Preferences of Otome Gamers // A survey of Japanese players

Like most things I write here, I had a random curiosity and then used my free time to see if there had been any research done into my random curiosity.

A while ago, I was curious about the thought processes that went into designing an otome game the way they are and wanted to know if there was any basis for the heroines’ personality written into games since quite a few are hit and miss personally. I know in the wider Western otome game fandom, this is also a recurring topic of discussion, but I wanted to know about Japanese games since that makes a bulk of the games I play.

I came across some presentation material given at a game conference some years ago which had some detail on writing otome games which got me thinking about how this came to be and what the current otome game climate is like, since the material was dated.


When I researched this topic I found something far more informative than what I was looking for.

Nakamura Sachiyo* had conducted some surveys of otome gamers in Japan to assess the current consumer base on both Twitter and Google surveys last year. She had initially done some presenting and explanation of the results with her background in game developing, which I will translate (prefaced by Nakamura said/stated etc.) and further assess by comparing the Twitter and Google survey results which she had not done. From the data I will attempt to draw any meaningful information while talking about other explanations too.

Interesting survey design choice…((it’s not perfect))

Please keep in mind that market analysis and whatnot is very foreign to my actual academic background (science) so if I make mistakes, let me know/take the explanation with a grain of salt. Also the resulting data is kind of not great in many cases.

I have rearranged the question presentation to answer what I wanted to know the most.


Data acquisition

All the data presented here are from responses Nakamura presented on her blogs [1], [2] and from her twitter polls [thread].

Some of the data from the Google survey was collated into less options due to capitalisations and alternative spellings being split off into separate groups. Data that was unidentifiable from the original posts due to its small value were disregarded.

Usually for these types of surveys, I’ve learnt that there is usually something at the beginning called a screening question. These questions are designed to ensure you’re speaking to the correct target audience, for example, to make sure you’re talking about otome games to people who are actually interested in otome games. These questions are usually included upfront, so you can ‘screen out’ people who aren’t interested in the topic (they won’t get asked the main survey questions), so they won’t dilute your actual responses.

Nakamura included a question right at the end of her survey for fun which was “do you like otome games?”. Judging by the fact that some people (1.2%,~20 people),  answered various versions of “no”, I’ll use this percentage to adjust the resulting data for higher accuracy. The resulting amount of respondents for the google surveys were 1627 and 1635 for post 1 and 2 respectively after adjustment. Amount of respondents for twitter polls will be listed in relevant section. All respondents are assumed to be Japanese consumers of otome games .

Otome gamer demographics

Age breakdown

Otome gamer age breakdown

The 40s, 50s age brackets were combined with the 36 to 39 age bracket due to small individual sample sizes. This value, along with others could have been skewed due to the method the survey was delivered (via social media, only to her following) which could exclude a wide range of people.

From the survey, almost 60% of all respondents were in their 20s. This should come as no surprise as this is generally the age group that starts earning the most money with the most free time.

Personally I think its surprising that there is a high amount of respondents that are also 30+ and still playing otome games, forming ~30% of the remainder of the respondents.

Nakamura notes that the low percentage of teens (10 to 19 year olds) playing otome games may relate to the prices of consoles being out of reach for them.

How do otome gamers enjoy their otome games?

This was the main thing I wanted to find out; How people play their games; self-inserting or as an outside reader.

This question was asked on both Twitter and on the survey.

If we group the self-inserters and those who can “become the heroine” when playing an otome game, they only form less than 25% of all the players from both polls. This is surprising given the roots of otome games. Nakamura notes that when she began working in the otome game industry, the amount of self-inserters was higher.

As times progressed, it appears now there are more players interested in enjoying otome games as a 3rd party (52% in the Twitter poll and 68% in the survey result). I’m not quite sure of the differences between overseeing the happenings in the game (見守る) and enjoying the game in a voyeuristic way (第三者視点, sorry couldn’t think of a better way to say it lol), but ultimately I think they’re similar in that the player doesn’t think of the heroine as themselves (self-inserting).

Previously there had been an article posted on the Japanese-sphere about how age affects the ability of players to self-insert, with this particular example. In this article the writer states that she could no longer self-insert after growing older than the heroine and enjoyed the games in a way where she served as a guardian to the heroine and cheered for her. This aligns with the current results as 90% of the respondents were older than 19 and most packaged games feature heroines within that age range. There are a few exceptions where the characters are slightly older and in the 20s range but I have yet to come across any packaged game or more than 1 mobile game with a heroine above 30. This could explain why there are more self-inserters than the teen age respondents. However its also possible that some few players are able to self-insert regardless of age.

Where do otome gamers get their info about games from?

This is a bit of a miscellaneous question but I included it anyway for the sake of completion.

Its clear that in the age of the internet, information travels so fast that magazines and other paper publications have been rendered irrelevant. This result makes it very clear why Dengeki Girl’s Style ended its publication and that B’s Log is likely to follow suit. In order to stay relevant, magazines will need to offer some sort of content that is unobtainable online. Personally I enjoy short stories and comics related to upcoming games but this is probably economically unfeasible.

However its possible that this poll was also influenced by the fact that it was held online which automatically selects for users that use the internet to begin with.

Hardware survey

Moving on to the hardware that otome gamers want to play games on. The survey was conducted before the switch lite was announced which is reflected in the numbers. I’d be interested to see how that may have changed now if possible though.

What devices do otome gamers own?

Since not all the numbers were listed in the original blog post, I can only use the data that was available.

Devices owned by otome gamers expressed as a percentage of the number of respondents after adjustment.

Unsurprisingly, most people own a smartphone. But surprisingly, slightly more people seem to own PCs than the Vita, the main otome game console in recent years. On average, players own about 3 devices from this data.

The low ownership rate of PS4 amongst otome gamers makes it increasingly obvious why Otomate’s move to the PS4 never took off.

Given the current console otome games are focused on the Switch, it’s interesting to see that there is only 40% ownership rate. From curiosity, I remembered that when the market made the move from the PSP to the Vita, sales for otome games dropped quite dramatically. From my old post, it showed that the average sales for PSP otome games were 14316.8 sales per title, while the Vita had average sales of 5828.5 per title. It shows that there are risks developing for a console that hasn’t established a strong owner base.  Perhaps this is the trend that was being referenced by Idea Factory director Satou Yoshiaki. If I had data for the PSP move to Vita as well, I could make more substantial claims about the console moves costing 60% of the consumer base but this preliminary finding is interesting.

What device do otome gamers want to play otome games on?

Otome gamers preferential devices for otome gaming expressed as a percentage of the total respondents after adjustment.

From this question, you can see that the preference still lies heavily on the Vita despite its end of production. While Switch also has second preference, it is split evenly with the smartphone. From this you can assume there is some amount of otome gamers open to moving to the Switch even without owning one, but also that approximately half of the smartphone owners also do not want to play otome games on their smartphone. This may be due to the reputation mobile otome games have as being ‘social games’ that are not self-contained the same way packaged games are.  In the same vein, about half of the PC owners also do not prefer to play on PC. Perhaps these people view their PC as a tool for working rather than relaxation.

Game preferences

These are various elements of a game like the scenario, characters etc which directly affect the enjoyment of a game.

What is most important when picking an otome game?

This was something I was interested in finding out before too so I’m glad the question was included.

Important factors for picking an otome game

Twitter polls only allow for one choice, which is why the results look so much more polarising, while the online survey also introduced many other factors.

Among the shared options across both surveys which otome gamers place importance on, both polls showed the highest importance on writing and lowest on voice actor. Character and art showed mid-tier importance. Theme of the game ties in very importantly writing as well since it is the initial idea of the plot that consumers will see and grab their attention. But without a good writer, good themes can very well flop as well…..stares at a certain game in my blog history.

Strangely enough, the company the game comes from also seems to be of some importance to gamers. Given a majority of scenario writers are freelance, and that there was a high importance placed on writing, I imagined this number would be minor. I suppose some players would associate a certain quality of game with each company.

What kind of scenario do otome gamers prefer?

I think the answers for this question are a bit poorly designed because when someone plays an otome game, they play it with the expectation of romance. Romance in fiction is always accompanied by the so-called heart-pounding events. Otherwise what is the point in reading romance?

Scenario preferences of otome gamers. The numbers for comedic and pure were eyeballed since they had no value attributed to them.

It goes without saying that emotional scenarios are those that probably leave the deepest impression so seeing it come up second is not surprising. The difference between dark and serious scenarios is also a very fine line (to me at least) so this ties in well with emotional.

What strikes me is the low demand for comedic scenarios. I’m not quite sure why this is the case but the only thing that comes to mind may be that comedic romance is more difficult to master as a creator. Perhaps what players consider comedic differ very greatly.

Do otome gamers enjoy perverted content?

Given the large amount of older (of age) respondents, this is both unsurprising and interesting. A resounding 52% would enjoy perverted (in the rated sense) content in their otome games. However backtracking to the console preferences of the same audience, an estimated half of those players are unlikely to play rated games which are only available on PC.

A further 36% appear to be open to the idea depending on the scenario, which totals 88% of the otome game audience which are willing to play games with this content. This makes me wonder why rated games for girls are not more prevalent.

The otome gamers’ ideal number of love interests per game

This result isn’t surprising either assuming you have pure console otome gamers being surveyed. On average, most games have between 5-6 love interests/routes which is reflected here in the average gamers’ expectation.

Nakamura states that when you add more characters into a game, the scenario tends to get thinner. In the end it might depend on the ability of the writer to bring out all the desirable traits of a characters in their route because there have been games were I felt a certain character had gotten more attention than others, but I have also experienced the reverse.

The ideal route length

The options in this question were not very clear. It is unclear whether ‘within 10 hours’ is inclusive of all options before it (thus, meaning anything from 2 hours up until 10 hours) or means ‘between 5 and 10 hours’. I am assessing this question based on the assumption of the latter.

It seems the average gamer expects it to take between 5 to 10 hours to clear a character. Assuming they mean the lower end of the “within 10 hours” scale, I think on average it used to take me about 6 or so hours to clear one character in the average game which fits right there. Honestly once it hits 10 hours I think I’d have zoned out and lost interest.

Nakamura says that it makes sense because people want to spend as long as possible with their favourite char, but personally I think it’d then mean people are suffering through the rest of the chars that they may not like so much.

If a game is well below the average of that 6-7 hour mark per route, I’d expect it to be cheaper but we all know that isn’t always the case….

What price is most accessible to most otome gamers?

It’s really a shame that we can’t access the original data for the survey because it’s hard to say anything about this result without being able to link it to previous answers. Its hard to say if those 26% that can only buy games under 5k yen are cheapskates that expect a full game with the 5-6 chars or a game which only has 1-3 chars which do fall into that price range.

For the 31% that will pay to about 5k yen, this price is equivalent to the download edition of some Vita otome games (Otomate lists its digital Vita games at 5.8k yen). But with the jump to Switch, digital games are now in the mid-6k yen price point. Since only 34% of all surveyed gamers are willing to pay this amount, price of games (on top of console cost) may also be a huge reason why players refuse to change platforms.

How many otome gamers fully complete their games?

Numbers are expressed as percentages of total respondents.

An overwhelming amount of people full comping their otome games is an expected result, after all its the only way to make the most of the money paid for the game.

Nakamura says that as creators, they pour everything into every character so they want players to play through every character. She also says that she has never stanned a character aside from the one she had her eyes on in the first place. But I think it’s more likely for the average player to need to go through every story, route before they can decide on their favourite, which would also require 100% completion of the game.

The heroine

This section is on the character which the player controls, or the heroine of the story.

Should the heroine have a personality?

Number of respondents for Twitter poll was 2359 people. Number of respondents for the google survey was 1627 people. Twitter poll number was revised by screening out all the people who chose “no heroine needed” as that would no longer be an otome game.

A heroine with a personality is a feature that makes it difficult to self-insert into her shoes. Given that only 14% of all otome gamers self-insert, it would make sense for the remainder of the players to want a heroine with a personality. Both the twitter poll and the survey showed a high preference for a heroine with a personality, with a lower preference for a heroine with less presence.

Types of heroine personalities

* These values were eyeballed since the original data was not available. Numbers were estimated to keep the order of preference consistent.

There is a high preference for strong heroines with firm resolve with calm and gentle following.

Nakamura comments that this may be due to the increase of games where heroines ‘counsel’ boys; without a strong heroine with firm resolve, you can’t fix their problems. She adds that, but of course, since the heroine is who your favourite character falls in love with, its natural to want the heroine to have a good personality.

Should other characters say the heroine’s default name out loud?

This refers to the option of having the other characters say the heroine’s default name out loud when the player chooses to use the default name. Interestingly, Nakamura shares that when the option to have the heroine’s default name voiced is given in the case where the heroine’s name is changeable, voice recording for those lines is doubled to reflect the two patterns. Because of that extra recording, the time spent is increased and the debugging work is increased. This is probably the reason many games in the past opted to not have names being said out loud.

The ‘better it is’ option indicates that they don’t really care but prefer her name voiced if given a yes or no situation…so ~70-80% of respondents in both polls prefer the heroine’s name be said out loud.

Should the heroine be voiced?

This refers to whether players want the actual player character herself be voiced.

Players seem to not really care whether she is voiced or not. However with the heroine being not voiced as second most preferred option, it would appear that there is not really an incentive for companies to voice their heroines. As voicing a heroine would likely alienate those self-inserting players while not providing any benefit to those who have no preference, it makes sense that most games on the market retain an unvoiced heroine in this case. However it is also possible that having no preference may be a direct result of the market not having many voiced heroines to begin with.

Reading Nakamura’s post, it seems like she misread the results. However she does make a good point that it increases the budget when you need to voice the heroine so in this case, it probably is not worthwhile pursuing for the a majority of the audience.

Should the heroine be in CGs?

This question asks whether the player character should appear in the stills that appear as rewards for reaching a certain part of the scenario. As with a couple of the other questions, the options are not particularly great which leads to a sub-optimal result. A better way to ask this would’ve been should the heroine appear in a romantic CG or something because the most sensible answer would always be depends if it is necessary for her to be there.

Disregarding the depends option, a majority of otome game players prefer that the heroine be featured in CGs. Since only 14% of players self-insert and are likely to not want the heroine to appear in CGs (which corresponds to the prefer her not be in CGs option), it is unsurprising that the remaining players would like the heroine to appear.

The love interests

This section is about the love interests, the characters that the heroine can eventually be romantically involved with in the otome game.

What is the otome gamer’s favourite male archetype?


  • Yandere: The type that is obsessively in love with someone that it makes them sick in the head. Example: Toma from AMNESIA
  • Tsundere: Hot and cold. Example: Akito from NORN9
  • Mysterious: The type that people never know what they’re truly thinking. Example: Akechi Mitsuhide from Ikemen Sengoku
  • Fleeting: The kind of character that seems like they’ll disappear at any moment, sometimes paired with a weakly constitution. I’m not too familiar with this type I’m afraid. Examples: Mio from Storm Lover (? maybe), Chigasaki Mamoru from Charade Maniacs (according to google at least).
  • Oresama: The type where they are the center of the universe, very often paired with sadistic in otome games. Examples: Oda Nobunaga from Ikemen Sengoku, Kazama Chikage from Hakuouki.

* Values were estimated from the graph to ensure ranking consistency since the number wasn’t available. Removing those options would have skewed the analysis so I opted to keep them in.

This result was an interesting one with otome gamers preferring the yandere archetype and a lower preference for oresama characters. I’m curious whether different age groups prefer different archetypes.

According to Nakamura, in the past oresama type were the most popular, which then tsundere type overtook. But it seems right now yandere type is the number 1.

What age gap with the heroine do otome gamers like?

Do players like love interests that are older or younger than the heroine?

The majority of otome gamers like love interests that are older than the heroine with 51% choosing this option. This seems to just reflect the preferences of the average woman as documented in past academic studies (Kenrick DT, 1992) rather than an otome game-centric thing. Yes, I just cited an academic paper.

Younger love interests are the least popular but as Nakamura puts it, it is difficult for them to gain popularity so as writers, they’re forced to make the younger characters more mature to match the older love interests.

What age is preferable for the entire cast?

This question refers to the age range of the main cast including the heroine.

The highest preference is that the cast be of working age, which is quite a rare setting in packaged otome games. Like Nakamura, I agree that this is likely a reflection of the older age of the respondents. The closer the cast age is to the player, the easier it is to feel more immersed.

Second more preferred age is high school age. Nakamura equates high school with the peak of youth which is something that even older people yearn for. People play games to experience things that they could not in real life so playing games set in high school would give them a feeling of youth that they never experienced. Personally I think this is probably also a reason why many anime and manga are also set in high school.

Nakamura thinks that university age is less popular since in Japan not everyone who finishes high school will continue on to university, so this may not resonate with some people.


Boy that was a lot to take in. But if you couldn’t be bothered reading it all, fret not for here is the general points (its still very long).

The most popular preferences of a Japanese otome gamer from these surveys:

  • Prefers not self-insert but to oversee the happenings of a game
  • Gets information about games from the internet
  • Prefers to game on a Vita
  • Picks their game based on theme and writing
  • Likes an emotional story
  • Is willing to read perverted game contents
  • Wants 5 or so love interests in a game
  • Wants a game to be 5-10 hours long per character
  • Is willing to pay about 5K yen per game
  • Is likely to play an otome game to full completion
  • Wants a heroine with a personality who is strong with firm resolve
  • Prefer other characters to say the heroines name out loud but do not really prefer a voiced heroine
  • Want the heroine to be visible in CGs
  • Like yandere boys the most but likes oresama boys the least
  • Want the love interests to be older than the heroine
  • Want the entire cast to be of working age

How does that look to you? Keep in mind these are the preferences of the Japanese otome gamers surveyed, it may not necessarily align with the average English-speaking otome gamer.


*Nakamura Sachiyo

is an ex-Honeybee employee who is now working as a freelance creative. Her penname was Yuzu Mikan for scenario writing and also does producing and directing of otome games such as Starry Sky, Ayakashi Gohan and Dynamic Chord. She is currently part of Animal Planet, a doujin circle producing otome contents such as Hoshikuzu Heliograph. [Her twitter]


Filed under Otome Game

11 responses to “Preferences of Otome Gamers // A survey of Japanese players

  1. angelrenoir

    Heh. Other than the yandere love interest most of this matches to my preference. Although one thing that I wish had been asked is “prevalence of side characters”. I always liked it when the MC has a good female friend to confide to in the game, particularly when it’s a fantasy/historical setting, and I find that a lot of the time it’s just the MC surrounded by men. Especially in the mobile games. Maybe it’s just because a lot of mobile games can only have so many assets, and it’s a bit awkward having a prominent character not having a face attached, but still…

    • The lack of supporting friend characters is probably mainly a mobile game thing…? Most of the console games I’ve played have a friend character unless the plot called for the heroine to be alone (like if she were a lone wolf/raised super sheltered). That’s just my impression though.

      • angelrenoir

        Ooh I see. So it’s not really an issue across the board. Being that I don’t have consoles I can only have access to PC ports when they have ’em, so I guess my reference is just limited. Thanks!

  2. Dude the sheer amount of research you put into this is insane. Thanks for taking so much time to do this, I’ve been very interested in this same topic for years but never really delved too far into it. It’s so nice to see the love for the vita is still going strong despite the betrayal it went through. Is there a reason why most people preferred yanderes? I personally never understood the appeal, but it seems to be a really popular trope and I’d love to learn why so many people love it.

    • Hmm…with the current quality of data, I don’t think I can make any conclusions on the preference for yanderes. It might be quicker to just ask some people who like yandere to find out their appeal lol. Thanks, I’m glad you enjoyed reading!

  3. Favorite types of characters being yandere and older is surprising to me, since I feel like that’s the least offered type… because these are my favorite types as well!

    Thanks for putting up all the effort in translating and grouping all results like that, it’s pretty interesting.

    • Just to be super clear, the ‘older’ doesn’t specify how much older, it could be 1 year or it could be 10 lol. I’m glad you found this interesting.

  4. This is extremely thorough and well researched. You mentioned that most of this was pulled from preferences of Japanese otome gamers, I know a few indie devs have done polls of English speaking players and I’d be interested in seeing a comparison of the two one day.

    • Unfortunately the quality of this data wasn’t fantastic so there’s still room for improvement and I’ve been informed by an expert that without the right kind of response, the data may not even mean anything hahaha orz. If someone wants to provide me the raw data I wouldn’t be opposed to the idea however 😛 Thanks for reading!

  5. aheadofthenight

    Awesome post btw. Really interesting results. Omg, Misoji wa Tsurai yo…

Leave a Reply

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

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

Google photo

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

Twitter picture

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

Facebook photo

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

Connecting to %s