same book on Amazon!

Let's look at the number of reviewers for the two systems, corresponding to the number of people who've read the book and recorded their review and/or star rating. First, GoodReads:



Well that seems reasonable - the number of ratings tails off in the middle, then picks up towards the end as you'd probably expect given the ratings we've seen. And generally speaking, each book has lots of ratings - the lowest count is for the final book, probably a reflection that A Memory of Light has only been out for a year compared to the other books.

Oddly though there are more ratings for book twelve (The Gathering Storm) than for quite a few of the earlier books ... more GoodReads users rated that book than earlier volumes.

This likely reflects the sad fact that volume eleven (Knife of Dreams) was the final volume completed by the original author Robert Jordan, who passed away in 2007. Volume twelve is the first volume written by Brandon Sanderson, the author Robert Jordan chose to finish the series. RIP Robert.

How about Amazon reviewer counts then?



On Amazon, there are far more reviewers for the books that received the really low scores. This suggests that the really low scores are actually a result of frustrated readers motivated to express their concerns, rather than a reflection of relative enjoyability or quality per-se.

GoodReads makes it extremely easy to submit a rating for a book - one click is all it takes. Amazon seems almost to discourage reviews - the "Write a Review" button is halfway down the page, and you must provide a title and description for your review. The net result being that input of everyday browsing users won't be captured on Amazon - only motivated reviewers (such as the frustrated reader) will be bothered to jump through all the hoops.

*

Overall, therefore, it seems sensible to expect a dip in the enjoyability of The Wheel of Time series, from book eight to about book eleven.

But perhaps that dip isn't as severe as suggested by Amazon, whose ratings are likely skewed by frustrated readers. My guess is that many readers reach the later volumes and are frustrated by a change of pace; this certainly matches my experience with the Game of Thrones series where events seemed to slow to a crawl in the most recent books. The problem is compounded when there are long gaps between books being published, making it harder to pick up the story.

Thankfully, the final few books get much higher ratings across the board, so I'm expecting that it's worth getting through the slower books to reach the finale. At least, that's the story I'm telling myself ...

Only Time will tell.

26 comments:

  1. Enjoy the series, it is (mostly) pretty great ... and I think your analysis is really interesting! :)

    ReplyDelete
    Replies
    1. Thanks. Looking forward to it - just started volume 2 so I'm on the way.

      Delete
  2. Hi, James --

    I love the use of review data to look into the series, and I think some of the reasons for discontent can be found in the series' publishing dates -- the first six books all came out with a spacing of 10-12 months, while six to ten have spacings of 19-34 months apart. Nearly 23 years of the series, 13 of which are spent on the last third of it, can make some people grumpy -- I know I was! But having been there for all of it, from start to finish, I think it was worth it.

    ReplyDelete
    Replies
    1. Definitely makes sense ... I read the first three Game of Thrones books in quick succession, then came back to the fourth volume about ten years later. I couldn't remember much at all, except that there were wolves :)

      Delete
  3. I read the whole series and fully concur with the amazon graph, not due to frustration, but just due to pacing. Jordan seems to spend paragraphs describing someone's dress. It just drags on. like...

    We're just about to do this big exciting thing! No we're going to soon. We promise we're going to go after that right away! Here we go we're getting ready! Well wait I have to do this one small thing first. Well that took fifteen chapters, even tho nothing really happened. Okay in the next book we will totally do that exciting thing. Well, okay maybe actually in the NEXT book.

    That's a bit what it's like. It seems nothing happens for the middle several books.

    ReplyDelete
  4. Love the analysis. Glad you've decided to give the series a try. Here's a few other variables you may consider over the 23 year publication history.

    Fantasy series may be rated on the overall feeling towards the series as well as the individual book. Less enjoyable books may get an upward boost just because favorite characters are in them. Later books in a series may feel the drag of earlier disappointments, or get a boost from exceeding lowered expectations. I suspect the same applies to when a favorite author's name is on the cover.

    Brandon Sanderson completed the series and has a very prominent web presence. A bump in number of reviews for Book 12 may be related to his name being on the cover. Not that Jordan's fans didn't have obsessive websites built around his books, but the fanbase for a 21st century author may have different views on promoting books they like through Goodreads and Amazon, neither of which existed in the early 90s.

    In general, the decreasing number of ratings for each subsequent book probably shows people who have left off completing the series, or left off commenting after having rated the first book. Shouldn't the first book of a series always have more reviews than later volumes?

    Your conclusion about Amazon's system favouring extreme opinions seems sound. Our fan website's theory-rating system seemed designed to drive negative ratings, all but a handful never scored above 6.5/10.

    Not having to wait for publication gaps of several years improves the reading experience greatly. Goodreads may give higher rankings for older books because they are all 'old' by the time Goodreads came out, while Amazon's ratings benefitted from immediate fan disappointment that they didn't get more after a long faithful wait.

    There are a number of lengthy series out there, all with differing publication dates with respect to the rating tools available at the time. I suppose it would be feasible to contrast an old series vs a current one to examine some of the date/delay variables. Stephen King's Dark Tower also had a lengthy gap between books, from pre-internet to full internet era. You already mentioned GRRM, whose work is alleged to have suffered from the same disease as Jordan's.

    Please follow up on your post with your own ratings.

    Thanks for this wonderful analysis!

    Tyler

    ReplyDelete
    Replies
    1. Very insightful, thanks. Good idea on a a follow up - I'll start a log to record my reading experience, and follow up once I'm done.

      Also I like the idea of doing a deeper analysis covering other long running series, and taking publication dates into account, I'll have a think about how that could work.

      Delete
  5. I would say the drop in quality is as severe as Amazon makes it look. Suddenly you're reading books where the plot makes no progress. Literally nothing important happens. HUGE SPOILERS with worldwide consequences happen, and then aren't even mentioned in the next book. Be prepared.

    ReplyDelete
    Replies
    1. Some thoughts on the data. GoodReads data declines so that by the really bad books, it is likely a much higher percentage of really-invested super fans rating them. Some may just rate all the books in a series they like highly because even if they don't like the specific books, they like that TYPE of book, so it may be useful for recommendations (I've certainly done this sort of thing on Netflix). People (like me) who became completely disgusted with the books and gave up on them might not want to rate them at all on Goodreads.
      Regarding the uptick in readership in book 12, by the time the books around #10 came out, many people may have been so disgusted with Jordan that they wouldn't read anything else he wrote. I wouldn't. They may have been willing to give Sanderson a shot to see how it ended, though.

      Delete
  6. In my experience, people who would think that the Amazon stats paint an accurate picture are those who actually had to wait for the books. There are exceptions, but from years of reading fan anecdotes, especially in the years since RJ's death, it would seem that the lag doesn't usually bother people who can read straight through very much at all. What does bother these first-time readers is the cast expansion that starts to really add up with book 6. I would suggest using Encyclopaedia WoT as you read, but the spoiler-segregated version isn't finished yet.

    Also, it helps to know that the slower books are also shorter books. We know from interviews that RJ's health started to decline while he was working on book 6; after his initial 6-book, book-a-year contract was up, he wrote significantly shorter books in significantly longer periods of time. That contributed greatly to the impression that 'nothing happened' in those books—it's not so much that nothing happens as that working plotlines didn't get resolved as quickly (in real world time) as readers would have liked.

    I think GLOTD is correct that the pick-up for the last few books is at least partly due to Brandon Sanderson's web presence. Another factor was the fact that people have been waiting for this ending for 20 years. When RJ died, we believed we wouldn't get it, so that depressive element contributed to the eventual mania. I think it would have been similar if RJ had lived, but it's hard to sort out all the factors involved with any confidence.

    I would guess that the peak at TGS (12) has something to do with that manic-depressive element again, but it's probably more to do with the fact that people wanted to give Brandon a show of support (both long-time fans of the series and Brandonfans who wanted to counteract those who weren't going to be happy with anything less than RJ himself). There were some structural issues in TOM (13) that contributed to a dip in ratings; these issues were especially confusing to the more casual readers of the series.

    ReplyDelete
    Replies
    1. You may be correct. I wait a yearish for a new book, it's building towards some sort of climax that never arrives. I'm disappointed, but the next book must be awesome! A yearish later the next book arrives. It starts off slow, but seems to be building towards a big climax...which again never arrives. I give up.

      Delete
    2. Just as a personal counterpoint, I read books 1-6 all together, then the rest as they were released, and although books 7-8 were a bit slower, books 10-11 felt like milking the readers. When the author mentioned that the series (which at some point was supposed to go to 12 books) got split into 13 books (and later into 14 books), it really did seem to make the case for the "milking" theory.

      Delete
    3. The 'milking' argument has never made much sense to me. While it's obviously true that RJ slowed down deliberately after book 6, I really do think that was solely due to his health. He wanted to finish the series even more than his fans wanted him to finish it, difficult as that might be to believe. He wanted to be done with it, and move on to new ideas, but his health wouldn't allow him to spend 10 hours a day writing 7 days a week any more, and that's what it took for him to put out books like 4-6 once a year (that, and a bit of a buffer zone that he had when the first book was published which was depleted by the time he finished book 6). For the record, when RJ died, there was supposed to be one more book (book 12). Brandon signed a contract for one book, but after writing more than a book's worth of material, he realized it was impossible to cover the outline RJ left in one book. In retrospect, I think he could have done it in 2 books with better planning and more rigorous editing, but there is a good argument to be made that Brandon's style (which was already established) made it impossible for him to cover that material even in two books. At that point, I think one book would have been rushing it either way.

      Delete
    4. Oh, I was going to clarify that it was never going to be 13 books: book 12, which was to be the last, was split into 3 books. Brandon avoided talking about the split until it was officially announced, and it was 3 books when it was announced. So it went straight from 12 to 14. RJ was always optimistic about how much material he could fit into one book, and I think he really tried to keep to those estimations and failed every time. It's not a problem that is limited to WoT; this happens to a lot of authors who work in a similar way (i.e., from a rough outline, with most of the detail made up as he goes along).

      Delete
    5. One more thing: most long-time fans think that KOD (11) was a satisfying book. One might argue that's because COT (10) lowered the bar, but many fans put KOD in their top few favorite books of the series. Every now and then I see someone such as yourself lump KOD and COT together in this sense, but it's very rare.

      Delete
  7. I don't think waiting for the books was a major factor, but like others commenters, I felt the plot had virtually no advancement in books 10 and 11, and slow advancement from 6-9. It had a very strong feeling of the author trying to milk the readers out due to the immense popularity (for a fantasy series) of the early books. I couldn't bother reading the last few books, that's how discouraged books 10-11 left me.

    I'm sure the series could have been written, with no loss to story or character development, in 6 books.

    ReplyDelete
  8. I recently finished book 9 and am enjoying the series as much as I did at the beginning. Yes, you have to pay attention. If you do, it's worthwhile.

    I wonder how much the dip reflected worry that people had invested time and energy in the series and, as book was in the works, didn;t know if the series would or could be completed satisfactorily. I Know I was worried.

    ReplyDelete
  9. Amazon and Goodreads are also going to differ in how one perceives their own reviews. With Googreads it's easy to look over your whole library and see what ratings you've given everything, and you're going to feel more pressure to be 'consistent'. You may wince a bit more at giving low ratings on Goodreads because in that interface and context, it will look harsher. Whereas Amazon's primarily there to buy things from, so your ratings aren't as front and center as they are on Goodreads. You're expected to be more critical there because you're there to tell people whether something is worth their money.

    In fact, that separation makes it clear - Goodreads is about appreciating books, Amazon is about parting with or saving money. It makes sense that a different mindset will come with ratings on each site.

    ReplyDelete
  10. This is what happens when you marry your editor and she doesn't want to hurt your feelings by cutting anything.

    Robert Jordan "planned" on the series as a trilogy! Clearly he had trouble deciding when to cut content.

    Don't get me wrong, I loved the series. But I'd be able to convince more people to read it if it wasn't so darn long.

    Now fold your arms under your breasts and sniff at me if you must. :P

    ReplyDelete
  11. Books 8 to 11 are indeed crap. A sudden deluge of uncharismatic characters that have no effect on the plot.

    I think the peak for me was around 5 or 6, but then the graph doesn't reflect that.

    ReplyDelete
  12. No, I think the Amazon ratings are actually pretty accurate, at least from my readings of the books -- although I only made it a few chapters into book 11. Books 8, 9, and 10 were just exceptionally slow, and felt like a total waste of time to read. I actually re-read the series twice in high school, up through book 7 then book 8. For books 9 and 10 I didn't re-read the series, and then on book 11 I gave up.

    ReplyDelete
  13. Nice post! Made me think, perhaps trying to calculate the standard deviations on the grades would help giving a more nuanced picture?

    I made a small python script (pastebin) that makes these images (imgur). (feel free to do whatever with these...)

    I tried amazon.co.uk in addition to amazon.com, the trend is the same there. Including the 1 sigma error band makes the dip seem a little less significant, but especially amazon.com is really dipping. It is also interesting how the amazons show very similar behaviour - indicating that it is the amazon experience rather than the amazon.com readers causing some of the dip... Still book 10 is low even on goodreads.

    From the fraction plots you see that clearly the first and last books are the most popular ones.

    ReplyDelete
  14. I loved the first 6 or 7 books, but as the series wore on I began to feel that the series was never going to end. They could have and should have wrapped up the series much earlier, but instead decided to let it drag on endlessly as a cash cow.

    I decided to pull the plug on everything by book 8 and while I hate not knowing how everything resolves, I'm content that I'm finally free of the seemingly endless series.

    ReplyDelete
  15. The first time through I also had the impression that the plot was crawling along midseries.
    Now that I've reread the entire series (up to book 13, awaiting the paperback release of number 14 coming january), I actually relished the middle books. Yes, the overall plot didn't get on like a blockbuster movie but there was a bleepton of character development, foreshadowing and events creating this uncomfortable sense of danger and despair leading up to that certain event at the end of book twelve. I wonder if your opinions also differ after the first and the second reading?

    ReplyDelete
  16. I like the charts they're pretty accurate compared to my own WoT reading experience. Books 1-6 were fantastic. 7-10 were definitely struggles, 10 being the low point of the series. Things start to pick with 11 and finally Sanderson finished off the series (12-14) to the best of his ability.

    ReplyDelete