A Rough Overview of Novelupdates, and possibly Asian Novel Fan Translation and other things

This started last summer when I wanted to know the proportion of Chinese novels being translated with female main characters versus male main characters. I tried to do that manually, it didn’t work. I still don’t know that ratio, but after learning some new skills, there are other things that novelupdates.com, with its amazing directory, can show.

Dataset: Obtained from novelupdates.com on 2017-08-22 at approximately 11pm UTC. The entire data set of translation projects has 2888 entries.Some entries are be incomplete which means they are included from some analyses but included in other ones where they do have data. Some entries which may have releases are not included such as dead links, or releases with no date/group/url attached to them.

The total number of novel translations over the years

(I would really love for someone to go and put a language for the two projects that don’t have a language of origin. “Storm of Prophecy” and “Heir of Kiri”)

Overall number of novel translation projects colored by language of origin. (Bar plot)
Figure 1. Overall number of novel translation projects colored by language of origin.
2003 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017
Number of Novels 1 3 4 5 9 13 35 50 90 100 366 774 1170
% Growth Rate 33 25 80 44 169 43 80 11 266 111 51

The figure above shows the overall increase in the number of projects, indexed on NovelUpdates, according to the year of first release. The total number of translation projects is 2888. The number of new projects started each year has increased, though particularly high growth occurred in 2011 and 2015 with a 169% and 266% increase in the number of new projects over the previous year, respectively. Growth has also been uneven, with particular lows in 2014 and 2012 relative to preceding and following years. 2017 is the year with the highest number of new translation projects started even though the year has not yet finished.

Cumulative Monthly Growth

Figure 2. Number of newly started projects by month, and colored by language of origin.

The figure above shows the cumulative number of novels on a month to month basis. The three most translated languages are Japanese, Chinese, and Korean.

Chinese, Japanese, and Korean Monthly Growth

Figure 3.

Of the three most commonly translated languages, Japanese novels are still the majority of projects translated due to a sustained growth starting in the late 2000s. Chinese novel translation projects started increasing rapidly in 2015 but has only surpassed Japanese projects in the number of new projects started each month in July 2017. Korean novels, the third largest language category, is showing increased growth in recent months but is far behind Japanese and Chinese novels.

Completed Projects

Language Total Completed Incomplete % complete
Chinese 1130 149 928 13
Filipino 25 10 14 40
Indonesia 12 1 10 8
Japanese 1558 201 1160 13
Korean 144 7 126 4
Malaysian 3 1 2 33
N/A 6 0 2 0
Thai 8 0 7 0
Vietnamese 2 1 1 50
Figure 4. Completed versus ongoing translation projects for each language category.

In total, there are 370 completed projects out of 2620 projects which had both a first release date, and last release date. Novels without either were excluded from the analysis. Of the three most common languages, Korean translation projects have the lowest completion rate while Chinese and Japanese projects are similar.

Figure 5. Completed projects by year of origin and colored by origin language.

Of all 370 completed projects, only 267 had a year in country of origin. In the Novelupdates database, the oldest novel with a complete translation is “The Book and the Sword” by Jin Yong, followed by most of his other works. The oldest Japanese novel translation is “Greetings of the Universe” by Hoshi Shin’ichi. The oldest Korean novel to be translated is “Invisible Dragon” by Duchibak. In more recent times, most completed works were Japanese, and the majority of completed novels were works from the early 2010s.

How long does it take for a project to be completed?

Language Days
Chinese 112.55705
Filipino 54.30000
Indonesia 142.00000
Japanese 137.10448
Korean 47.57143
Malaysian 9.00000
Vietnamese 0.00000

On average, it takes 122.5837838 days for a project to be completed. Of all languages with more than one completed project, Korean projects are the fastest to be completed with an average of 48 days, or approximately 7 weeks. Chinese, Indonesian, and Japanese projects usually take over 4 months to complete.

Figure 6. Time to complete from first release to last release for all completed novel translation projects.
Figure 6. Time to complete from first release to last release for all completed novel translation projects.

However, 103 of the 370 novel translation projects were completed in one day. This may occur for a variety of reasons such as short stories. Excluding these projects that are completed within one day, most projects take longer than a month to be completed. The novel that took the longest time to translate was “Love Times” by Konohara Narise at 1799 days or roughly 4 years and 11 months. The Chinese novel that took the longest to be completed was “Half Prince” by Yu Wo at 1710 days, or approximately 4 years and 8 months.


Top Original Publishers

461 novels did not have an original publisher filled in. Some novels had more than one publisher. Works with multiple publishers were not included as it may inflate numbers due to double-counting.

Original Publisher Works
Syosetu 471
Qidian 368
jjwxc 133
Ascii Media Works 125
17k 80
Kadokawa Shoten 75
Fujimi Shobo 74
Munpia 72
Media Factory 69
zongheng 60

Syosetu is the most popular Japanese original publisher, Qidian is the most popular Chinese original publisher, and Munpia is the most popular Korean original publisher. Additionally, Munpia is the only Korean publisher to make it onto the list of top ten.

English Licensing Status

Language Works Licensed Total Number of Works % Licensed
Chinese 162 1130 14.336283
Filipino 10 25 40.000000
Indonesia 1 12 8.333333
Japanese 250 1558 16.046213
Korean 9 144 6.250000
Malaysian 1 3 33.333333
N/A 1 6 16.666667
Thai 0 8 0.000000
Vietnamese 1 2 50.000000

The majority of works licensed in English are Japanese works. However, Chinese, Japanese, and Korean novels have the lowest percentage rate of licensed projects to total projects.


English Licensee

Licensing Party Works
Qidian International 72
Yen Press 45
Wuxiaworld 27
June 19
Tokyopop 19
Volare Novels 19
Viz 15
Gravity Tales 12
J-Novel Club 11
Seven Seas 9

There are 33 unique licensees. Of the top ten licensees with the most licenses, six (Yen Press, June, Tokopop, Viz, and J-Novel Club, Seven Seas) publish Japanese projects. Qidian International, Wuxiaworld, Volare Novels, and Gravity Tales are English publishers of Chinese projects.


Most popular month to start a project

Figure 7. Frequency of projects started for each month for each of the three major language categories.
Figure 7. Frequency of projects started for each month for each of the three major language categories.

The most popular time to start a project overall is in August. The general trend for the three most common languages is higher starts during the summer months (June-August), most likely due to the end of the school year. There is a decrease in the number of projects started in September, also a possible result of the start of the school year. One factor to keep in mind is the number of new Chinese novel projects that have been started in recent months. They may be inflating the summer month values compared to later months that have not yet passed.


Tl;dr: The number of novel translations is going up, possibly doubling in number this year, the number of new Chinese translations has surpassed the number of new Japanese novel translations in July for the first time in a long time. The mean time for a translation to be finished is approximately three months. Syosetsu is the original publisher with the highest number of works translated. Qidian International holds the largest number of English licenses.


And now, translator ramblings.

Way back last August was when I decided to manually sample and see the proportion of female to male main characters. It was too time consuming and I gave up. Earlier this week, I had some free time, and tried to get data from novelupdates. The tags were too intimidating so I went for stuff that was easier. The data set is missing a lot of stuff. For example, if the links were dead and hidden, I did not get the release dates. Or if they were just in text and didn’t have a link, I didn’t get those releases either. At the end of the day, no database is going to be perfect, and I don’t have the skill to copy all of novelupdate’s data to analyze.

Some more interesting things I don’t have the time to look at: Most popular update times, the number of projects per group/individual etc etc.

Liked it? Take a second to support Dreams of Jianghu on Patreon!
Become a patron at Patreon!

37 thoughts on “A Rough Overview of Novelupdates, and possibly Asian Novel Fan Translation and other things”

  1. This takes me back to my papers lol.. Not a fun time

    I find Chinese translated novels, even with their constant time travel troupe as more superior than j or k novels. The translated j novels just seems so juvenile for my taste, tho I really enjoy Kenkyou kenjitsu as a gem, and can outweigh tons of c novels.. And I haven’t found any k novels that entice me to start reading

      1. Thankfully yes! And I get horror flashback reading your analysis lol

      2. Haha, it would be an easier read if I made it more of a story, but I can’t find the motivation.

    1. I find K-Novels are a nice middle ground between JP and CN novels. CN novels usually have stupid, brutal, unfeeling protagonists that will wipe out their enemies to 7 generations or whatever and aren’t below raping and pillaging all the time, whereas JP novels usually have spineless, stupid MC’s that can’t do anything alone and are constantly crying (both with exceptions obviously). KR novels like BEM in my experience tend to have level-headed, smart MC’s with the badassery of CN protagonists and the emotions of JP characters. They usually have pretty decent comedy (as compared to CN novels where it’s typically nonexistent (WoC is an exception to all these!!!) or JP novels with horrible unfunny cliches e.g. MC walks into bath and gets slapped by heroine, MC is super dense, etc) KR novels tend to just be generally better (though unfortunately they focus a bit too much on the comedy aspects).
      I would recommend Reincarnator (a bit like a CN novel with brutal/badass MC), The King of the Battlefield (MC is a little similar to Reincarnator MC), Dragon Maken War (MC is hero, but not like typical spineless JP hero who hides his identity and doesn’t really care about saving the world or whatever for some flimsy reason), Overgeared (just get past the first 100 chaps for the meaty bits), Book Eating Magician (overall a pretty decent read but a little bit of Harem/JP tropes), and a personal favourite, The Tutorial is Too Hard (awesome MC and characters, character development, interesting plot that keeps you guessing, so on so forth).

  2. Wow, that must have been a lot of work! I too would be interested in the ratio of male versus female protagonists. But yeah, thank you for your hard work xD

  3. Wow this is really interesting. I’d love to see the female/male stats as well. Although I rarely read stuff aimed at guys, haha….

  4. Wow, half prince that brings back some memories.

    Also it might be worth mentioning that the highest publishers for Japanese (Syosetu), Chinese (Qidian) and Korean (Munpia) are all for web novels only (or at least I think they are).

  5. I have to say – despite some of my favourite works having a female lead – I very rarely get to read something good with a main female character. Often I find this is because whoever has written it whether it be a Chinese writer or English, they get too caught up on the fact that they’re finally doing something with a female lead which can cause many other parts of the writing to fall short.
    Has anyone else noticed this?
    And can you reccomend some works that avoid such problems?

    1. “they get too caught up on the fact that they’re finally doing something with a female lead which can cause many other parts of the writing to fall short”

      With English writers, I’ve seen this happen occasionally (particularly in YA literature), but it’s really not a big issue in traditional (i.e. not self-)publishing because of the inherent quality controls there. Writers are pushed (and push themselves) hard to grind and smooth their weak spots–if they want to get published/win a readership/retain said readership/increase their sales.
      With Chinese writers… Yeah, I agree 100% with you: I rarely get to read a good C-novel with a female MC these days. But as I see it, the reason for this is absolutely *not* because there do not exist fantastically written female-led C-novels–but because these only rarely (as in: almost never) get picked up for translation.
      Translating is difficult and time-consuming. Translating good, complex writing is a herculean task. So I completely understand why [most] fan translators would rather go for the easy choice when picking their projects. So no wonder the proliferation of these 2000+ chapter long novels in the translation community, novels which in most cases (there are always some exceptions, and I don’t mean to offend anyone) are the worst offenders when it comes to subpar quality (not quality of translation, but of the actual story content). The common traits of |most| of these newly ubiquitous novels? For one, their chapters are [very] short, a perfect fit for the great-sounding 4-7+ promised releases per week framework that’s increasingly popular these days (and further fitting well into the sponsorship framework). 4-7 chapters of a half-penny novel are easier to translate than even just 1 chapter of a layered, complex novel, and 4-7 weekly chapters |sounds| so much more impressive than 1 weekly release. Second, the character are cardboard cutouts, the plot advancement inconsistent (or non-existent). All of this comes packaged in simplistic, repetitive writing–which again, is less taxing on a translator’s brain cells and time when translating. So yeah, no wonder.
      Still, I suppose we’re grateful for any new translation project these days, because it leads to the growth of the community and thus hopefully to an increased possibility of better novels picks in the future. Not to mention, the translators are mostly fans (not professionals) and they should be able to just translate whatever they want to translate in their free time–and if their projects are able to hit on the easy AND lucrative combo, more power to them. The people who are translating the layered, well-written novels and/or are translating purely just out of the love for a good novel/the community/… are also sadly the people who will most often quit along the way, because of how much harder to translate complex writing is, and because of how much harder it is to do something for free longterm.
      So yeah, that’s sad and frustrating for everyone. I know it’s not feasible for readers to persuade/chocolate-bribe/… translators into translating the novels they |really| want translated (which, for me personally, would be novels like those on shushengbar, like Minglan, and Rise of Phoenixes and Drunken Exquisiteness…, and not 2000+ chapters long novel which will probably never get translated until the end anyway–and even putting aside the question of quality, what’s a story without an ending?). But I really wish chocolate bribes could work. Or that professional publishing would enter the market. God, that’d be a dream come true. They’d only pick novels which were published in book format in China (already an inherent quality-control), the half-penny web novels would be deservedly ignored, and I would be so happy to buy a good quality, complete translation of novels like the ones on shushengbar that I mentioned before. A “Minglan” copy of my very own! And “Golden Age Legitimate Fei”! And and…

      1. Thanks. I didn’t properly realize how much I’d written until I hit “post comment,” and then I was like Eeek!! Then I wondered if the comment came across as too much of a rant, which was not intended… Oh well.
        P.S.: I love the translation projects *you* choose, ChocolateCosmos. 🙂

      2. I enjoyed reading your commentary. Looking back since I first found about Shusheng Bar was around 4 years ago? (I realized how long I’ve been reading novels…) Back then there were barely any translations and most of the people who did translations were mostly sticking to modern version or doing teasers/previews of ancient series. If not, they would go for scene translations (which I’m grateful for). I think Koala who did the translations of Tong Hua’s novels along with cdramas adapted from novels (by now it’s has become so common to adapt a novel into a tv show) encouraged many translators.

  6. You used ggplot!!!
    This was such a satisfying read. I must really thank you for this post, though. I chose the Data Analytics elective this sem and this has given me some nice ideas for my project.
    I’m also pretty motivated to do some practice on novelupdates datasets.

      1. Yes, I used ggplot. Prettier than base r since I don’t want to spend time finding colors.

        I did a really quick and dirty scrape of novelupdates. You can scrape using rvest. Also by Hadley.

  7. A great amount of those chinese novels have hundreds of chapters, so its impossible to complete in a short period. So those percentages are quite deceptive :p. Another graphics that would be nice are: “Average number of words per novel”, “Average number of chapters per novel” and “Average number of translated chapters per month”
    Great job with all those graphics!

    1. You won’t be able to get average words with just novelupdates data. I didn’t want to use chapters per novel because the novels are ongoing and a lot of them have volumes rather than chapters. Average updates per month is already something that NU has.

      1. Create a bot that enter every site on NU, count the words and associate it to the novel title saving it in a database so you can create the graphics XD
        Im just kidding :p
        Great job!

  8. While I’m late to when this post was made, I would like to inform a mistake you made. You stated that invisible dragon was the oldest Korean novel that was translated. That statement is wrong. Dragon Raja and The Stone of Days are both 1998 novels that have been translated.

    1. I based this on the data from Novelupdates. If the website doesn’t have those entries with the proper release dates then I would not see it.

  9. Wow! That’s some fun data analysis…(insert intelligent comment here)

  10. Wow. I’ve been wondering about this myself. Thank you for taking your time to do this research. Due to the amount of time it takes to translate a Chinese novel, I’ve started to learn Chinese myself. Though some say that it is one of the most difficult language to learn, I still think that there is merit to learn Chinese to help translate novels that need to be appreciated.

Tell me something

This site uses Akismet to reduce spam. Learn how your comment data is processed.