Using QCA to identify what makes a happy ending for Fruits Basket characters

Cover for the first volume of Fruits Basket Collector's edition.

Fruits Basket is a Japanese manga (comic) series by Natsuki Takaya that goes through the themes of love, belonging and change. In this series, the heroine Tohru meets and befriends members of the Sohma family who are afflicted with the zodiac curse, where they turn into animals when they are weak, stressed or embraced by anyone of the opposite sex. By drawing on her experiences of being with her mother before her death, Tohru helps the Sohma family members overcome their past and present struggles to become better versions of themselves. In return, Tohru receives love and support from members of the Sohma family which culminates in her wanting to break the zodiac curse to free them.

Fruits Basket has a lot of characters that undergo different journeys throughout the series to achieve different endings. These characters vary in terms of how much they change throughout the series, and how open they are to interacting with other people. In this blog post, I wanted to explore the key factors that determine whether a character in the Fruits Basket manga achieves a happy or sad ending. To answer this question, I used qualitative comparative analysis (QCA) to understand each character, and to identify different ways that characters can achieve a happy or sad ending.

A brief overview of QCA

QCA is a case-based research method that draws on case studies to identify multiple pathways to the outcome of interest. Each pathway consists of a combination of factors that work together to produce the outcome. I used QCA in my project for a number of reasons:

We can look at multiple factors at the same time, and simplify them to a set of pathways that lead to the same outcome. This allows us to describe how the outcome can occur in different ways.
We can test both the presence and absence of an outcome, deriving different pathways in the process.
We can preserve the detail of our case studies, drawing on them as examples to explain how a specific combination of factors leads to or does not lead to the outcome of interest.

QCA works by first understanding the case studies, scoring them on the presence or absence of the outcome and factor, and using them to identify a set of pathways. This process can be described in six steps:

Outcome and factor definition: I initially defined the outcome and factors for my QCA, as well as how they will be scored, based on background research of what may affect a young person. I placed the definitions and scores in a rubric which was constantly updated as I read the manga. In the end, I came up with the below factors:
- Character-parent relationship (relationship): What is the relationship between the character and their parents like?
- Social interaction (interaction): How much does the character interact with other people?
- Conformity to authority (conformity): How much does the character follow the orders of authority figures (such as their parents, the Sohma family head…)?
- Character development (development): How much does the character change throughout the manga?
Case selection: I listed some Fruits Basket characters that I wanted to investigate further. I included the main characters of the manga, including the Sohma family members who are part of the zodiac curse, as well as some side characters. In total, I looked at 21 characters in the Fruits Basket manga series.
Case description: I read the manga and made notes on most chapters. I also read the pages of the Fruits Basket Wiki to collect information on each character. Information from both sources was coded for the outcome and individual factors.
Case summary: Based on what I collected from the manga and the Fruits Basket Wiki, I summarised information about each character in an Excel spreadsheet.
Data analysis: For each character, I scored each outcome and factor based on the rubric. I then imported the scores into fsQCA software. The software uses Boolean algebra to simplify the scores into a set of pathways that leads to the presence or absence of the outcome.
Case linkage: I checked each pathway against the case studies to test the link between the factors and outcome and to produce an explanation behind each pathway.

Throughout the process, I went back to earlier steps of QCA to change some things depending on the results I received. These included redefining the rubric by changing definitions or scores for the outcome and factor, adding, removing or combining some factors, or re-reading the manga or Wiki pages to collect more information about the character.

In the end, I identified three pathways that can help characters achieve a happy ending. I also discovered a pathway that led to a sad ending for the character. Each pathway is represented by a combination of factors that work together to achieve the outcome of interest (represented by the multiplication sign (x)). I will explain what each factor combination means.

Pathway 1 to a happy end

Interaction x development -> outcome

Characters who are willing to interact with other people and change themselves throughout the manga often get happy endings. The majority of the characters analysed (13 out of 21 characters) fell under this pathway, with two of the main characters being exemplars of this pathway.

Tohru

Tohru is the main heroine of the manga, meeting all the members of the Sohma family.

As a child, she felt lonely, with other children excluding her from games. Despite this, she made two close friends in middle school (Arisa and Saki) who support her. During the manga, she interacts very positively with all members of the Sohma family, providing support and unconditional love to them. This love even extends to Akito, the Sohma family head who was antagonistic towards her for most of the manga. Despite this, Tohru still reached out to Akito when she became despondent about everyone leaving her. These examples show her willingness to interact with other people throughout the manga.

Tohru is already a positive character at the beginning of the manga, being open-minded, cheerful and optimistic. However, these hide behind her struggle to be honest with her feelings, particularly her love towards Kyo, a member of the Sohma family who is part of the zodiac curse. This conflicted with her need to not forget her recently deceased mother. However, she gradually becomes more honest with her feelings towards Kyo which fuels her desire to break the zodiac curse to free him. She moves on from her mother’s death by accepting her feelings towards Kyo and reciprocating Kyo’s love in return. Eventually, she moves out of town with Kyo to support his martial arts training in the countryside. This shows the amount of character development Tohru underwent as she moved on from her past to be with the person she loved.

Kyo

Image of Kyo from the anime (2019 series)

Kyo is an orange-haired boy who is afflicted with the curse of the cat. He is often referred to as the ‘black sheep’ of the Sohma family as the other members look down on him.

He is socially awkward, wanting to keep other people at arm’s length as he is tired of other people pitying him. Consequently, he only associates himself with a few people, particularly Kazuma, a martial arts master, who acts as a supportive foster father towards him. He also gradually warms up to Tohru, telling her to take care of herself and to be more honest with her feelings. He eventually falls in love with her as he received unconditionally support from Tohru throughout the manga. These examples show that despite Kyo interacting with few people, he views Kazuma and Tohru as important people in his life.

He experienced a lot of guilt and self-loathing from past events in his life. These included his mother’s suicide, his estrangement from his biological father and him running away from Tohru’s mum as she lay dying from being hit by a car. He initially coped by pinning all the blame on Yuki, his cousin, and wanting to beat him in battle. However, thanks to Kazuma and Tohru, he gradually opens himself up to other people and to confront his past in an effort to move on. This culminates in Kyo facing his biological father and, in the face of his father throwing insults at him, declares that he will not give up on himself to live a normal life.

He eventually confesses to Tohru and becomes her boyfriend. He later moves out of town with Tohru to continue his martial arts training in the countryside so that he can succeed Kazuma in the future. These events highlight his journey from being a negative person at the beginning of the manga to a positive person at the end.

Pathway 2 to the outcome

Conformity x development -> outcome

Despite characters conforming to their parents or authority figures in the past, if they are open to changing themselves throughout the manga, they can still achieve a happy ending. From the manga, there are 5 characters (out of 21) who fall under this pathway.

Yuki

Yuki is a shy, silver-haired boy in the Sohma family who is afflicted with the curse of the rat.

As a child, his mother ordered him to stay by Akito’s side all the time so that she could benefit from his son being close to god. Yuki stayed by Akito’s side all the time, not interacting with other members of the Sohma family. He was also verbally, emotionally and psychological abused by Akito, and was often locked up in a dark room for days while Akito tormented him. His mother also neglected him, even going so far as slapping him and telling him to go back to Akito when he came to her for support. This shut off any avenue of rebelling against his mother or Akito, forcing him to conform.

As a result of his neglected and traumatic childhood, he suffered from severe self-esteem issues and loneliness. However, through the support of people such as Tohru and Ayame (his older brother), he gradually finds the courage to stand up and improve himself. A prime example of this is his defiance towards his mother during the parent-teacher conference, where he told her not to make decisions about his future on his behalf. He also learns not to blame other people, something that is shown when he found the courage to forgive Akito despite her abusive past. As a result, he becomes more confident of himself, takes more control of his life, and is close to many people, particularly his girlfriend Machi. This shows the massive amount of growth Yuki experienced from the beginning of the manga, and how he benefited from it.

Machi

Machi is a girl who is suffering from the aftereffects of an inheritance war.

Machi conformed quite strongly to her mother in the past, being pressured to be perfect in every aspect of her life so that she can be the next family heir. However, when her little brother was born, she was cast aside as her mother now had a male heir. When she was wrongfully accused of trying to murder her little brother, she was forced to follow her mother’s orders to move out of the family house and live in an apartment alone.

As a result of her upbringing, she did not have time to discover who she was as an individual. Consequently, she had low self-esteem, and felt that she did not have a personality. The pressure to be perfect at a young age also caused her to despise perfection. This hatred drove her tendency to destroy anything that she considers too tidy such as wrecking her apartment or the student council room and breaking a brand new box of chalk. However, thanks to Yuki’s support, she starts to learn what she likes as an individual and tries her best to change. As a result, she becomes more approachable, expressive, honest and communicative. She also falls in love with Yuki, something that he reciprocates by the end of the manga. This shows how much Machi had changed in the manga, allowing her to have a happy ending.

Pathway 3 to the outcome

-interaction x conformity x relationship (x development) -> outcome

Even if the character does not interact much with other people and conforms strongly to their parents or authority figures, if they have a positive relationship with their parents they can still achieve a happy ending. Only 3 characters out of 21 fell under this pathway, and even then it was difficult to explain how this pathway led to a happy ending. Adding the character development factor helped explain this pathway more clearly.

Kisa

Kisa is a blond, middle-school girl in the Sohma family who is afflicted with the curse of the tiger. As a result of other middle school students teasing her about her blond hair, Kisa goes mute and shuts herself off from other people. Even when she talks again, she only interacts with a few people, becoming attached to Tohru and Yuki and slowly becoming close to Hiro, her childhood friend, again.

She was not rebellious towards Akito or her parents. Rather, she was slapped by Akito after Hiro confessed that he loved Kisa. Despite this, Kisa loved her mother, not wanting to trouble her by telling her that she was being bullied. This left her no choice but to run away. Even though Kisa’s mother panicked when Kisa ran away, her fears were allayed when Tohru explained why Kisa did not want to talk about being bullied with her. Kisa’s mother also loved her dearly and wanted to keep in contact with her, shown by her contacting Tohru every day to pass on some tips of how to take care of Kisa.

However, it was thanks to Tohru and Yuki sharing their experiences with Kisa that she realises that she needed to change or things would get worse for her. As a result, she starts to return to school and face her fears. Consequently, she not only becomes more confident in herself, but also grows closer to her childhood friend Hiro, supporting one another throughout the manga, particularly after their curses break towards the end of the manga.

Ritsu

Ritsu is a blond-haired, cross-dressing male who is clumsy, over-apologetic and afflicted with the curse of the monkey.

He is afraid of interacting with other people for fear of offending them. Consequently, he only interacts with a few people. However, he received parental love from his mother who accepted him unconditionally despite his clumsiness. In return, he loves his parents, to the point of apologising to them for feeling useless in the world despite their assistance. He also decides to succeed his mother in being the host of the Sohma onsen of his own free will. Also, thanks to Tohru, Ritsu thinks about what he wanted to do in his life, and to find his reason to live in another person. Eventually, he not only becomes more confident in himself, but also becomes friends with, and eventually dates, someone who had a similar personality as him.

What leads to a sad ending?

-Relationship x -Interaction x -Conformity x -Development -> -Outcome

A combination of all of these factors can lead to a sad ending for the character:

The absence or negativity of the character’s relationship with their parents.
An unwillingness to interact with other people.
A tendency to rebel against their parents or authority figures.
An unwillingness to change throughout the manga.

The above factors can explain how two characters had sadder endings than other characters in the manga.

Akito

Akito is the head of the Sohma family who controls the other members of the zodiac curse as their ‘god’.

Akito had opposing relationships with her parents, loving her father but hating her mother. Her father was kind and affectionate towards Akito, telling her that she is special and that she is born to be loved by everyone. In contrast, her mother hated Akito for stealing her husband’s attention from her. Upon her husband’s death, she taunts Akito by saying she is unneeded and worthless and that other people will abandon her. Akito responds by being defiant towards her. She tries to show her mother that she is still loved and needed by abusing many members of the zodiac curse and telling them to come back to her.

As a result, despite warnings from other characters that things are changing, she still clings onto the belief that everyone will come back and respect her. When the other characters leave her as their curses break, she breaks down. It was by meeting Tohru that she learns she does not have to be special and that she can be herself. As a result, she willingly breaks the rest of the zodiac curse to free everyone.

Nevertheless, she faces significant challenges at the end of the manga as she atones for her past actions. These included excluding herself from future family events and continually being berated by her mother.

Isuzu

Isuzu is a young woman in the Sohma family who is afflicted with the curse of the horse.

Her relationship with her parents is non-existent. Her parents felt Isuzu was a burden to their lives, abusing and abandoning her after she questioned whether they were happy. They eventually disown her as they did not know how to love her anymore. Additionally, she is thrown out of the window by Akito after she confessed that she loved Hatsuharu who is afflicted with the curse of the ox. Consequently, she is driven to break the zodiac curse for her boyfriend, even going so far as accepting a request from Akito’s mother to find Akito’s secret. She is found out by Akito who punishes her by cutting her hair and imprisoning her in a room.

As a result, Isuzu is a negative person for most of the manga who does not meaningfully change throughout the manga. She is also distant towards other people, pushing them away as she did not want to show weakness in an effort to break the zodiac curse. She only changes after she is freed from Akito and falls into Hatsuharu’s arms.

Consequently, she becomes a broken person by the end of the manga, crying when her curse breaks. In contrast to the other characters who move on with their lives, Isuzu is stuck to her past, not forgiving Akito for what she did to her and not knowing what she wanted to do in her life.

Conclusion

Poster of the Fruits Basket anime series (2019 version)

I conducted a QCA to identify some ways that characters in the Fruits Basket manga series achieve happy or sad endings. From this analysis I identified three pathways that characters can follow to achieve a happy ending. In all three pathways, character development, describing how open the character is to changing, plays a key role in achieving a happy ending. I also identified a pathway that led to a sad ending for the character. This pathway relates to the social isolation and lack of development the character experiences as they focus on achieving their goals (Akito in proving her mother wrong, and Isuzu in breaking the zodiac curse). If the characters had been more open to receiving help and support from other people, or in changing themselves in the face of shifting circumstances, their endings may have been happier.

By identifying different ways that characters can achieve happy or sad endings, we can identify key results that we can apply to real life. Namely, by surrounding ourselves with people who support us and by being open to change and self-development, we can find a happy, fulfilling life for ourselves.

AI statement

All content is human-created and verified. No AI tools were used in collecting or analysing data, or in writing this blog post.

Graphing in I CAN Network’s 2023 Social Impact Report

I CAN Network is an Autistic-led social enterprise that aims “to prove what Autistics CAN do”. It is working towards this aim by running a variety of programs that engage Autistic people and their supporters (for example, parents, carers and teachers). In particular, I CAN Network runs peer mentoring programs in school and online environments for Autistic young people aged 5-20 years to come together and build their social connections, improve their self-esteem and develop life skills. It also runs professional development programs for schools to create safe environments for Autistic young people to be themselves and engage in school.

Over the past 10 years, I have been working on internal evaluations for I CAN Network. I planned and conducted evaluations measuring outcomes of I CAN Network’s programs and wrote numerous reports describing the results of these evaluations. These reports describe the outcomes I CAN Network has achieved in its programs, particularly peer mentoring programs for Autistic young people. While most reports are delivered to state governments and funders, I also wrote Social Impact Reports that are publicly available on I CAN Network’s website, with the 2023 edition being recently released.

With the 2023 Social Impact Report being the last evaluation report I will write for I CAN Network, in this blog post I will describe why I have designed the graphs in a particular way to make them easy for the reader to interpret and understand.

What are the aims of I CAN Network’s programs?

I CAN Network runs school and online peer mentoring programs for Autistic young people aged 5-20 years. At school, I CAN Network runs I CAN Imagination Club^® and I CAN School^® mentoring programs for Autistic young people in primary and secondary schools respectively. These programs are run during school hours. In comparison, I CAN Network runs I CAN Online, delivering online mentoring programs after school for Autistic young people across Australia.

All peer mentoring programs aim to achieve outcomes in the following three areas:

Self-esteem: The peer mentoring programs aim to improve Autistic young people’s views of themselves and their Autism.
Social connection: Autistic young people in the peer mentoring programs interact with each other to establish social connections and friendships, as well as their parents, teachers and mentors to develop support networks.
Skill development: These relate to Autistic young people acquiring and improving life skills such as communication and stress management so that they are more likely to advocate for themselves and manage their own lives.

These outcome areas are supported by a positive program environment, where Autistic young people feel safe to be themselves without feeling judged. This allows them to open up and interact with other people.

The below figure summarises the outcomes that I CAN Network’s peer mentoring programs aim to achieve.

Venn diagram of the three outcome areas of I CAN Network's peer mentoring programs, encompassed by a bigger circle representing 'positive program environment'. — A visual summarising the aims of I CAN Network’s peer mentoring programs

These aims provide a structure in which the peer mentoring programs can be assessed. One way this can be done is by running internal evaluations within the organisation.

Why I have not used mean scores in the report

I CAN Network runs internal evaluations by distributing surveys and polls to Autistic young people before and after the program. Each survey or poll contains statements relating to the outcomes of the program, where mentees rate on a 3- or 5-point Likert scale how much they agree with the statements. We then compare responses before and after the program to see whether mentees have changed during the program. These can be shown both numerically and graphically.

One way to compare responses is to convert them into a score, calculate the mean score before and after the program and compare them to see whether that has changed. For I CAN Imagination Club^® mentoring programs, I can convert the yes, maybe and no responses into scores of 1, 0.5 and 0 respectively and calculate a mean score out of 1. Similarly, for I CAN School^® mentoring programs, I can convert the responses into scores ranging from 1 (for strongly disagree) to 5 (for strongly agree) and calculate a mean score out of 5.

We can compare the mean scores before and after the program in a table along with the standard deviation (SD which measures the spread of scores) and the total number of responses received.

Statement	Before			After
Statement	Mean score	SD	Total # responses	Mean score	SD	Total # responses
I can try new things	0.74	0.29	331	0.80	0.27	303
I know what makes me special	0.70	0.37	323	0.79	0.34	302
I think my brain is awesome	0.69	0.36	327	0.76	0.34	298

We can also visualise the scores in a bar chart, placing the mean scores before and after the program alongside each other. We can rule black solid lines to separate the statements, making it easier to compare mean scores before and after the program within a statement.

Bar chart comparing mean scores before and after I CAN Imagination Club program over three outcomes — Mean scores before and after the program when the scale starts from 0

There are two main problems with using mean scores to compare outcomes. Numerically, calculating mean scores simplifies the responses too much. This makes it difficult to explain a specific mean score (“What does a mean score of 0.74 mean?”) or changes in a mean score (“What does a mean increase of X units mean?”) and relate them to the outcome. Furthermore, because we design our own statements instead of using a standardised tool, it is not possible to compare the changes to a specific standard to explain how relevant they are. The only way to explain the changes is to use statistics to assess whether the changes are statistically significant and relevant. This makes it difficult to explain to a lay audience how large outcomes have changed.

Bar chart comparing mean scores before and after I CAN Imagination Club program over three outcomes when the scale starts from 0 — Mean scores when the scale starts from 0

Bar chart comparing mean scores before and after I CAN Imagination Club program over three outcomes when the scale starts from 0.62 — Mean scores when the scale starts from 0

Graphically, it is hard to accurately visualise changes in mean scores. Scaling the scores from 0 makes it very hard to differentiate mean scores before and after the program. In contrast, scaling the scores from a higher base value distorts the differences before and after the program, deceiving the reader. For example, in the ‘I can try new things’ statement, the mean score only increased by 0.06 after the program. That difference is quite small when scaling the scores from 0, but it is magnified when scaling the scores from a higher base value (specifically 0.62), distorting the difference.

For the above reasons, I decided not to describe changes in outcomes using mean scores. Instead, I represented the changes in a different way.

Why I have used 100% stacked bar charts in the report

To represent changes in outcomes among mentees attending the peer mentoring programs, I decided to compare the distribution of responses before and after the program. For each timepoint, I calculate the proportion of responses belonging to a particular category out of the total number of responses received (excluding not sure and missing responses). I then bring the categories together to form 100% stacked bar charts. I use black solid lines to separate the statements and place the distribution of responses before and after the program next to each other to make it easier to compare them.

100% stacked bar chart comparing the distribution of responses before and after I CAN Imagination Club over 3 outcomes — Changes in I CAN Imagination Club^® outcomes, expressed as proportions of the total number of responses

Using the distribution of responses is a better way to represent changes in outcomes compared to using mean scores. Numerically, it is easier to explain changes in outcomes by saying how much responses in a particular category have increased or decreased after the program. For instance, in the I CAN Imagination Club^® graph above, we can compare the proportion of ‘yes’ responses before and after the program and calculate how much it has increased. We can see that compared to before the program, mentees coming into the program show an:

11% increase in trying new things (‘self-confidence’);
14% increase in knowing what makes them feel special (‘self-acceptance’); and
9% increase in thinking their brain is awesome (‘neurodiversity acceptance’).

Representing outcomes as percentage changes makes it easy to explain to the reader how Autistic young people have changed in the peer mentoring program.

100% stacked bar chart comparing the distribution of responses before and after I CAN School over 4 outcomes — Changes in I CAN School^® outcomes, expressed as proportions of the total number of responses

Graphically, showing the distribution of responses makes it easier to visualise how different responses shift over time. For instance, in the I CAN School^® graph above, we can see that the proportions of ‘strongly disagree’ and ‘disagree’ responses have fallen, while the proportions of ‘agree’ and ‘strongly agree’ responses have risen. Visualising the distribution of responses in this way makes it easy for the reader to interpret the result to see whether Autistic young people are improving as a result of the program. It also does not deceive the reader as the graph is scaled from 0% to 100%, ensuring that the changes are properly represented.

Hence, using 100% stacked bar charts to show the distribution of responses allows us to retain all the information we have collected from our surveys and makes it easier to explain changes in outcomes to a lay audience.

Visualising mentee demographics using different graphs

In the 2023 Social Impact Report, I have also visualised some demographic data to describe who participates in I CAN Network’s peer mentoring programs. I used two different graphs to visualise demographic data: pie charts and bar charts.

There has been some talk as to why pie charts are bad visualisation tools and that bar charts should always be used when looking at proportions. In my opinion, pie charts are still useful for visualising proportions. However, there are some conditions that need to be met:

It should show as few categories as possible (five at most).
It should be as visually simple as possible, with the parts clearly signposted.
The pie chart should be in 2-D instead of 3-D to prevent distortion of the data.

Pie chart showing the proportion of I CAN Online mentees belonging to different genders — Gender among I CAN Online mentees as a pie chart

For gender, I used a pie chart as it only has three categories (male, female and trans and gender diverse), making it easy to show the sizes of each gender as a whole. I have also kept the pie chart as visually simple as possible.

I have earmarked each section with a solid black outline and different colours representing each gender.
I have also included some data labels indicating what each section represents, along with the number of mentees and the proportion out of the total number of mentees.

Using a pie chart provides a clear picture of how mentees are split up over the three gender categories.

Pie chart showing the proportion of I CAN Online mentees belonging to different Australian states and territories — States and territories among I CAN Online mentees as a pie chart

In contrast, I used a bar chart instead of a pie chart to represent Australian states and territories where I CAN Online mentees live. A pie chart is not appropriate as there are eight states and territories in Australia, resulting in numerous sections being crammed in one space. Using different colours to represent each section and data labels to identify different sections further increase the visual complexity of the graph. Hence, it is hard for the reader to comprehend what proportion of mentees live in a specific state or territory.

Bar chart showing the proportion of I CAN Online mentees belonging to different Australian states and territories — States and territories among I CAN Online mentees as a bar chart

In comparison, using a bar chart provides more room to show different categories. It is also easier to look at the length of each bar and align it to the y-axis to see what proportion of mentees live in a specific state or territory. The limitation of a bar chart is that it is hard to see how different states and territories can be pieced together to form a whole like what can be done in a pie chart.

Hence, pie charts have their uses, but their benefits and drawbacks compared to bar charts need to be considered when deciding how to visualise the results.

Conclusion

In this blog post, I explained how I have designed the graphs that appear in the 2023 Social Impact Report. People may have different opinions of how data should be analysed and visualised. From my perspective, I place great importance on presenting the results as simply as possible so that everyone can easily grasp what they are seeing. This facilitates their understanding of how I CAN Network’s programs are making an impact to Autistic young people attending them.

Personal disclaimer

This blog post was written by James Ong in his personal capacity. The content, views and opinions represented in this blog post are solely my own and do not reflect those of I CAN Network Ltd.

The Fall of SIFAS: A sacrifice for the greater good?

April 30^th 2023 saw the end-of-service (EOS) announcement of Love Live! School Idol Festival ALL STARS (hereafter ‘SIFAS’), a rhythm action RPG that shut down on June 30^th 2023 for both the Japan (JP) and Global (GL) servers. This followed the shutdown of Love Live! School Idol Festival (hereafter ‘SIF1’), a pure rhythm game, for both the JP and GL servers on March 31^st 2023. Around that time, Love Live! School Idol Festival 2 MIRACLE LIVE! (hereafter ‘SIF2’), the successor to SIF1, was released to JP players on April 15^th 2023, with a global release coming in 2023.

From ambitious beginnings, SIFAS has fallen by the wayside. Behind a flawed game design and mismanagement by KLab, MyNet Games and Bushiroad, the game experienced declines in player counts and revenues in both the JP and GL servers. Having written four blog posts on the decline of SIFAS, I thought it fitting to provide a final update of the last few months of SIFAS, and what motivated the shutdown of the game.

A final update on player counts

Surprisingly, after the constant falls in player counts over the last 6 months of 2022, player counts have stabilised or even improved slightly in the first 3.5 months of 2023.

Line graphs of player counts of SIFAS in the Japan server over 2022-23. — Green represents Love Live! Nijigasaki Season 2 anime airing, red represents 3 year anniversary period, blue represents 3.5 year anniversary period and red dashed line represents end-of-service (EOS) announcement

In the first three months of 2023, the JP server had around 46,000 players for item exchange (itex) events in the first half of each month and around 45,000 players for story events in the second half of each month. After the 3.5 year anniversary period (blue region), April 2023 saw a small decline of around 1,000 players for the itex event, but a massive fall by 5,000 players (or around 11% of the player base) for the story event. These declines in player counts occurred even before the EOS announcement.

After the EOS announcement, both itex and story events experienced massive falls in player counts in May 2023. In the JP server, there was a 16% decline in player counts for itex events from 45,279 to 38,162 players. Similarly, for story events, there was a 10% decline in player counts from 40,759 to 36,506 players.

In the GL server, player counts increased slightly in the first four itex events, rising from 15,731 players in January 2023 to 16,497 players in April 2023. For story events, player counts stabilised around 15,000 in January and February 2023 before increasing by 1,000 players to 16,089 players in March 2023.

Similar to the JP server, even before the EOS announcement, we saw a 10% decline in player counts in the GL server for the April 2023 story event, decreasing to 14,534 players. After the EOS announcement, there were massive falls in player counts in the GL server in May 2023. Player counts for the itex event decreased by 17% to 13,755 players. The decline in player counts for story events was smaller than the JP server, only decreasing by 6% to 13,670 players.

Both the JP and GL servers experienced alarming declines in player counts even before the EOS announcement, and decreased even more afterwards. These results indicate that even the stabilising player counts could not save the game. Bushiroad decided in advance that SIF2 would replace both SIF1 and SIFAS, owing to the declines in player counts in both games over 2022.

A final update on revenues

Revenues (in Japanese Yen millions) in the mobile game SIFAS over 2021-23. — The revenues of SIFAS over 2021-2023 under KLab’s and MyNet Games’ management

2023 continued to see a trough in revenues in the JP server. The game hovered around 100 million JP Yen in January 2023 before falling below that mark for the rest of the game’s lifespan. Not even the 3.5-year anniversary celebrations (which occurred in March 2023) could restore SIFAS’ revenues back to 100 million JP Yen. Matching the declines in player counts in April 2023, we saw a massive fall in JP revenues to 66.3 million JP Yen, the lowest monthly revenue ever generated by SIFAS. The massive fall in revenues appeared even before SIFAS’ EOS announcement.

Comparing SIFAS’ shutdown to other games managed by MyNet Games

In part 3 of the ‘Decline of SIFAS’ series, I put forward the benchmark of 2 years, or 24 months, as the benchmark for determining whether mobile games transferred to MyNet Games would survive or be shut down prematurely. That benchmark is consistent with what MyNet Games presented in their financial reports. I explained that a whopping 85% of games are terminated within the first 2 years of MyNet Games’ operations, highlighting that period as one with the highest risk of a shutdown.

Using 2 years (or 24 months) as the benchmark, SIFAS falls below the mark, only lasting 18 months after being transferred from KLab to MyNet Games before being shut down. From looking at the final player count and revenue graphs, it is hard to fathom whether SIFAS would survive to its 4 year anniversary as a result of low player counts and revenues.

How Bushiroad killed SIFAS

For most of part 3 of the ‘Decline of SIFAS’ series, I described how the transfer of SIFAS to MyNet Games does not bode well for the game’s future. At the same time, I also described how Bushiroad can contribute to SIFAS’ further decline and eventual downfall. That is because Bushiroad, as the publisher, has the final say on whether SIFAS continues on or is terminated. Bushiroad’s actions over the past few months can be seen as them wanting to rapidly move SIF1 and SIFAS players in both JP and GL to SIF2.

The major sign of SIFAS’ imminent shutdown came when the main story was rushed. The main story consisted of six seasons of content. The first three seasons, lasting from September 2019 to September 2022, lasted 9-11 chapters each, with new chapters being released monthly. These are divided by intermission sections that lasted 1-2 chapters, allowing each season to last a year.

A timeline of main story chapter releases over October 2022 to June 2023. — The timeline of main story chapter releases over October 2022 to June 2023.

However, the last three seasons were very short and crammed in between October 2022 and May 2023 without any intermissions. Season 4 (in green) lasted from October to December 2022, with 3 chapters being released monthly. After the last chapter of season 4, season 5 (in blue) started right away, with 4 chapters being released monthly from January to April 2023. By the time of the EOS announcement, season 6 (in purple) was started, with the remaining 3 chapters crammed within a month. Ironically, if those chapters had been released monthly, there would have been enough main story content to tide SIFAS along until July 2023, meaning the game would have been shut down by the end of August 2023.

There were other red flags indicating that SIFAS was going to be shut down:

In mid-April 2023, Voltage Ranking songs for itex events were announced for May 2023 but not June 2023. Other similar notices would normally announce 2 months’ worth of Voltage Ranking songs. This indicated that SIFAS would be shut down around June 2023.
The reduced emphasis on new game modes and the increased reliance on existing game modes became apparent when MyNet Games introduced voltage ranking campaigns in mid-April 2023, a few days before the EOS announcement. MyNet Games did not introduce any new game modes or improved gameplay that kept the game fresh throughout its whole management period.
There was never any intention from Bushiroad to run SIFAS alongside SIF2. Not only did Bushiroad plan SIFAS’ shut-down in advance despite stabilising player counts in 2023, but they also made the point clear in the EOS notice:

“We regret to inform you that on June. 30, 2023 16:00 JST, we will be discontinuing the service for ‘Love Live! School Idol Festival All Stars’. With this, the ‘Love Live! School Idol Festival’ series of apps will be consolidated into ‘Love Live! School idol festival 2 MIRACLE LIVE!’.” (emphasis added)

The irony of shutting down SIFAS

As previously mentioned, Bushiroad had enough main story content to prolong the availability of SIFAS to August 2023, when it would be shut down. This implied that Bushiroad wanted to move SIFAS players onto the global version of SIF2 as quickly as possible around mid-July 2023 or the beginning of August 2023.

However, the rush to release remaining SIFAS content before its shutdown was unneeded because as of Anime Expo 2023, there was no official date of when SIF2 would be released worldwide. Bushiroad could have taken its time to release the remaining chapters of the main story monthly. By keeping SIFAS open for a longer period of time, it would buy time for the SIF2 developers to optimise the game for a global audience. Without SIFAS, there is no momentum to direct global SIF1 and SIFAS players to SIF2, where they would be lost to other mobile games. This would make it difficult to bring Love Live players back together to SIF2 when it eventually releases globally, presenting a massive lost opportunity for Bushiroad.

Conclusion

This blog post marks the sad end of SIFAS, and the conclusion of the ‘Decline of SIFAS’ series. Under both KLab and MyNet Games’ management, SIFAS has seen declines in both its revenue and player counts, with no recovery in sight.

There were many problems that have plagued SIFAS over its lifetime. The flawed game design tried to straddle the RPG and rhythm game genres without being good at either of them. This was made worse by the controversial season 2 story, stagnant gameplay and pay-to-win mechanics that neither KLab nor MyNet Games fixed. KLab also saw SIFAS as a ‘dead weight’ that did not live up to its potential, prompting its sell-off to MyNet Games. What prematurely ended SIFAS; though, was Bushiroad moving too quickly to get players to play SIF2 when it was not yet ready for a global audience.

The ‘Decline of SIFAS’ blog series provides a cautionary tale of what happens when a game with a flawed design is released to the world, and problems are not responded to right away to keep players in the game. Given the mad rush by game developers and publishers to release mobile and live service games, the story of SIFAS’ downfall is timely indeed.

The Decline of SIFAS (2022 update): Stable condition hides the milking of golden geese

In my previous posts of the ‘Decline of SIFAS’ series, I described why Love Live! School Idol Festival ALL STARS (SIFAS) is failing in KLab and what my thoughts were in transferring development of the game to MyNet Games. In 2022, the situation in SIFAS had somewhat stabilised, with the fall in revenues now stopped. However, SIFAS faced numerous issues in 2022 such as lower revenues, a shrinking player base and the move towards pay-to-win (P2W) tactics. In this post, I would like to provide an update on how SIFAS is going, and provide one possible reason why MyNet Games was facing the current situation in SIFAS over 2022.

SIFAS player counts in 2022

Line graph tracking player counts in SIFAS in the Japan server over 2022 for itex and story events. — Player counts in SIFAS over the Japan (JP) server in 2022

SIFAS continued to see declines in player counts throughout 2022. In Japan (JP), the number of players playing the game has decreased by 22-28%, from 63,361 in the January 2022 item exchange (itex) event to 45,871 players in the December 2022 itex event. Similar results were found for the story events, decreasing from 61,266 players at January 2022 to 47,869 players in December 2022.

Line graph tracking player counts in SIFAS in the Global server over 2022 for itex and story events. — Player counts in SIFAS over the Global (GL) server in 2022

In the global (GL) server, the decline in player counts was just as dramatic, decreasing by 22-31% throughout 2022. From 22,548 players in the January 2022 itex event, only 15,541 players participated in the December 2022 itex event. For story events, January 2022 saw 22,534 players, but this decreased to a low of 16,015 players in November 2022, before coming back to 17,653 players in December 2022 due to the Christmas celebrations. Based on the percentage decreases calculated for 2022, I predicted that by the end of 2023, there would be 33,028-37,338 players in JP and 10,724-11,371 players in GL.

The anime and anniversary campaigns attracted new players to SIFAS (represented by green and pink rectangles respectively in the graphs). This resulted in a temporary increase in player counts that was not sustained after the campaigns ended. The half-year anniversary campaign in March 2022 was associated with an increase in player counts for story events in both JP and GL servers (by 3% and 6% respectively). These increases were sustained in story events that were run throughout April-June 2022. For itex events, there was initial strong interest towards the game, as indicated by more players playing SIFAS in April 2022 compared to January 2022 in both servers. However, that interest was not sustained, with 8-9% players from April 2022 not participating in the May 2022 itex event and a further 1-3% players not participating in the June 2022 itex event. As soon as the anime season ended, interest towards SIFAS quickly fell apart. In particular, from the June 2022 story event, 13% and 15% of JP and GL players respectively did not participate in the July 2022 story event. This result most likely indicates that MyNet Games failed to capitalise on the anime season to retain new players in SIFAS.

During the anniversary celebrations that occurred from September to October 2022 (pink areas), itex events in both servers failed to show increases in player counts, either stagnating or decreasing by up to 4% of the player base. In contrast, the anniversary celebrations had different effects on player counts in the story events for both JP and GL servers. The JP server saw a temporary upswing in player counts, increasing by 9% from 50,942 players in August 2022 to 55,387 players in September 2022. However, that increase was not sustained, decreasing to 52,164 players (6% decrease) in October 2022. In contrast, player growth was sluggish in the GL server, only increasing by 2% from 18,247 players in August 2022 to 18,688 in September 2022 and 18,735 players in October 2022. After the anniversary period, player counts fell precipitously, falling by 7% for itex events in both JP and GL and, for story events, by 10% in JP and, more worryingly, 15% in GL. The player counts for both JP and GL in November 2022 were lower than those of August 2022 (before the anniversary celebrations), indicating that MyNet Games had failed to retain both new and existing players in SIFAS.

A number of things can be drawn from looking at player counts for both JP and GL servers. There were spikes in player counts during the anniversary periods, indicating that the anniversary rewards were attractive enough for new players. However, the increases in player counts were not sustained after the anime and anniversary periods ended, indicating that MyNet Games failed to retain new players.

At the same time, the precipitous declines in player counts in the GL server were very worrying. That put into question the viability of running a separate server for GL players as it would cost money to run two servers. There was a strong chance that MyNet Games will merge the JP and GL servers into one server in 2023, similar to what had happened in Love Live School Idol Festival. This would have major repercussions on retaining meta players as GL players will find it harder to compete in voltage ranking against JP players who had more time and opportunities to obtain and upgrade powerful cards.

SIFAS revenues in 2022

Revenues in SIFAS (Japanese Yen millions) over 2021 and 2022. — SIFAS revenues when the game was run under KLab’s and MyNet Games’ management (pink and blue respectively)

Two arguments can be made on the effect of MyNet Games on SIFAS revenues in JP. On the one hand, MyNet Games have stopped the constant declines in JP revenues when SIFAS was run by KLab (shown in pink). If KLab had continued to run SIFAS in 2022, SIFAS revenues in JP would have hovered around 60 million Japanese Yen by the end of 2022. Its continued decline would have led to its termination by 2023 or 2024. In contrast, since MyNet Games took over SIFAS (shown in blue), revenues remained steady, averaging around 125 million Japanese Yen per month.

However, that flat trendline masks a pattern of peaks and troughs in revenue that were worse than KLab’s. Unlike KLab, MyNet Games did see an ‘anniversary bump’ in revenues, increasing by 41% from August to September 2022. The revenue peaks in 2022; though, only matched KLab’s worst-performing month in November 2021. MyNet Games also experienced much lower baseline revenues than KLab. Half of 2022 had monthly revenues that only hovered around the 100 million JP Yen mark

Collectively, these results indicate that although revenues stabilised in 2022, they did not increase during the year. MyNet Games would become increasingly reliant on anniversary celebrations to collect much of their revenue, with the rest of the year stagnating around a baseline level of revenue. That baseline revenue may fall during 2023; though, as meta players either stop paying or playing the game and are not replaced by new paying players.

MyNet Games burning the bridge for paying players

The stable revenues experienced in 2022 despite a shrinking player base were driven by MyNet Games transitioning SIFAS towards a pay-to-win (P2W) strategy. Their actions, particularly during the second half of 2022, heavily supported whales (players who spend a lot of money on the game) at the expense of free-to-play (F2P) players and low and middle spenders. These included:

Making limited resources such as High Memories and Shikishi Boards unlimited by selling them as paid packs. This allowed whales who were willing to spend to purchase them to quickly boost their cards and idols while shutting out F2P, low and middle spenders.
Increasing the tap cap from 50,000 to 100,000 per tap in Advanced songs. This destroyed the meta in the Advanced difficulty as song requirements were trivialised and voltage ranking these songs heavily favoured whales who had strong cards. The change also reinforced the existing meta, placing SP filler cards above all other cards due to their ability to reach the tap cap often.
The increased tap cap led to continual powercreep over the last months which made older cards increasingly redundant. This pressured players to get new cards, making it difficult to keep up with the meta without investing a significant amount of money. These effects increased player burnout and fatigue towards SIFAS.
Combining actions 2 and 3, there was a sudden change to the meta of one event (Super Big Live (SBL)) which drastically marginalised a card that was released just beforehand (found below). That card was useful for voltage ranking in SBL under the old tap cap of 50,000 per tap. After the SBL and the gacha to get that card ended, MyNet Games released a notice saying that the tap cap for SBL would be raised from 50k to 100k. This relegated the card from an essential card in every SBL song to a niche role of scoring well in one attribute and swapping subunits in certain songs to maximise the score. This change upset plenty of meta players who have obtained that card in an effort to compete in the old SBL meta. This was evidenced by the fact that four JP whales who used to compete in SBL voltage ranking left the game altogether.

Details of the card, with stats and abilities, that was referenced in point 4. — Details of the card that is reference in point 4 of the above text.

Altogether, MyNet Games implemented these changes in an effort to extract money from the whales while the game was going well, maintaining steady revenues. This came at the expense of new players who were not retained, as well as existing free-to-play and paying players who left the game due to the burnout they experienced in SIFAS. The shrinking of the player base not only precluded increases in revenues, but also resulted in a slow but continual decline in revenue.

This is in contrast to Genshin Impact which not only maintained high revenues throughout 2022, but have also promoted practices that are friendly to both casual and meta players:

Hoyoverse’s commitment to not release end-game content for Genshin Impact, while controversial, actually positioned it as a casual game to be enjoyed at its own pace instead of being a competitive game like Tower of Fantasy. Players do not feel pressured to get the latest character in an effort to keep up with the meta. This not only reduces stress towards the game, but players can still enjoy playing their favourite characters without it being powercrept.
The introduction of the dendro element in Version 3.0, along with its associated elemental reactions, improved the gameplay of existing characters. For example, electro characters such as Keqing now have an elemental reaction (aggravate) that increased their damage output compared to previous versions.
Hoyoverse also gives players time to decide whether to pull a new character. They not only drip-market new characters before they are released, but they also give players the opportunity to trial new characters before pulling for them. This allows players to decide whether the character fits their playstyle before pulling for the character in the gacha.

Hoyoverse’s practices in running Genshin Impact have allowed them to generate high revenues without resorting to P2W practices or alienating any part of their player base. Their practices are better than MyNet Games which have placed all their chips on extracting money from the whales at the expense of other players before SIFAS’ eventual decline and termination.

Conclusion

I provided a balanced picture on the situation in SIFAS in 2022. The stabilisation of revenues in 2022 was good news for SIFAS. However, this is contrasted by a declining player base, where the meta players leaving the game are not being replaced by new players entering the game. This motivated MyNet Games to move towards P2W tactics which heavily favoured whales at the expense of other players. The result was short-term gain for long-term pain for all players in terms of revenues and player counts. The next blog post explains the termination of SIFAS, and the reasons why it was terminated.

The decline of SIFAS (Part 3): how MyNet Games (and Bushiroad) will put the final nail in SIFAS’ coffin

In the previous two parts of “The Decline of SIFAS” series, I explained how Love Live! School Idol Festival ALL STARS (SIFAS) has become unpopular and why, in the face of its financial loss, KLab had to sell off SIFAS to MyNet Games. Now, we delve deeper into MyNet Games, and how its management and performance of previous games might have affected SIFAS’ future prospects. I did a lot of research into MyNet Games which included gathering and machine translating lots of Japanese documents, articles and websites and tracking the performance of MyNet Games’ currently operating and terminated games. From my research, I argue that SIFAS’ transfer to MyNet Games and Bushiroad does not bode well for its future. This post will provide background information on MyNet Games and game development before describing how the work of MyNet Games and Bushiroad, the developers and publishers of SIFAS respectively, might influence SIFAS’ prospects.

What is MyNet Games?

MyNet is a Japanese company that does a variety of things in the digital transformation (DX) sector. Its core business is in game services via its subsidiary MyNet Games. MyNet Games does not currently develop its own games. Rather, it acquires and manages existing mobile games from other developers, with the aim of having them last for a total of 10 years (in both the original developer and MyNet Games). MyNet Games is a main driver of MyNet’s diversification into other businesses which include:

An AI and marketing consulting firm to help businesses in DX;
Management of two Japanese sports teams: the soccer team FC Ryukyu and the basketball team Shiga Lakestars via data analytics and DX; and
Establishment of fantasy sports in Japanese professional basketball and baseball.

MyNet Games claims that they have a large number of operating professionals in their team, allowing them to accept any mobile game regardless of genre or platform. Consequently, as of January 2022, MyNet Games currently operates 21 games simultaneously, including SIFAS, plus other games that have not been disclosed publicly. There are two ways that MyNet Games acquires mobile games:

Purchase: MyNet Games purchases an existing game from a game developer and wholly takes over the game’s operation, either with its own staff or with the original development staff. A developer may want to sell the game to MyNet Games to:
- Sell the company to MyNet (For example, C&M Games sold its games to MyNet Games by placing them in its spin-off company Crooz which they subsequently sold);
- Exit the mobile game sector to focus on other businesses; or
- Raise funds or free up staff to develop new mobile games.
Collaboration: MyNet Games operates the game with a publisher and shares the profits. This is the case with SIFAS, where Bushiroad has taken over publishing duties while MyNet Games takes over development and operating duties.

After acquiring the game, there are three main ways that MyNet Games manages games to boost and maintain revenues. The first way is data analytics. By acquiring a total of 80 mobile games, MyNet Games claims that they have built a database and a range of indicators and parameters to optimise the user experience and raise the profit margin of games. The second way is increasing sales by redesigning how games are managed and applying sales knowledge to market the game. The last way is cost reduction, where they reduce costs as much as possible by sharing illustrations, assets and users; outsourcing work; automating processes and merging servers.

The business practices and management of mobile games by MyNet Games does not bode well for the quality, profitability and survival of SIFAS. However, Bushiroad, as the publisher of SIFAS, may also contribute to SIFAS’ further decline, and has the final say on whether SIFAS gets terminated. To understand what can go wrong in how SIFAS is currently managed, we must first understand the difference between game developers and publishers, and how either (or both) of them can contribute to a bad game.

Differentiating game publishers and game developers

List of publisher and developer in SIFAS Japan and Global. — The publisher and developer of SIFAS after the changeover in January 2022

A game developer is a person, group or organisation that designs and creates video games. Game developers differ in size, ranging from one person who creates the whole game (for example, Stardew Valley) to teams or even companies of people that work on different parts of a video game. In contrast, a game publisher is a company that markets video games to consumers. They have the marketing and public relations expertise to promote the game to consumers and networks to distribute the game globally. Game publishers often provide funding to game developers in exchange for having a say on the game’s development. For example, they can provide input on what to include or exclude in a game, set sales targets to reach or propose timelines on when the game needs to be released.

Often, problems with or negative reception towards a video game can be the main fault of the developer, the publisher or both. Here are some examples of how developers and/or publishers have contributed to some bad games in 2021:

Fault of developer: Balan Wonderworld, a platform game whose development was led by Sonic the Hedgehog creator Yuji Naka. The game did not sell well, with revenues falling well below expectations, and received unfavourable reviews from everyone. Criticisms of the game include clunky controls and gameplay, bloated, outdated graphics and an uninspiring, confusing story. The fault lies with the developers as they brought forward flawed ideas to a game that cannot be fixed by the publisher.
Fault of publisher: Battlefield 2042, a first-person shooter that was negatively received by fans as the game was very buggy at launch and dropped a lot of features from previous Battlefield games. This is because DICE, the developers of the game, only had 15 months to develop the game, a very short timeframe for AAA game development. The fault lies with EA, the publishers of the game, as their actions adversely affected Battlefield 2042’s development. This included letting experienced Battlefield developers leave DICE, suddenly pivoting to an entirely different genre in the middle of the game’s development and not delaying the game long enough for DICE to get the game to a playable state.
Fault of developer and publisher: Grand Theft Auto: The Trilogy – The Definitive Edition, which has received backlash from both critics and fans for its numerous bugs, terrible character models, framerate issues and lazy rendering. The game is the fault of both the publisher Rockstar Games for rushing development of the game to make money and developer Grove Street Games for remastering the games from existing mobile ports instead of remaking them from scratch.

In the context of SIFAS, although their actions are being heavily scrutinised, it is not enough to focus solely on how MyNet Games will further fuel SIFAS’ decline. Although I will be explaining them at length in this post, we must also consider how Bushiroad, the game publisher, can exacerbate SIFAS’ problems. I will explain their role later in the post.

How do games operated by MyNet Games fare?

Reading comments from the SIFAS subreddit, some people believe that MyNet Games has garnered a bad reputation for being the graveyard of mobile games. Receiving games from other developers, MyNet Games can shut down unprofitable games. Seeing how SIFAS has been transferred to MyNet Games, I decided to investigate how long games transferred to MyNet Games last and measure the chances that a game gets shut down.

I first compiled a list of all mobile games that MyNet Games has operated. For each game, I noted down the dates that the game was released under its original developer, was transferred to MyNet Games and, if applicable, shut down. I then used the dates to calculate the number of months that each game was operating in total and under its original developer and MyNet Games. The amount of time each game was operated by MyNet Games was tracked up to January 2022 (for a currently operating game) or its termination date (for a discontinued game). Afterwards, between currently operating and discontinued games, I compared the number of months that games operated in total and under the original developer and MyNet Games. I also tested whether differences in operation periods were statistically significant (at p < 0.05) via a Mann-Whitney U test. Statistical differences in operation periods mean that these differences did not arise through chance but usually for specific reasons.

Looking at the number of months that each game was operating, I found the following results.

Boxplot of total operation time between discontinued and continuing games. — Total operation time between discontinued and continuing games

First, looking at the total amount of time that each game was operating, games that were discontinued operated for significantly less time (by a median of 30 months) compared to games that are still operating (p = 0.027).

Boxplot of operation time under the original developer between discontinued and continuing games. — Operation time under the original developer’s operation between discontinued and continuing games

This was not due to the original developers, as the operation time under their original developers were similar between currently operating and discontinued games (p = 0.182).

Boxplot of operation time with MyNet Games between discontinued and continuing games. — Operation time under MyNet Games’ operation between discontinued and continuing games

Rather, the lower operation time among discontinued games may most likely be due to MyNet Games. When looking at the periods of time that games were operated by MyNet Games, discontinued games spent less time in MyNet Games (by a median of 14 months) compared to currently operating games. Although the difference is not statistically significant (p = 0.094), the result seems to suggest that MyNet Games played some role in determining how long a mobile game lasts.

I further investigated the operating periods of games by sorting currently operating and discontinued games into different categories. First, I sorted each game based on whether they lasted less or more than 2 years with MyNet Games. I chose 2 years as this is the benchmark set by MyNet in their financial presentations.

Table showing the number of games currently operating or discontinued under a MyNet Games operation period of under or over 2 years.

The number and proportion of games that were currently operating or discontinued (as of January 2022) between games that operated less or more than two years under MyNet Games

Looking at the contingency tables, the most striking result was when looking at the termination rates under different operating periods. Within the first two years of MyNet Games’ operation, the termination rate is 85% which is very high. This is reduced to 53% when the game is operated by MyNet Games for more than two years. This result indicates that even though there is a high chance of MyNet Games terminating a game, the risk is highest in the first two years of operation. This is particularly true for games that were unprofitable or not promising which may explain MyNet Games’ willingness to quickly shut them down.

Just as interesting is how long SIFAS would last under MyNet Games’ operation. To predict this, I sorted currently operating and discontinued games based on how long each game was run by its original developer. I chose 2 years 3 months as the benchmark as this was the amount of time that KLab managed SIFAS.

Table showing the number of games currently operating or discontinued under SIFAS' original operation time of under or over 2 years 3 months.

The number and proportion of games that were currently operating or discontinued (as of January 2022) between games that operated less or more than two years three months under the original developer

Out of all the games acquired by MyNet Games, games that have lasted as long as or shorter than SIFAS under its original developer had about a 16% higher chance of being shut down, compared to games that lasted longer with their original developer. This result shows that SIFAS has a slightly higher risk of being shut down while being operated by MyNet Games.

Taken together, the data suggests that MyNet Games may have an influence on which games continue for a long time, and which ones get shut down. These results fit with their objective of having games last for a total of 10 years. To sustainably achieve this objective, they need to shut down bad games quickly and use their resources to promote good games. This may explain how MyNet Games has garnered a bad reputation for being the graveyard of mobile games. This might be worrying for SIFAS as they have a slightly higher risk of being shut down, but that is only one part of the story.

Tracking revenues of games managed by MyNet Games

Other people say that for each game acquired, MyNet Games tries to generate as much profit as possible by either reducing costs or making the games more grindy and pay-to-win, and terminates them as soon as they become unprofitable. To test these claims, I tracked the revenues of two games that were transferred to MyNet Games in 2019 and were still operating:

Flash God Ceremony: Come Light Live (Shinshiki issen kamuraitoraibu, 神式一閃カムライトライブ), an action RPG; and
Future Family Tree Tsugu-me (Mirai-ka keizu tsugu me, 未来家系図つぐme), a farming simulator.

With revenue data available from 2018 to 2021, we can see how monthly revenues have fared before and after the transfer to MyNet Games.

Line graph of revenues for the Flash God Ceremony video game. — Monthly revenues of the Flash God Ceremony game

Line graph of revenues for the Future Family Tree video game. — Monthly revenues of the Future Family Tree game

Looking at both games before the transfer, we can see that revenues fell from their peaks in 2018. This was particularly true for Future Family Tree which saw a precipitous decline in revenues throughout 2018. Consequently, both games were transferred to MyNet Games: Future Family Tree in April 2019 and Flash God Ceremony in June 2019.

The transfer to MyNet Games affected the revenues of both games differently. Flash God Ceremony continued to show declines in revenue even after the transfer, reaching a low of 5.9 million Japanese yen in October 2021 before bouncing back to 38.9 million Japanese yen in December 2021. In contrast, Future Family Tree saw an increasing trend in revenues after the transfer. From a low of 3.6 million Japanese yen in April 2019, revenues increased to 33.2 million Japanese yen by December 2021.

Let’s now look at a game where revenues have stagnated after MyNet Games took over operation duties: Yakuza Online (Ryū ga gotoku ONLINE, 龍が如く ONLINE). Yakuza Online is a card-collecting RPG that recounts the events of the Yakuza series, but starring Kasuga Ichiban, the main character of Yakuza: Like a Dragon. Sega initially developed and operated the game which was released on 21^st November 2018. However, the game was co-operated by Sega and MyNet Games from March 2021 due to Sega’s inexperience in operating mobile games for the Yakuza series.

Line graph of revenues for the Yakuza Online video game. — Monthly revenues of the Yakuza Online game

Before the co-operation period, Yakuza Online showed unstable revenues. Not only is there a small decreasing trend in revenues, the game also did not experience a bump in revenues during its anniversary periods in November 2019 and 2020, a warning sign that the game is not doing well. Revenues stabilised after co-operation with MyNet Games but otherwise did not grow. In fact, the game experienced a dip in revenues to 54.9 million Japanese yen in October 2021, only recovering to 90.6 million Japanese yen by December 2021.

Collectively, these results show that the transfer to MyNet Games can have different effects on game revenue. While some games can experience increases in revenue after the transfer, other games can show stagnation or even declines in revenue. Most likely, SIFAS will follow Yakuza Online’s path, stagnating or even experiencing declines in revenue. This is because Yakuza Online and SIFAS are based on popular IPs, allowing them to maintain a player base. However, they did not perform well under their original developers, not generating an ‘anniversary bump’ in their revenues. This does not bode well for SIFAS which desperately needs to improve and innovate to stay afloat in a competitive mobile gaming industry. MyNet Games will not provide that platform for SIFAS to succeed.

The implications of transferring SIFAS to MyNet Games

MyNet Games is, first and foremost, a game services company, not a game developer. It acquires as many games as possible to fuel MyNet’s growth, at the cost of maintaining the game’s quality. This does not bode well for SIFAS’ growth, quality and survival. Here are three key problems of MyNet Games’ game management and how they relate to SIFAS.

More limited resources and staffing

MyNet employs less people than KLab, with only a subset of them working on mobile games. As of September 2021, MyNet only employed 387 staff in total, spread out across game services, sports management, fantasy sports and consulting. This is in contrast to KLab which had 654 employees in September 2021, with 526 of them working on mobile games. This means only a subset of MyNet Games’ employees are actually working on mobile games. Worse, these staff members are simultaneously working on at least 21 mobile games over various genres, with vastly different coding, gameplay and monetisation systems. In contrast, KLab operates a smaller number of games over set genres, each using similar systems.

Even more concerning is that in its December 2020 financial presentation, in an effort to cut costs, MyNet announced that they will reduce the amount of work done by full-time employees and transfer their work to part-time and outsourced workers. This thinly spreads out game operations to many people, many of whom will have less stable jobs, have to work on multiple games and are less experienced in game management.

All of these problems are detrimental to SIFAS. First, under MyNet Games, there are less resources and staff to operate and improve SIFAS, both in terms of fixing SIFAS’ problems and innovating to keep players stimulated. Second, as staff are working on multiple games, less time will be spent maintaining SIFAS. Any bugs or issues in SIFAS will take longer to identify and fix as staff will be preoccupied by other games.

MyNet Games are less familiar with SIFAS

Given the complex game mechanics of SIFAS, it is imperative that the new development team in MyNet Games are sufficiently briefed on how the game works. Although we do not know who is currently working on SIFAS (as the staff list was not updated after the transfer), it is likely that not many people have transferred from KLab to MyNet Games to keep working on SIFAS, leading to new staff being recruited to operate and manage the game. Despite claims that MyNet Games has a large number of operating personnel to accept any game regardless of genre or platform, the staff would be less familiar with SIFAS’ codebase and gameplay. This has resulted in two problems that cropped up in SIFAS.

First is based on the staff’s unfamiliarity with SIFAS’ codebase. Given that MyNet Games staff were not the original developers of SIFAS, they may introduce bits of code that inadvertently alter the game or introduce new bugs. For instance, in one of the recent School Idol Channels, the song for the Voltage Rally changed from Torikoriko PLEASE!! to Mirai no Bokura wa Shitteruyo, the song from two weeks ago. What might have happened is that a MyNet Games staff member accidentally changed the code for the Voltage Rally song, shifting the Voltage Rally back by two weeks. This example not only shows how unfamiliar MyNet Game staff were with SIFAS’ codebase, but the reduced resources and staff time towards SIFAS meant that the bug was not fixed in time before the School Idol Channel ended. This highlights that it would take longer for MyNet Games staff to identify and fix SIFAS’ bugs and problems which would demotivate players from playing the game.

Second is based on the staff’s unfamiliarity with SIFAS’ gameplay. MyNet Games staff have little to no understanding of how the game works due to less staff time being dedicated to understanding the game. Hence, they would be less familiar with SIFAS’ gameplay (which is vastly different from other rhythm games) or its meta (i.e., which cards are powerful or not powerful).

Comparison of stats and abilities between two groups of event cards.

For instance, in the Year of the Tiger Unit, Assemble! marathon event in January 2022, all three URs, including the event UR, had the exact same attribute, type and skill set as their respective UR cards in the Pure White event (March 2021 in Japan, April 2021 Worldwide). The only differences were in their stats and their art work. By copying UR cards from a previous event, it shows a lack of understanding from MyNet Games of SIFAS’ meta. This makes it likely that they will not keep the meta balanced, either introducing underpowered cards that have no impact on the meta or overpowered cards that can disrupt the meta (and making players pay to get them).

Given the terrible working conditions in MyNet Games, it is not surprising that no KLab staff would want to transfer to MyNet Games to keep working on SIFAS. Even if KLab’s financial situation is dire, they would rather stay in KLab to work on a new game than move to MyNet Games to work on something whose future is uncertain.

An analytics approach to game management cannot solve everything

A data-driven approach to game management cannot solve everything as it neglects the inherent flaws of SIFAS and the reasons why the game is failing. MyNet Games uses analytics and AI to manage its games, analysing different kinds of data such as user loyalty data and game costs based on the performance of past games. It then uses the results of these analyses to take different actions on various games, from marketing promising titles to reducing costs to boost profit margins or even terminating unprofitable games. This approach, based on past performance of previous games, forces MyNet Games to adopt the same strategies to manage different games which stifles innovation. It also ignores the unpredictable behaviour of humans and the unique gameplay elements of each game. This can be seen by the fact that after the transfer to MyNet Games, some games become tougher to play with more pay-to-win elements. These actions are in an effort to quickly milk as much revenue as possible before it falls, leading to their discontinuation.

Combined with the reduced time budget to manage each game, MyNet Games will not have sufficient time to understand and fix the problems of each game to improve quality and restore profitability. In the case of SIFAS, this is aggravated by KLab being less invested towards SIFAS as they believe that it is on the decline. This will be worsened by MyNet Games as they will have less time to understand the problems of SIFAS and make meaningful improvements on the game. Hence, SIFAS is under the mercy of MyNet Games’ ruthless operation and management procedures which is not adapted to the unique elements of SIFAS.

Explaining Bushiroad’s role in SIFAS’ further decline

The saving grace of SIFAS is that it is being managed by Bushiroad. As the publishers of SIFAS, Bushiroad can direct MyNet Games to improve certain parts of SIFAS, set up promotional campaigns (such as allowing SIFAS players to enter a lottery for online viewing tickets of Niji’s 4^th live if they clear L!L!L! in chapter 30, hard mode), and has the final say on whether SIFAS continues or is terminated. However, Bushiroad has been laissez faire on SIFAS while KLab was still running the game. While Bushiroad was not the publisher at the time, it did recruit a producer to oversee SIFAS. As one of the gaming partners of SIFAS, it should at least intervene to get SIFAS back on track. This did not happen when season 2 of SIFAS’ story went out of control, where Bushiroad did not intervene in time to defuse the situation. If Bushiroad and Sunrise had foreseen this and forced the writers to rewrite the story, the fate of SIFAS might have been different.

With publishing duties handed over to Bushiroad, they now have more power to decide where SIFAS should go, and how it should be promoted. It is unclear how responsible Bushiroad will be to keep SIFAS running smoothly. For instance, Bushiroad decided to speed up BanG Dream! Girls Band Party on the Global server to catch up to Japan. This has caused a lot of consternation among players as events are out-of-sync with their respective holidays and less time was available for staff to prepare and implement updates and fixes, ironically causing delays. This highlights how problems in games can be not just the fault of the developer but also the publisher.

Additionally, SIFAS is at the mercy of Bushiroad who can decide whether to continue the game or terminate it. This is not helped by the announcement and release of Love Live! School idol festival 2 MIRACLE LIVE! (SIF2) which introduced a lot of uncertainty on whether SIFAS can keep running in the future. As soon as the game was announced, Bushiroad poured all their efforts in promoting the game while neglecting SIFAS. Consequently, SIFAS’ popularity and player base dropped as players eagerly await for the new Love Live game, putting it in a further downward spiral that may lead to its termination.

Hence, just as much as SIFAS’ further decline can be driven by MyNet Games’ actions, we must also consider how Bushiroad may contribute to the decline.

Conclusion

The transfer of SIFAS from KLab to MyNet Games is regrettable given that KLab had given up on SIFAS towards the end of them operating the game (and the fact they have resorted to NFTs to recoup their losses instead of fixing how they develop and operate their games). It is reasonable that, with more time and effort poured into it, SIFAS would bounce back. It is also reasonable that, should it be transferred, SIFAS can be given to a developer that is passionate about Love Live and is willing to improve the game to restore its growth and reputation.

Instead, KLab have handed it off to MyNet Games, a game services company that has less resources to operate or improve SIFAS. As highlighted in the post, SIFAS was at the mercy of MyNet Games’ management which is known for spreading its staff thinly across many games and for ruthlessly terminating games it does not see as profitable. Bushiroad might also aggravate SIFAS’ decline, given how unpredictable they will be to keep the game operating smoothly or promote the game in the face of a new Love Live game. Hence, at best, SIFAS would survive, albeit with no prospects for future growth. At worst, SIFAS would be shut down, to make way for new games that are more faithful to Love Live or have better gameplay than SIFAS. That is when the final nail in SIFAS’ coffin will be hammered.

References

MyNet. (2021). Financial Results Briefing Materials: 3^rd quarter of the fiscal year ending December 2021. MyNet Co. Ltd., Tokyo. https://pdf.irpocket.com/C3928/gK0G/VKkO/O6Xm.pdf.
https://gamebiz.jp and https://www.4gamer.net were used as the main Japanese websites for collecting dates on the launch, transfer to MyNet Games and termination of games operated by MyNet Games.

A brief overview of QCA

Pathway 1 to a happy end

Tohru

Kyo

Pathway 2 to the outcome

Yuki

Machi

Pathway 3 to the outcome

Kisa

Ritsu

What leads to a sad ending?

Akito

Isuzu

Conclusion

AI statement

Share this blog post to your social networks:

Like this:

What are the aims of I CAN Network’s programs?

Why I have not used mean scores in the report

Why I have used 100% stacked bar charts in the report

Visualising mentee demographics using different graphs

Conclusion

Personal disclaimer

Share this blog post to your social networks:

Like this:

A final update on player counts

A final update on revenues

Comparing SIFAS’ shutdown to other games managed by MyNet Games

How Bushiroad killed SIFAS

The irony of shutting down SIFAS

Conclusion

Share this blog post to your social networks:

Like this:

SIFAS player counts in 2022

SIFAS revenues in 2022

MyNet Games burning the bridge for paying players

Conclusion

Share this blog post to your social networks:

Like this:

What is MyNet Games?

Differentiating game publishers and game developers

How do games operated by MyNet Games fare?

Tracking revenues of games managed by MyNet Games

The implications of transferring SIFAS to MyNet Games

More limited resources and staffing

MyNet Games are less familiar with SIFAS

An analytics approach to game management cannot solve everything

Explaining Bushiroad’s role in SIFAS’ further decline

Conclusion

References

Share this blog post to your social networks:

Like this: