#naruto stats

On the first glance the graphs look peculiar: there is always sort of a double maximum for “0 Comments” and “2-5 Comments” divided by a minimum for “1 Comment”. But once we remember that the Comments on AO3 include also author’s responses the situation becomes clear: the majority of fics have 2 comments: namely 1 real Comment from a reader and one answer from the author.

To put this hypothesis into a test I checked how many fics have 2 Comments, and it is 316 out of 839 of SasuSaku fics in “2-5 Comments” bin, 84 out of 204 fics for NaruSaku and 104 out of 315 fic for KakaSaku – which gives always ca. one third of comments in the “2-5 Comments” bin.

At that point I deliberated for a while whether I should modify my bins to include “2 Comments” as a separate category, but I opted out of it for the sake of (future) comparisons with ff.net (and because I think by bins are very good, thank you).

As a side note: on the first 4 graphs we can also see that MadaSaku has a different distribution than all the other ships, but it is worth mentioning that it was also an outlier on ff.net. We will come to this later.

Peculiarity of distribution aside – the feedback on 3 main Sakura’s ships is consistently miserable. But, I reasoned, AO3 is famously M/M oriented so maybe the F/M audience is simply not there. Therefore, I set off to check all the Naruto ships in the order of their popularity, which btw is the following:

So, working down the list we get this:

And, WTF Naruto fandom? Most of the fics for hugely popular, established ships have no comments??!! Or one comment and an answer from the writer??!!

That was a really disappointing exercise…

Until I found that:

MadaTobi was the first ship that displayed a different pattern of readers interactions! It has a maximum in “21-50 Comments” per fic which is perfectly decent and corresponds well with the values for the ships I examined before on ff.net.

Encouraged by this discovery I dug further: (And tbh for a moment I thought I have here a “Madara-effect”, since both MadaTobi and MadaSaku displayed encouraging results 😉 )

But Madara-effect was not the case (HashiMada displays the same distribution as all the other ships) and further data were even more disappointing. (Let me repeat myself: WTF, Naruto fandom?!)

Until I found another instance of good readers’ participation:

KakaGai! So, at that point we had three ships that were behaving differently than the norm: MadaTobi, KakaGai and MadaSaku.

I looked further to see if I can find more examples of fandom positivity, but, spoiler alert – No. For the sake of appearances lets look at the data though. GaaLee is sort of a borderline case, so let’s flag it as “tentative” for now.

So, what could be the reason for such a drastically different behavior of distinctive pockets of the fandom? Over the course of last year, I was observing how MadaTobi was rapidly gaining popularity, so that hinted me to follow this line of reasoning.

For each ship, I broke down the fics numbers into years when they were published. I ignored all the fics published in 2020 (because the year has only started and including those data would skew the image). To be able to compare the ships (which vary widely in term of fic numbers) I calculated the percentages of fics published in each year. I.e if a ship had 100 fics published until December 2019 and 30 of them were published in 2019, then 2019 had 30% of all published fics. It was at that point irrelevant how many fics were published in 2020, they weren’t counted either the way. I included only data from 2011 on, because earlier ones are very fuzzy due to back-dating customs that were in fashion at that time (for explanations see here).

Let’s look at the data. Tadaaa!!!

Isn’t it absolutely beautiful?! (yes, I know!) Does it look like a tangle of knitting yarn? (yes, I know that as well!)

So, let me break it down for you:

See that, that general trend? This is the growth curve that most of the ships follow – pretty linear, with maybe a bit of acceleration in 2019.

Now let’s focus on outliers: (3 most popular ships: SasuNaru, KakaIru and SasuSaku are included for reference – they illustrate the general trend pretty well).

The ships that grew more than 30% in 2019 are: SasuHina, MadaTobi, GaaLee, KakaObi, HashiMada, KakaGai, MadaSaku and InoSaku. So, among those eight, 4 (3 obvious, and 1 tentative) are the ones which showed outstanding readers’ engagement!

 So, are newer, more energetic parts of fandom also the ones with best readers’ engagement? That would be a very optimistic message… But what about four ships that grew tremendously, but still show very poor readers behavior?

SasuHina is an easy case. If there exist a ship that has a captain, then it is this one. And the captain is @365daysofsasuhina​ who contributed staggering 438 fics (out of total of 966) to SasuHina! (Seriously, a round of standing ovation, because this is amount of dedication, work and talent that we can all only dream about. Talking about being a change you want to see!). And 90% of that contribution was done exactly in 2019, which explains the growth spurt of the ship. Sadly though, the audience response didn't keep up with the supply of new fics, therefore great many of them remain with very few comments.

So what about the remaining ones: KakaObi, HashiMadara and InoSaku?

Scrolling through the fic lists I noticed that great many of KakaObi fics are written in Chinese, so I decided to look into it deeper.

I analyzed the percentages of Chinese fics for all the examined ships (and percentages of English ones, just in case there was a ship with overrepresentation of some other language):

And we can see that both KakaObi and HashiMada have a very strong presence of the Chinese fics!

At that point I thought I have it, and proceeded to look at the distribution of Comments for Chinese KakaObi and HashiMada fics. And indeed, the reader’s engagement is sadly, almost nonexistent (165 out of 182 HashiMada fics and 478 out of 630 KakaObi fics have no comments at all). The reasons for that may be multiple; one that come to my mind is that the Chinese readers are still to follow their writers to AO3, or that the fics published recently are in reality not new (which for sure they are not, given the sheer amount of them) and received already their share of feedback on other platforms.

But, even if one subtracts the Chinese fics from the overall fics, the distribution remains the same! ☹ KakaObi and HashiMada still display the standard Comment distribution, despite having grown 45% and 41% in the last year!

It looks a bit better, but still follows standard, not very flattering trend of most of other ships.

So, then I thought that maybe if I subtract the Chinese fics the sudden growth in 2019 will disappear?

The answer is unfortunately, no. The growth spurt is still there, language independent.

So, at this point I don’t know... I am convinced that there is some effect coming from the influx of Chinese fics, but I cannot put my finger on it. (It is also a beautiful example of how correlation does not equal causation.) I also have no explanation for InoSaku behavior…

  Nevertheless, I think it is possible to draw some conclusions, and those conclusions are of vital importance for the health of fandom as it is currently on the move from ff.net to AO3. And I would hate to see the good habits that are in place among ff.net readers getting lost during this move!

It seems that the readers response is better in the pockets of fandom that experience rapid growth, which in itself is very positive news. But one would think that the entirety of Naruto fandom is experiencing such growth (as the fandom is moving to AO3, remember…?). Unfortunately, it isn’t the case.

There can be yet another factor in play: namely the question if the ship is well represented on ff.net, and most of the ships are. Let’s take a final look at our three outliers: MadaTobi, KakaGai and MadaSaku.

Quick look into ff.net (by quick look I mean: sum of two main Ship Names plus Pairing Option minus fics that were double-tagged) shows that there are 174 KakaGai fics, 132 MadaTobi fics and 230 MadaSaku fics (my carefully calculated MadaSaku fic number was 232 in December 2019, so now I believe this number will be slightly higher).

MadaTobi and KakaGai are definitely underrepresented on ff.net (though MadaSaku is not). Therefore, new fics are not being published on ff.net in parallel to AO3, but rather on AO3 exclusively. It is a speculation only, but maybe if readers are given a choice between commenting on ff.net and on AO3, they are choosing the older platform.  That would imply that the authors who make a complete move can still count on feedback, as with absence of ff.net-uploaded fics the readers would have only one place to comment on. At least that’s my hope…

Methods

All Naruto fics and Sakura’s fics

Data was collected from FF.NET (filters set on: Ratings: All, Sorted by: Published date).

For the description of process of total Naruto fics numbers collection see this post. Briefly, the archive was browsed and the first pages featuring a new year as Publish Date of the fic at the top of the page were taking note of. Knowing the number of archive pages produced per year and the number of fics per page (25 fics on FF.NET) the number of fics published in every given year was calculated. Data obtained through this method have error of +/- 24 fics per year. For detailed description of limitations of this methods see here.

To find out the numbers of Sakura’s fics, I chose Haruno Sakura as Character A in Filter Menu, and performed an analogical analysis as for total Naruto fics.

Ships – Pairing Option

Romantic pairings are more complex issue. From October 2013 a ship can be indicated by “Pairing Option”. Pairing Option makes the characters’ names appear in square brackets, and users can search for the fics tagged with a specific ship through the Filter Menu.

I searched fics tagged with Pairing Option between Haruno Sakura and following characters:

Uchiha Sasuke

Uzumaki Naruto 

Hatake Kakashi 

Uchiha Itachi 

Gaara 

Yamanaka Ino 

Uchiha Madara

The choice of characters was based on this analysis. I chose Top 5 ship partners of Haruno Sakura, accompanied by Yamanaka Ino (Sakura’s 7th most popular ship and her most prominent F/F ship) and Uchiha Madara (Sakura’s 9th most popular ship and my OTP 😊 ).

For every ship, I sorted the fics by Publish Date and counted the number of fics published per year.

Ships – Ship Names

Before the introduction of Pairing Option other solutions were invented and utilized by the fandom. In order to allow the readers to search and find shipping fics, authors were putting specific phrases into the summary, and readers would use Search Menu. Such phrases involved various variants of “Ship Names”: character names mergers (example: SasuSaku), usage of letter “x” between character names (example: Sasuke x Sakura, SasukexSakura) and combination of both (example: SasuxSaku). For details see this analysis.

In this analysis the following phrases were searched, and number of ships published per year was collected (on example of ship Uchiha Sasuke x Haruno Sakura):

SasuSaku

SakuSasu 

SasukexSakura 

SakuraxSasuke 

SasuxSaku

SakuxSasu 

Sasuke x Sakura 

Sakura x Sasuke

For detailed description of the process of searching various ship names see this post.

Fic number retrieved from every Ship Name search were summed up year-by-year, giving a total number of fics per a given ship per year. Of course, this procedure opens up a possibility for fics featuring several Ship Names in their summaries to be counted multiple times. For detailed elaboration on limitations of the method please see here.

Results

First off, it is interesting if the number of Sakura’s fics reflects the general changes of Naruto fandom. Fics tagged with Haruno Sakura and total Naruto fics are presented in Figure 1. (Please note that Sakura’s fics are plotted on the secondary vertical axis! I will use secondary axis a lot in this analysis. For those unfamiliar with it, a short how-to-read-it: Total Naruto Fics reached their maximum in 2008 when ca. 47000 fics per year were created. Sakura’s Fics also reached maximum that year, but in her case ca. 11000 fics were published (you need to look at right-hand axis). Such presentation allows us to see that Sakura’s fics creation follows very exactly the output of the entire fandom. If I would plot Sakura’s Fics on one axis with Total Fics, her graph would look very flat, and this observation would have been lost.)

In 2013, one year before the end of manga, all Naruto fics show stabilization of published fics numbers, while fics featuring Sakura show a small, local maximum. The end of the franchise and canonization didn’t halt the general negative trend that was present in the fandom since 2009.

Sakura’s fics represent ca. one fourth of total Naruto fics (102K fics vs 407K fics). If one breaks down the percentages year-by-year (Figure 2.), it can be seen that the 25% was pretty constant over the years, with the exception of years 2003-2005. When I was performing Ship Names searches, it turned out that during those years there were many more “shipping” fics with Sakura than the fics tagged with “Haruno Sakura” (example: in 2003 there were 34 Sakura’s fics, while in the same time 54 SasuSaku fics and 53 NaruSaku fics!). So, it is clear that during the very early years people weren’t tagging the characters too diligently. Therefore, data form the years 2003-2005 will need to be excluded from certain further parts of this analysis.

And now, finally, after all the boring intro: THE SHIPPING!!!!!

SasuSaku

Let’s start with Uchiha Sasuke, Sakura’s most prominent ship (Figure 3.). As reference, Sakura’s fics are plotted on the secondary axis (this figure and the next ones). Fics numbers retrieved from Ship Names searches and Pairing Option search are shown separately and shouldn’t be added up. It is not clear how many of the fics are both tagged with Pairing Option and a Ship Name, especially those published in the years 2013-2019. I am planning to address this issue for Sakura’s main ships in a separate analysis.

Generally, SasuSaku ship development follows pretty well the number of Sakura’s fics, with a maximum in 2008, and a steep decline afterwards. It seems that the decline of SasuSaku fics is steeper than the general decline of Sakura’s fics (and of the Naruto fandom in general) though. There is a small local maximum in 2013 that reflects an overall increase of Sakura’s fics (2013 maximum may be longer and larger if Pairing Option fics are to be included – but this needs prior deconvolution) and afterwards we have a steady decline.

NaruSaku

Uzumaki Naruto x Haruno Sakura ship shows similar trend as SasuSaku ship until 2012 (maximum in 2008 and then decline). The local maximum around 2013-2015 seems to be more pronounced though than the one of Sasuke’s ship and probably than in fandom in general (Figure 4.).

KakaSaku

The ship with Kakashi behaves differently from the two previous ones (Figure 5.) and often goes against the general trends of the Naruto fandom. Namely, it shows maximum in 2007, a local maximum in 2011 (and no maximum in 2013), and, going against the decline of entire fandom (!!!) another local maximum in 2017. Pairing Option for KakaSaku was also rather on a constant level in years 2016-2018 when for other ships it was already showing a decline. One need to take under consideration though, that KakaSaku is by far not as numerous as the two previous ships, and relatively small number of fics can produce effects that are well visible on the graph (example: maximum of year 2017 is only 173 fics).

ItaSaku

Ship with Itachi shows maximum in 2009, and a second, relatively strong maximum in 2013 (the strongest among the ships examined so far). The fics tagged with Pairing Option exhibit maximums in 2014 and in 2017, but otherwise the general shape of the curve reflects the trends in the fandom (Figure 6.).

Gaara

Ship Gaara x Haruno Sakura had its maximum in 2008. The local maximum from 2013 is the weakest out of the ships examined so far, as if the temporary rise of activity in the fandom cause little increase of Gaara x Sakura fics production (Figure 7.).

InoSaku

Ship Ino x Sakura follows general trends of the fandom, with an exception of a local maximum appearing in 2017 (Figure 8.). The Pairing Option shows stabilization in years 2015-2018, despite the general fandom decline. However, relatively small size of this ship (ca. 20-40 fics per year in the recent years) makes all relatively small variations appear as more significant than they might really be.

MadaSaku

Ship Uchiha Madara x Haruno Sakura is a bit of a special case, as Uchiha Madara appeared in the franchise on-panel in 2008 only, and in person in 2011. This is reflected through the first MadaSaku fics showing up in 2008 only. There is a maximum in 2010 – interestingly, before the Tobi/Obito reveal, when Tobi persona was still assumed to be Madara. It would be interesting to see if the fics published this year take place in original Narutoverse or are of a time-travel sort. Years 2012- 2013 show a decline followed by a steep increase (against the general trends of the fandom) with maximum in 2016 (for Ship Names) and 2017 (for Pairing Option). In 2019 another increase can be noted, after a relatively weak 2018 (Figure 9.).

So, now it’s time for some comparisons between the ships.

Ships with Sasuke, Naruto and Kakashi can be plotted together (Figure 10.). The ranking of ships remains unchanged over the entire timeline, however NaruSaku used to be ca. 4 times less popular than SasuSaku during the fandom heyday, and around the times of canonization it became only 3 times less popular (a relative increase in popularity). This is unfortunately also the time of introduction of Pairing Option, so this effect may be an artifact of parts of fandom using tagging systems in different ways. I will try to address this very question in the future.

Dynamics of ships tagged with Pairing Option (Figure 11.) show as well the 3 : 1 ratio of SasuSaku : NaruSaku ships in years2014-2015. From that relative high, the decline of Naruto’s ship seem to be more dramatic than this of Sasuke’s ship in the recent two years (in 2019 the ratio was 4,3 : 1).

But the numbers of fics created per year don’t tell the entire story, especially with the number of total fics changing so rapidly. Therefore, I examined what percentage of Sakura’s fics a ship amounted for over the years (Figure 12.) Years 2003-2005 are unsuitable for this kind of analysis, as the numbers of shipping fics exceed (and in 2003 vastly so) the number of Sakura’s fics.

The results from Ship Names searches are presented in Figure 12. For presentation purposes, SasuSaku ship is plotted on the primary vertical axis, while all the other ships – on the secondary.

SasuSaku ship used to make up to 30% of all Sakura’s fics during the fandom rise and heyday (2006-2008), only to drop to 15-17% in more recent years. The decline reached its local minimum in 2011 (for orientation: Sasuke’s and Sakura’s infamous reunion under the bridge in Iron Country was published in March 2010). During the Fourth Ninja War chapters SasuSaku stabilized on 16-17%. The canonization of ship in November 2014 didn’t seem to influence its popularity, and recent years brought a decline with all-time minimum of 14,3% in 2018.

However, it is possible that data from years 2013-2019 are influenced by existence of Pairing Option In the most optimistic scenario, the two tagging systems would be additive and we would have ca. twice as much SasuSaku fics. This question will be addressed in the future.

Data from Pairing option show only a fraction of fandom history (Figure 13.), and after a steep rise (caused by spreading of Pairing Option awareness among the writers) SasuSaku remains on a steady level of 14-18% of all Sakura’s ships.

NaruSaku ship was at 8% in 2007, and later it dropped to 4,3% in a manner similar to SasuSaku ship. During the Fourth Ninja War Arc popularity of NaruSaku rose once again to almost 7% and remained relatively high also post-manga end (up to 2015), only to drop to its minimum of 3,6% in 2019. The Pairing Option shows a similar drop in NaruSaku fics in last two years.

KakaSaku started off at 4% in 2006 and afterwards suffered a small decline, interestingly - with a minimum during Fourth Ninja War Arc in 2013 and not like those of Team 7 boys in 2011. From 2014 KakaSaku is on a relative rise as reflected by both Ship Names searches and Pairing Option.

ItaSaku had its maximum relative popularity in 2009, when 3,5% of Sakura’s fics were ItaSaku ones, and another peak in 2013. Last year brough a small, but visible rise of Itasaku fics.

GaaSaku started at 3% in 2006 and from then on it is on a steady decline according to Ship Names searches. Pairing Option shows a modest increase of published fics in two recent years.

InoSaku doesn’t show dramatic changes, though a low period can be noticed during the years 2010-2015.

MadaSaku remains on very low level throughout the entire timeline, but a rising trend can be noted in recent years (both tagging systems).

So, let’s start from this graph:

As I already explained here and here, on FF.Net there exist two types of independent systems for tagging romantic relations ships: Ship Names and Pairing Option (for explanation see here). The graph in Figure 1. shows the dynamics of SasuSaku ship where Ship Names are marked dark blue, while Pairing Option is light blue. However, without manual check there is no way to know how many fics tagged with one system are also tagged with another. I.e. we cannot know if we can, and if yes then to what extent, add these values together.

This issue was bugging me for a very long time, especially as the introduction of Pairing Option (October 2013) lies unfortunately close to the end of manga and canonization of ships (November 2014). And it’s super interesting to know how did the fandom (the Sasusaku part of it, as well as those parts of fandom whose ships weren’t confirmed) reacted for the end of manga and canonization.

So, I set off to untangle the double-tagging issue, and I’ll start with SasuSaku. NaraSaku and KakaSaku are next in line.

Additionally, since it seems that the center of gravity of Naruto fandom shifts towards AO3 (as can be seen in this analysis), I included SasaSaku dynamics on that platform as well.

Methods

FF.Net

I based my analysis on already collected data (published here) on ship dynamics by year. Those data gather number of fics published in a given year for the following tagging systems:

Pairing Option

Ship Names (Ship Names examined: SasuSaku, SakuSasu, SasukexSakura, SakuraxSasuke, SasuxSaku, SakuxSasu, Sasuke x Sakura, Sakura x Sasuke, Sasu x Saku, Saku x Sasu)

Basing on those data, I set off to excluded the double-tagged fics.

For practical reasons (minimalization of amount of clicking around I had to do), I approached the issue from two different sides.

A.     To exclude double-tagged fics tagged with Pairing Option and with Ship Name SasuSaku (main Ship Name used by the fandom) I used the Filters Tool (settings: Character A: Haruno Sakura; Character B: Uchiha Sasuke; Pairing Option: On, Sort: Publish Date). I went through resulting list of fics page-by-page, taking note of the year of publication and searching for the phrase “sasusaku” (using Ctrl+F key combination in the browser). I took note of number of times the phrase “sasusaku” appeared on each page and summed up those numbers for the fics that were published in the same year.

B.     To exclude double-tagged fics tagged with Pairing Option and with minor Ship Names, I searched for the minor Ship Name of interest (for example: “Sasuke x Sakura”) using Search Window, and, after sorting the results according to Publish Date, I searched the for the occurrences of square brackets (“[“) on the result page. Square brackets are used by FF.Net as an indicator of characters paired together through Pairing Option. However, it also was and still is used by the authors in summaries to describe romantic pairs, writing status of the fic, tropes, AUs and various other, mostly house-keeping purposes. Therefore, it was necessary to manually check if occurrence of a square bracket meant Pairing Option (and if the romantic pair indeed was SasuSaku!) or if it was merely featured in the summary. For all the minor Ship Names, I manually checked each occurrence of square bracket and took note of those that indicated romantic pair Uchiha Sasuke with Haruno Sakura, along with the publication year of the fic that was double-tagged.

Using this method, I searched through years 2006-2019. I skipped years 2003-2005 due to previously made observations about general messiness of the data from the very early years of the fandom (people weren’t tagging particularly diligently…)

Using the data collected as described above, I could subtract the double-tagged fics from the sum of fics tagged with Pairing Option and fics tagged with Ship Name.

To get better data resolution for the decrease (years 2011-2012) and subsequent rise of SasuSaku popularity (years 2013-2014), and in order to be able to corelate fic production rate with canonical events, I performed the same analysis month-by-month for the years 2010-2015.

AO3

I searched the AO3 archive by clicking into Uchiha Sasuke/Haruno Sakura tag and sorting by Publishing Date, and took note of how many fics were published per year. I did the same for the character Haruno Sakura (as I didn’t have these data yet). On AO3, authors can freely set a publication date, and the chosen date is then displayed in the header of the fic when one clicks inside it. However, the user-defined date doesn’t influence the sorting order, i.e. fics are still sorted according to their date of addition to the archive. When such an “unclear case” happened on a border of two years, the fics of uncertain publishing date were assigned to the new year (i.e. if a fic dated as from June 15th 2006 appeared between fics dated December 31st 2017 and January 1st 2018, it was assigned to the year 2018).

Limitations of the method

The elimination of double-tagged fics concerns only fics tagged with Pairing Option and one of the Ship Names. It does not eliminate the fics tagged with more than one Ship Name. I.e. a fic that has in its summary two of the Ship Names it will be still counted twice. Also, if it has one of the Ship Names in the summary and another in the title, it will also be counted twice. In an extreme case, if author used all 10 Ship Names, then the fic would be counted 10 times.

In the same time, due to mechanics of elimination method A. if a fic (that was additionally tagged with Pairing Option) contained multiple occurrence of phrase “sasusaku” in the summary or title, such fic will be subtracted multiple times (as many times as there have been repetitions of the phrase). Also, if an author had a phrase “sasusaku” in their name, their fic will also be falsely subtracted even if it wasn’t double-tagged at all (falsely, because my initial harvest of “sasusaku” fics was defined as “search story” and not “search writer”).

Results

Figure 2. presents the number of SasuSaku fics tagged with different systems (shades of blue) and the unique number of fics published per year (magenta).

Figure 3. compares the numbers of fics published on FF.Net and AO3. Please take note that simple addition of fics from the two platform can actually be misleading and incorrect. This is due to the fact that many authors publish in parallel on both platforms, and fics uploaded to FF.Net are de facto the same fics as those published on AO3. Additionally, some authors move from the older FF.Net to the newer AO3, where they re-post their older works in bulk.

Unique fic numbers on FF.Net show a main maximum in 2008 (in line with the maximum of total Naruto fics, see here) when 3308 SasuSaku fics where published, followed by steep decline with a minimum in 2012 when only 1409 were produced. There is a significant increase in fic numbers forming a secondary maximum in years 2013-2015 when ca. 2000 fics were published yearly. After this increase in activity happening around the end of manga and canonization of SasuSaku ship, a steep decline is continuing until the present day. In 2019, SasuSaku ship noted its historical minimum when only 540 fics (on average 45 per month) were published on FF.Net.

The steady increase of SasuSaku fics on AO3 is in line with the general increase of activity of Naruto fandom on this platform. It is noteworthy that in 2019, for the first time, there were more SasuSaku fics published on AO3 (779 per year, i.e. 65 per month) than on FF.Net.

Due to highly fluctuating numbers of Naruto fics published per year it is difficult to assess ship popularity basing on published fics only. The percentage of shipping fics in relation of total Naruto fics, or total fics featuring one of the romantic partners (I used Sakura) are better indicators of ship popularity. For this part of analysis, see Discussion.

To analyze the effect of canon events of the fic production, better resolution is necessary. Figure 4. summarizes SasuSaku ship dynamics during the years 2010-2015 on monthly bases.

It can be seen that with introduction of Pairing Option, the main Ship Name tag (SasuSaku) slightly lost on prominence while the Pairing Option gained acceptance, and already in January 2014 there were more less same number of fics using Pairing Option as Ship Names. However, most fics used either one type of tagging or another (double-tagging ratio was 25%-33% in years 2013-2015). Interpretation of the curve and correlations between the canon and fandom events and monthly publishing rate will be elaborate on in Discussion.

Discussion

Figure 5. shows what percentage of all Sakura’s fics consisted of SasuSaku fics in years 2006-2019 for FF.Net and 2011-2019 for AO3. This approach allows for correction for the effect of general fandom activity decline (see this analysis) that took place after year 2008.

SasuSaku of FF.Net shows two maxima, divided with a strong minimum in years 2011-2012. On AO3, after a slow start (12% in 2011) SasuSaku quickly reached the level of 20% which it held since 2013. Currently, both platforms host a very similar percentage of SasuSaku fics.

I would like to focus primarily of FF.Net, as it covers entire timeline of publishing of the manga.

The way how canon influences fandom was one of the focal points in this analysis. Sakura’s ship with Sasuke was very prominent during the earlier years of the fandom (2006-2008) when it made more than 30% of all published Sakura’s fics. And if one takes into account that not all the fics are necessary shipping ones, the dominance of SasuSaku ship becomes even more massive.

However, during later years SasuSaku lost at popularity, reaching its minimum in 2011 and 2012 when only 17,1%and 17,5% of all Sakura’s fics featured romantic pairing with Sasuke. In 2013 we see already an increase, that continues until 2015, when SasuSaku reaches a second maximum (almost as high as the first one) achieving once again 30%. After 2015 there is a decline, as this effect is independent from the general fandom decline. In last years, SasuSaku makes ca. 23% of Sakura’s fics. That might be an indirect evidence of popularity increase of others, non-canonical and more exotic ships.

To put the timeline into perspective:

Sasuke kills Itachi in chapter 394 published in March 2008; and, informed about Itachi’s true motives, declares intention to kill Konoha Elders in chapter 404 published in May 2008.

Sasuke attacks Kage Summit in chapter 460 published in August 2009.

Sasuke’s and Sakura’s encounter and mutual assassination attempt on the Bridge in Iron Country happens in chapters 482-485 published in February/March 2010. Afterwards Sasuke virtually disappears from the manga for many chapters, appearing only briefly in chapter 553 (published in September 2011) with fully evolved Eternal Mangekyō Sharingan in his eyes.

His next prominent appearance is already after the beginning of Fourth Ninja War and implementation of Edo Tensei, when Sasuke encounters Itachi in chapter 576 (published in March 2012) and after Itachi’s revelations, declares the will to destroy Konoha in chapter 589 (published in June 2012).

After the meeting with the previous Hokage, in chapter 627 (published in April 2013) Sasuke changes his mind, declares will to protect Konoha and joins Shinobi Alliance four chapters later (May 2013).

Figure 6. presents percentage of unique SasuSaku fics (in respect to total Sakura’s fics) in context of manga events. During Sasuke’s quest for revenge the popularity of the ship was constantly dropping. Interestingly, it reached its low during the time period when Sasuke was basically absent from the manga. His declaration about destroying Konoha in 2012 happened also during this minimum. However, despite this declaration, already in 2013 the ship gained popularity, possibly in response to anticipated changing of sides that indeed took place in May 2013. After that, SasuSaku was on steady rise that continued throughout ship canonization.

In order to see the effects of specific manga events it is necessary to look at more detailed picture. Figure 7. presents number of unique fics published (not percentages, because gathering monthly data for total Sakura fics exceeded my capacities) and puts them in context of manga and fandom events.

Data of this high resolution tend to be choppy, nevertheless some trends can be detected. Most prominently, SasuSaku Months (fandom events held every July in years 2011-2019) give rise to sharp peaks. There is a similar peak in July 2010 but I didn’t find any evidence of SasuSaku Month that year. @sasuxsakumonth and @sasusakumonth – maybe you know more about that?

After Iron Country Bridge disaster, the number of fics indeed falls. The effect is blurred by the peak in July (if it was indeed an event), but afterwards there is a decline. During the second half of 2010 and the first half of 2011 the number of fics oscillated between 100 and 150 per month, and even SS Month didn’t produce a big increase (182 fics only). During the second half of 2011 and first half of 2012 the number was 100 fics per month.

Sasuke’s re-appearance in March 2012 seemed to revitalize the fandom and in autumn and winter of 2012 the fandom was already on the rise. In 2013 and 2014 there were already 150 fics regularly published every month, and SS Month in 2013 and 2014 reached almost 300 fics (of course not all the fics were produced specifically for the event).

A sharp decline in September - October 2014 may be an after-effect of a successful SS Month in July (and certain writers’ fatigue) or can be caused by canon events, as directly before the canonization, in chapters published in September/October 2019, Sasuke announces his “revolution”, places Sakura under death-by-Chidori-genjutsu and fights Naruto.

And finally - in answer to my initial question – canonization of SasuSaku ship in November 2014 resulted in a very prominent (comparable only with strongest SS Months), sharply rising maximum, that, additionally, tailed at least until January 2015.

As general observation, it seems that fandom reacts to “negative events” (canon moments decreasing the chance of characters canonically getting together, absence of character from the story) with a certain delay. It is understandable, as one can imagine that negative feedback from the side of canon doesn’t make people to abstain from publishing the fics that they’ve been already working on, but rather acts demotivating which results in a diminished output of fics in subsequent months. In contrary, positive events (like ship-supporting canon moments or canonization) bring very fast reactions, as writers tend to get inspired by new, ship-supportive material.

1.      I identified Top Ten most popular Characters through scrolling through all the characters (Pairing Option was Off at this stage) and taking note of their fic numbers. There are 49 characters that have more than 1000 fics to their name, and those characters will be considered further. Top Ten were (in order of popularity): Sasuke, Naruto, Sakura, Hinata, Itachi, Kakashi, Gaara, Neji, Ino, Shikamaru

2.      For each of the Top Ten characters I switched the Pairing Option On and scrolled through all the listed Character Bs, taking note of the ships with more than 100 fics. This way, I systematically checked all possible ships where at least one partner belonged to Top Ten.

At that point it should become apparent for everyone that here lies the greatest limitation of this method. Namely, there can be a ship, consisting of both partners outside the Top Ten that nevertheless is popular enough to rank high.

Therefore:

3.      Since I wasn’t ready to search systematically through further characters, I concentrated on ships featuring characters ranking 11-49 that struck me as possibly popular (canon pairs, Akatsuki teams, fandoms favs known to me) and checked manually the fic numbers for those fics. This check was by all means not exhaustive and was obviously biased.

Limitations of the method

The biggest limitation of this method is described in point 2., and the partial solution provided in point 3.

Additionally, this method is based exclusively on Pairing Option. Pairing Option was introduced in October 2013 and fics tagged with it tend to be new. The parallel ship tagging system (so called Ship Names, i.e. SasuSaku, NaruHina, etc.) was in place for the majority of the Naruto fandom history on FF.Net and is still widely in use. For more details see here. This analysis doesn’t include in any way the fics tagged with Ship Names (and the differences in rankings between the systems can be large; for example featuring Sakura’s ships see here).

Results

The Table below summarizes 42 ships that I identified applying approach described in point 2 and supplementing it with approach from point 3. This list is not exhaustive and I am pretty sure there are couple of ships towards the bottom of the ranking that slipped through my fingers.

Ships identified via point 3. are marked red. Those ships are: MinaKushi, SasoDei, BoruSara and MadaTobi. It is worth mentioning that MinaKushi is ranked 10th, and it had I not used the manual check, it’s place would have been taken by ShikaIno, which proves that also relatively high-ranking ships may escape the procedure from point 2.

Since FF.Net lists “Naruko” as separate character, ship Sasuke x Naruko appeared as a separate entity in my method. It has 178 fics and is ranked 28. If one adds those 178 fics to the classical SasuNaru ship, it introduces profound changes in the top of the ranking. Namely, SasuNaru with 4105 fics takes the 2nd place from NaruHina (with 3930 fic), as those two ships have practically identical fic number. Btw, I checked Hinata x Naruko, just to be fair, and since it has 29 fics, it wouldn’t make NaruHina win its 2nd place.

Discussion

The ranking is dominated by ships consisting of four players: Sasuke, Naruto, Sakura and Hinata. Almost all combinations of ships featuring those 4 characters (except of SakuHina) rank at the top of the list. The ships featuring those 4 characters are also the only ones that cross the boundary of 1000 fics per ship.

SasuSaku dominates the ranking with 5826 fics, followed by an almost-tie between NaruHina and SasuNaru with 3930 and 3927 fics, respectively (or the 2nd and 3rd ranks are reversed, see above).

Interestingly, there is a second almost-tie on the 4th and 5th place with NaruSaku (1733 fics) and SasuHina (1728 fics).

There are 23 F/M ships in the ranking, 10 M/M and 1 F/F. 8 ships (with OCs, Kurama and Naruko) are hard to categorize.

In Top Ten there are 4 canonical ships, while the fifth prominent one (InoSai) ranks 13. However, for every canonical ship, the uncanonical ‘options’ for either, or both partners, feature equally prominently and usually on similar place in the ranking (the ‘Top 4 Characters’ situation described above; ShikaTema (7th) vs ShikaIno (11th), InoSai (13th) vs ShikaIno (11th), and SasuIno (12th))

Several ships ranking very highly in AO3 rank low on FF.Net. KakaIru, which is the 2nd most popular ship on AO3 ranks 15 on FF.Net (though it can be an artifact of using exclusively the Pairing Option, I have an impression that this ship used to be far more popular in the past). KakaObi (5th on AO3) is ranked 33. MadaTobi which is ranked 7th on AO3 barely makes the cut.

MadaSaku is 36th, but since it’s towards the end of the ranking, I would treat it more as an orientation than as a set-in-stone number, as there might be a fic or two that I missed that would rank higher.

It is interesting to note that 5 high ranking ships are the ships with OC. The canonical ship-partners in these pairings are: Kakashi, Itachi, Gaara, Sasuke and Naruto.

And finally, the curiosities: there are two ships of Kurama that made it in: with Naruto (ok, I can understand that) and with Itachi (WTF? If anyone read Kurama x Itachi fic, please, let me know what it was about!)

Methods

I collected the data browsing through both archives. AO3 by default shows all available fics. Filters in FF.NET were set to display all Ratings (not Rating K -> T as in default). Fics were sorted by Publish Date in both archives.

I browsed through the archives and took note which consecutive page of the archive featured a first appearance of a “new” year as a Publish Date of the fic on the top of the page (see Figure 1).

In the example illustrated in Figure 1.  the first fic on the page 806 was published on 01.01.2018. The first fic on the page 807 was published on 31.12.2017. Page 807 was defined as the first page of 2017. Independently on how many fics published in 2017 could be found on the page 806, page 806 was still counted as “belonging” to the year 2018. Such simplified approach was necessary for streamlining of data collection.

Data collection was performed similarly on AO3. However, AO3 displays Publish Date only in the body on the fic (when browsing the archive, only Update Dates are visible, and those are different than Publish Dates for most of multichapter stories). Therefore, manual inspection of fics on the top of the pages was often necessary.

After identifying all “first” pages for every year, I calculated (via simple subtraction) how many pages of fics were produced per year. FF.NET displays 25 fics per page, AO3 – 20 fics, therefore the number of pages per year was multiplied by 25 for FF.NET and by 20 for AO3 to obtain the number of fics published in a given year.

Analogical approach was employed to examine the number of fics produced per month (the page number with first appearance of “new” month on top was noted) for the years 2018 and 2019, to gain more insight into details of fandom dynamics in most recent times.

Limitations of the method

Due to my definition of the “first page of the year”, each year can in reality have up to 24 fics more (or up to 19 more on AO3) on the page preceding “first page of the year” (page 806 in the example from Figure 1.). According to my method those fics would be included into the count of the following year (2018 in the example from Figure 1.). In the same time, a year can in reality have 24 fics less (19 less on AO3) than I calculated on its “last page” due to similar phenomenon concerning the year’s “end”.

Therefore, all the values reported for FF.NET have an error of +/- 24, and those for AO3, error of +/-19.

Additionally, as new fics keep being added, years’ borders may shift from one page to another. This factor should however be negligible, because current fic addition speed is ca. 1 fic per hour on FF.NET and 1,5 fic per hour on AO3 and my data collection lasted ca. 2h for FF.NET and 1h for AO3.

On AO3, authors can freely set a publication date, and the chosen date is then displayed in the header of the fic when one clicks inside it. The user-defined date doesn’t influence the sorting order, i.e. fics are still sorted according to their date of addition to the archive. However, the user-defined publish date makes application of my definition of the “first page of a year” very challenging. I often had to look at second, third or fourth fic on the page to find out to which year they belonged to (in order to correctly assign the page to the year). Fortunately, the publish date-changing seemed to be in fashion only during the early day of the archive (years 2009-2011). Therefore, the data from the years 2010 and 2011 may be contain larger errors than those from later years. Due to this difficulty, I also entirely opted out from reporting the data for 2009.

 Results

Number of fics published per year on FF.NET and AO3 is shown in Figure 2.

On FF.NET, after a steep rise in years 2006-2007, Naruto fandom was at its peak in 2008 (having produced more than 45000 fics that year). After 2008 a rather steep decline can be observed, with a very slight reversal of negative trend in 2013. In 2019 only less than 9000 fics has been added to FF.NET.

Situation looks very differently on AO3, which, having been established in 2008 and opened for use in 2009, has been gaining momentum ever since. In 2019 ca. 13000 Naruto fics have been added to AO3. It is interesting to note that the growth of Naruto fandom on AO3 was very slow for several years, but recently it accelerated.

To examine the details of this “growth spurt”, I analyzed data from 2018 and 2019 month-by-month. Results are visualized in Figure 3.

It can be seen that until July 2018, less fics per month were being published on AO3 than FF.NET. In the period August 2018 – November 2018 the fics’ numbers between the platform evened out, and AO3 gained advantage over FF.NET in December 2018.

 In order to check if the gain of fics on AO3 can counterbalance the loss of fics in FF.NET, I plotted the summed numbers of fics (those published on FF.NET and those published on AO3). The results are presented in Figure 4.

At the first glance, it looks as if AO3 could compensate for the losses on FF.NET and halt the decline. The case is not that straightforward however, and simple addition of fics from two platform can actually be misleading and incorrect. This is due to the fact that many authors publish in parallel on both platforms, and fics uploaded to FF.NET are de facto the same fics as those published on AO3. Additionally, some authors move from the older FF.NET to the newer AO3, where they re-post their older works in bulk.

To answer the question if the decline of Naruto fandom has been stopped, one would need to know the numbers of unique fics published per year. But, looking at years-long trend, I would lean towards the opinion that the decline is real.

But not to finish this post on a too depressing note, I would like to share one more graph – namely the cumulative number of fics on both platforms (Figure 5.). Even if the fandom is past its heyday, hundreds of thousands of fics have been produced and are there to read and enjoy (and comment on!). And, last but not least, it’s up to us to keep the fandom alive!

Methods

After choosing a character, a menu appears on the righthand side of the page that allows the user to choose ratings, categories, etc. The same menu allows to see which relationships appear most frequently in the context of the chosen character (example shown in Figure 1.)

Attention: This menu doesn’t provide with neither Top Ten most popular ship for a given character nor with actual number of any of the ships listed, because:

a) Many ships listed in the menu involve platonic pairings (“&” relationships according to AO3 nomenclature) and while one can argue whether those are ships or not, the aim of this analysis was to find the prevalence of romantic ships.

b) Many ships listed involve characters different than the character of interest. For example: Senju Hashirama/Uzumaki Mito is the third most popular pairing written in the fics that contain Uchiha Madara as a character.

c)  Not every author tags character in question when tagging him/her as part of a ship. For example: Senju Tobirama/Uchiha Madara features in the above menu with 1173 fics (at the moment of writing this text). When the ship “Senju Tobirama/Uchiha Madara” is chosen through clicking into the tag, AO3 provides us with 1250 fic. I.e. in case of 77 MadaTobi fics, Uchiha Madara was not tagged as a character. Therefore, to obtain the real numbers of fics for any given pairing, one has to always click directly into ship tag.

The ships listed in the righthand menu can nevertheless provide orientation and hints about which ships should be inspected manually in order to get true number of fics for a given ship.

In case of Uchiha Madara the menu features only three romantic ships involving Madara, namely:

Senju Tobirama/Uchiha Madara

Senju Hashirama/Uchiha Madara

Haruno Sakura/Uchiha Madara

How to find out more ships featuring Madara?

I tried two approaches:

Method 1

I excluded one relationship from the list of ten most popular and iterated the search excluding the first most popular relationship only, the second most popular relationship only, and so on, together 10 searches.

During every search I noted the “new” romantic relationships involving Madara that popped up in the list.

In the second step, I excluded 2 most popular relationships simultaneously and again, searched all possible combinations, together 45 searches because it is combination without repetitions:

Again, during every search I noted the “new” romantic relationships involving Madara that popped up in the list.

When excluding 3 relationships I would need to perform 120 searches to cover all combinations. Therefore, I limited myself with searches: excluding 1st, 2nd and 3rd ship; excluding 1st, 2nd and 4th ship; excluding 1st, 2nd and 5th ship; etc.

All things taken together, this method proved to be very laborious, and didn’t bring expected results, i.e. “new” romantic relationships were showing up very rarely in relation to the workload.

Therefore, a different method was employed:

Method 2

I excluded increasing number of ships from the list of ten most popular relationships. I.e. I excluded 1st most popular ship and checked for appearance of “new” ships. Then I excluded first two most popular relationships (1st and 2nd) and checked for “new” ships, then excluded 1st, 2nd and 3rd most popular ships etc…

This method proved to be both efficient (i.e. new ships appear relatively quickly in respect to the workload) and much less error-prone.

This was therefore the method of choice for this analysis.

Figure 2. illustrates an example where the most popular ship (Senju Tobirama/Uchiha Madara) was excluded (Figure 2.a.) and a new ship (Uchiha Izuna/Uchiha Madara) appeared during the search. (Figure 2.b.)

Table 1 summarizes the romantic ships of Uchiha Madara found using Method 2. Data was collected on 22.10.2019. 40 most popular relationships written in context of Uchiha Madara were excluded sequentially, and new ships were noted in order of their appearance. The searches that didn’t brought any new romantic ship involving Uchiha Madara were omitted from Table 1. (for example: excluding 1st and 2nd most popular relationships didn’t bring any new romantic ship for Madara, therefore this search was not listed in the Table 1.).

Disclaimer: the order of appearance in Table 1. doesn’t necessarily reflect the ship’s popularity. Because of the fact that the list is composed using ‘exclude’ filter, the order of appearance is heavily influenced by which pairings are written together with which other pairing (as AO3 allows tagging multiple ships in one fic).

The list of ships in Table 1. was used to obtain final, numerical results for popularity of the ships of Uchiha Madara (see: Results).

Additionally, I manually checked popularity of several other Madara’s ships, and two of them (with Pein and with Kakuzu) ended up scoring high enough to be included in final results.

Same approach was used to find most popular ships of Haruno Sakura.

Sakura’s righthand menu lists 5 romantic ship involving Sakura herself, namely:

Haruno Sakura/Uchiha Sasuke

Haruno Sakura/Hatake Kakashi 

Haruno Sakura/Uzumaki Naruto 

Haruno Sakura/Yamanaka Ino 

Haruno Sakura/Uchiha Itachi

To find further ships for Haruno Sakura I employed Method 2 as described above. Table 2 summarizes the results.

Additionally, other ships of Sakura where manually examined, and ships with Senju Tobirama and Uchiha Obito scored high enough to be included in final results.

Limitations of the used method

The employed method of searching for popular ship has its limitations. Due to the mechanism of finding the ships to check for popularity (usage of ‘exclude filter’) it cannot be excluded that certain ships have been overlooked. Four such cases were actually detected through manual checks: Method 2 failed to show Madara/Pein and Madara/Kakuzu as ships qualified for checking for actual fic number, but when those ships were checked, their respective fic number was high enough to earn them 14th and 17th places among Madara’s most popular ships (see: Results). Haruno Sakura/Senju Tobirama and Haruno Sakura/Uchiha Obito scored high enough to be Sakura’s 20th and 21st most popular ship even though neither was detected via Method 2. This is entirely possible that certain other ships remained ‘undetected’ and therefore not included in the final results, but the risk of non-detection concerns mostly the less-prominent ships.

Due to AO3 tagging system, if a ship is tagged as “background” or “past” it will also be detected, and not differentiated from proper shipping fics. Possibility of tagging multiple ships in one fic causes collections of multishipping fics to be counted multiple times (which skews the numbers when percentages are concerned - see Figures 5 and 6). From this perspective, the relatively meager tagging system of fanfiction.net brings cleaner results.

Results 

Numerical values of fic numbers for every given ship were collected on 26.12.2019.

Romantic ships of Uchiha Madara are shown in Figure 3.

Romantic ships of Haruno Sakura are shown in Figure 4.

Discussion

Sakura is the third most popular Madara’s ship partner, and his most popular F/M ship, however it lags far behind the more popular “Founders Cluster” ships. Sakura is by far most popular F/M ship of Madara, next being OFC with 54 fic and Mito with 34 fics.

Madara’s fics are very focused on ships with few partners (Tobirama, Hashirama and Sakura), to the extend that various “exotic” ships rank high enough to appear in this breakdown (OC with 101 fics, OFC with 54, Reader with 41 and Harry Potter (LOL) with 10 fics).

Madara is Sakura’s eighth most popular ship partner (or seventh if you don’t count the polyship Sakura/Sasuke/Naruto as both Sasuke and Naruto feature already on the list as separate ship partners). One has to take into account that Madara appeared on-panel in February 2008 (i.e. almost 9 years after Naruto started publishing) and in person in the story in October 2011 (12 years after start of publishing). Nevertheless, MadaSaku accumulated more fics than many characters with much longer on-panel time, and that even though Madara is considered villain of the story.

When one takes under consideration the total number of Uchiha Madara fics (3913 at the point of writing this text) and Haruno Sakura (13397) it turns that MadaSaku fics form 5,3% of all Madara’s fics and 1,6% of all Sakura’s fics (Figures 5. and 6.).

It is surprising that Sakura, being famously (and infamously) shippable character, has as much as almost 40% of fics where she is not part of a romantic pair. Such high percentage of “other fics” may also mean that she is part of great number of very rare ships that were not investigated in this analysis (the analysis was stopped after excluding 40 most popular ships).

Outlook

Employing a direct data-scraping script would eliminate risks of non-detecting of certain ships. Such script would potentially also allow to examine ship’s popularity development over time.

Since this analysis was done before MadaSaku Weekend, it will be interesting to re-examine the data some time after the event to see if our participation influenced the popularity of the ship.

Methods

I chose 4 ships: SasuSaku, NaruSaku, KakaSaku and MadaSaku.

For each ship, after choosing the relevant characters in Filters Menu and ticking the “Pairing Option”, I sorted the fics a) by Favourites and b) by Reviews.

I defined following bins:

>500

201-500

101-200

51-100

21-50

11-20

6-10

2-5

1

0For a) and b) approach separately, and assigned the fics to bins (i.e. I counted how many fics there were that had more than 500 reviews, how many of those that have between 201 and 500 reviews etc..)

Please note that the bins are not normally distributed, but it is a conscious choice because neither are the data. I couldn’t really check *how* are the data distributed (I would need to scrape them for it) but my guess is that it is logarithmic. Bins themselves aren’t logarithmic, because I wanted nice round numbers, but somewhat close to it.

I, on purpose, included bins “1” and “0” because it was important for me to know how many fics did receive extremely little feedback, or none at all.

Limitations of the method

This analysis is based only on the fics tagged with Pairing Option, which was introduced in October 2013 and doesn’t reflect all, and even not the majority of fics for any pairing. Fics tagged with Pairing Option tend to be “newer”, but even for the fics published in 2013-2019, they represents only ca. half of all the fics (the other half is tagged with Ship Names): 53% for SasuSaku (see analysis for SasuSaku here), 55% for NaruSaku and 49% for KakaSaku (unpublished). When one takes the entire fic number for a ship, Pairing Option-fics represent 23%, 24% and 20% for SS, NS and KS, respectively.

With my current skill-set it is not possible to analyze fics tagged with Ship Names, because ff.net doesn’t have options of sorting results of text-searches according to Favs or Reviews (only available options are: Relevance, Update Date and Publish Date).

Therefore, this analysis represents at best the recent state of the fandom, and definitely does not represent the situation during fandom heyday in years 2007-2009.

Sorting by the number of reviews is reliable, but for unknown reasons sorting by favourites is less so. The obtained lists were not perfectly ordered from highest to lowest values in their lowest ranges (for the fics with less than 15 favs). It might be caused by the fact that favs are granted relatively often, and the sorting corresponds to actual number of given favs that are already in the system but not yet visually displayed. Nevertheless, I oriented myself exclusively by visually displayed number of favs per fic. Therefore, I needed to make manual checks in the lower ranges of favourite lists, which increased the probability of errors.

Results

The analyzed ships vary widely in size. Using Pairing Option approach there were: 5905 SasuSaku fics, 1725 NaruSaku fics, 702 KakaSaku fics and 120 MadaSaku fics available for analysis.

Let’s start with favourites per fic (Figure 1.)

It can be seen that for SasuSaku, NaruSaku and KakaSaku the most popular bin is the bin 21-50. I.e. most of the fics (1511 for SasuSaku, 581 for NaruSaku and 214 for KakaSaku) have between 21 and 50 Favs. In the highest, the “crazy-popular” category (bin >500 Favs) 120 SasuSaku, 33 NaruSaku and 12 KakaSaku fics can be found.

Otherwise the data are quite symmetrically distributed, but one cannot say too much in terms of comparisons, because the ships are very different in size. But before we move further it is worth noticing though that 78 fics of SasuSaku have only 1 Fav and 52 fics have none!

To compare the readers behavior between the ships one needs to examine the distribution in percentages (i.e. how many percent of a given ship’s fics fall into each bin) (Figure 2.)

It is possible to see that even though the most occupied category is always bin 21-50; it is occupied by 25% of SasuSaku fics, 30% of KakaSaku fics and almost 35% of NaruSaku fics. I.e. the distribution is the “flattest” in case of SasuSaku.

The “popular” side of the distribution looks very similar for all the ships, with maybe a bit of overrepresentation of KakaSaku in bins 51-100 and 101-200. However, the less popular side of the graph features a prominent overrepresentation of SasuSaku fics. There is more SasuSaku fics in bins 2-5 and 6-10 (in bin 2-5 twice as much as other fics). And the last categories: bins “1 Fav” and “0 Favs”, are populated almost exclusively by SasuSaku fics.

So, we can conclude that it is equally probable to get very good response in terms of Favs for all the three ships (i.e. have your fic fall into bins >51 Favs). If you’re publishing NaruSaku and KakaSaku, you can be almost sure of receiving at least intermediate recognition (11-50 Favs). This is however at all not the case if you are writing SasuSaku, where as much as 10% of works get virtually no recognition (7,8%, 1,3% and 0,9% falling into 3 lowest bins, respectively)

 Let’s move the parameter held in highest regard by the authors – the reviews! Real numbers for SasuSaku, NaruSaku and KakaSaku are presented in Figure 3.

For SasuSaku fics, four middle bins are populated by similar number of fics (1000+ fics in bins 2-5, 6-10, 11-20 and 21-50). NaruSaku fics show maximum in the bin 11-20, and KakaSaku is similar to SasuSaku (but of course much smaller in terms of numbers).

On the high end of spectrum, there are 111 SasuSaku fics in “crazy-popular” category (>500 Reviews), accompanied with 18 of NaruSaku and 8 of KakaSaku.

However, on the low-end of the distribution, SasuSaku shows again (as with Favs) its less encouraging facet: 191 fics have only 1 review, while 91 - none at all. The same is the case for 31 and 6 NaruSaku fics and 16 and 6 KakaSaku fics.

To compare readers behavior between the ships, let’s take a look at the percentages (Figure 4.)

Almost twice as much SasuSaku fics percentage-wise fall into the “crazy-popular” bin in comparison to other two analyzed ships (1,9% opposed to 1% and 1,1% of NaruSaku and KakaSaku). On the other hand, even more SasuSaku fics rate very low in terms of reviews – 3,2% have 1 Review and 1,5% - none at all; while the percentages in lowest bins are much lower for KakaSaku and, especially for NaruSaku.

It seems that writers of NaruSaku can await a moderately high response (majority of fics localizing in bins 11-20 and 21-50), while readers of KakaSaku probably tendentially lower one (majority of fics in bins 6-10 and 11-20). The response for SasuSaku fics is very varied, with undefined, broad maximum between 2 and 50 reviews per fic.

 SasuSaku, NaruSaku and KakaSaku represent well established, popular and widely recognized Sakura’s ships. After analyzing what can be called standard readers’ behavior, let’s compare reader’s engagement with a relatively rare and exotic ship – MadaSaku.

MadaSaku is represented with 120 fics in the analysis. For the sake of clarity of graphs, I will present results only as percentages (as numerical differences between ships are so big that data would not be legible otherwise).

Favs per fic are shown in the Figure 5.; data of the three previously analyzed ships are included for comparison.

It is immediately apparent, that “crazy-popular” category is drastically overpopulated in case of MadaSaku (7,5% of MadaSaku fics have more than 500 Favs!). Other bins on the popular side of the spectrum are also significantly over-represented (13,3% in 210-500 Favs and 20% in 101-200 Favs). Even though the 7,5% from the highest category means in terms of real numbers only 9 fics, it is nevertheless noteworthy.

Further supporting the trend of increased inclination towards giving Favs among the readers of MadaSaku, the lower part of spectrum is virtually non-existent. There is only 1 (!) fic that has between 0 and 10 Favs, all other 119 fics have 11 or more!

So, how do the reviews look? Figure 6 compares MadaSaku reviews with other three ships.

The over-representation of “crazy-popular” category is again, as was the case for Favs, very prominent. There are 5,8% (corresponding to 7 fics) in the highest bin.

Second-, and third-best categories (bins 201-500, and 101-200) are also a bit over-represented. Towards the middle of the distribution things start looking more similar to the bigger ships. Nevertheless, the lower categories (especially bin 2-5) are underrepresented, pointing to the observations that reviews are (even though to the lesser extent than the Favs), given more eagerly to the MadaSaku fics when compared to the three major Sakura’s ships.

Conclusions

SasuSaku, the largest Sakura’s ship sports a very broad distribution of both Favs and Reviews. That means in practice that while it is maybe more probable for a fic to became crazy-popular, it is even more likely that it will go completely under radar and won’t be acknowledge in any form or way.

NaruSaku and KakaSaku readers seem to “reward” their writers in much more consistent fashion, probably due to smaller “supply” of fics, and almost all fics enjoy some form of readers’ engagement.

MadaSaku, being by far smallest ship, follows a very interesting trend – it contains as many “crazy popular” fics as KakaSaku (which is 6 times larger) and only 3 times less of those fics that NaruSaku (which is 14 times larger). It additionally has no fics “left behind” in terms of Favs. One can speculate that the audience of this ship is much larger than the writers’ base and because of the rarity of published material readers are more prone to show their support. Also, within such a small ship it is virtually impossible to go undetected or to disappear under the pile of newer updates, so every fic gets to be read.