João Vitor Tomotani¹ & Rodrigo Brincalepe Salvador²
¹ Independent Researcher. São Paulo, SP, Brazil. ² Museum of New Zealand Te Papa Tongarewa. Wellington, New Zealand.
Emails: tjvitor (at) gmail (dot) com; salvador.rodrigo.b (at) gmail (dot) com
The Fate franchise started with the computer game Fate/stay night back in 2004. To put it briefly, it is about mages fighting to obtain the fabled Holy Grail. The mages (a.k.a. Masters) summon heroic spirits (Servants) to aid them in their battle. A bit like Pokémon, but the servants are not kawaii monster mascots, but rather figures from worldwide myths, legends, history, and literature. For instance, Napoleon is a servant who’s a historical figure, while Heracles is a mythological one, and Sherlock Holmes is a literary one.
The franchise has grown quite a lot since its inception, with numerous novels, manga, anime, games, and all sorts of merchandise. Its current main product is Fate/Grand Order (henceforth FGO), a game developed by Delightworks (mobile version, 2014) and Sega AM2 (arcade version, 2018). It is one of the top-grossing mobile games worldwide, with most of the money being spent on it by the Japanese public (Chapple, 2020; Orr, 2021).
The popularity of the franchise can have a peculiar effect on the Internet because its characters are almost all taken from real-world material. As Salvador (2020) commented, many historians, archaeologists, and literature scholars will eventually curse Fate when they do a Google search on a given character and are flooded by Fate entries. So, we decided to see how biased by Fate Internet searches truly are.
We hypothesize that the most obscure characters will have more hits about their Fate incarnations than their original ones. Furthermore, those Google hits will appear sooner rather than later in the search. For instance, we expect Astolfo and Bradamante to be such cases, as the stories of Charlemagne’s paladins are not as well-known as, for instance, the (inferior) stories about King Arthur.
Conversely, widely popular characters will have fewer hits about their Fate incarnations and those hits will appear later in the Google search. In this category, we expect characters like Holmes and Moriarty, who have endless movie incarnations.
Our hypothesis is simple and intuitive, but we can’t just take it for granted. We need to test whether this “Astolfo Effect” is real. And so, we did.
Even though we are interested in the Fateverse as a whole, for our purposes we considered only the characters present on its currently most popular and top-grossing entry, FGO. We excluded servants from other games, novels, etc. We compiled a list of FGO characters from the mobile version of the game considering both the American and Japanese servers in their state in June 2021, when the last announced character was Mysterious Idol X.
We used Google Search to look for the name of each character. For those characters that have yet to be released in North America, we used the names as given by the Fate Grand Order Wiki (https://gamepress.gg/grandorder/servant-availability). We used the names as spelled in FGO, which can be a bit off sometimes; we will discuss this matter in more detail later on.
We did the searches on Google Chrome, with clean cache, cookies, and history, using an Incognito tab. A VPN client was used to obtain an IP address from Michigan, USA. We were interested in the Images search rather than the regular website search. SafeSearch was deactivated (we’re talking about Fate, after all). Search of images of any size and any publication date were allowed in the results.
After the results of the search were obtained, we counted the number of images that were about the FGO characters in the first 50 results on Google. Those images could be official artwork, fanart, memes, cosplay, figures, etc., provided it was clearly related to the incarnation of the character in the Fateverse. We also noted the position (1st, 2nd, etc.) of the first Fate-related image to appear.
Of the 308 servants in the game in June 2021, we excluded repeated entries of the same character, such as Alter, Summer, Prototype, etc. versions. We also excluded: original characters, such as best kouhai Mash, for obvious reasons; and servants that were based on real-world stuff but in a way that makes them exclusive to the Fateverse (Hessian Lobo, Kijyo Koyo, and Senji Muramasa). In total, we had 191 servants.
For some of the servants used in our searches, we had to tweak their names a bit. For Zhuge Liang (Lord El-Melloi II) and Sima Yi (Reines), we only used the names of the Chinese strategists, not of the Fate original characters. The same is valid for Ganesha (Jinako). For Henry Jekyll & Hyde, we used simply ‘Jekyll and Hyde’; and for Scathach-Skadi, simply Skadi (because there is already another Scathach servant). We reverted the names of Altria Pendragon and Altera to their actual forms, Arthur Pendragon and Attila; otherwise, only Fate images would pop up in our searches. The same logic applies to First Hassan, Beni-enma, and Miss Crane, for whom we used the names Hassan-i Sabbāh, Shita-kiri Suzume, and Tsuru no Ongaeshi, respectively.
Other potentially problematic names are those with non-standard spellings. Some of these cases are just weird choices, like Brynhildr or Asterios. But the real problem lies with Japanese servants because FGO does not make use of macrons for their romanization, preferring instead the use of two vowels. The unusual spelling resulted almost exclusively in FGO entries in our searches, while the standard spelling resulted almost exclusively in the “proper” historical/mythological character. As such, the following servants were excluded from our discussion below (standard names shown within brackets): Amakusa Shirou (Amakusa Shirō), Ashiya Douman (Ashiya Dōman), Asterios (Asterion or Asterius), Brynhildr (Brunhild), Fuuma Kotarou (Fūma Kotarō), Housouin Inshun (Hōzōin Inshun), Ibaraki-Douji (Ibaraki-dōji), Katou Danzo (Katō Danzō), Minamoto-no-Raikou (Minamoto-no-Yorimitsu), Okita Souji (Okita Sōji), Sakamoto Ryouma (Sakamoto Ryōma), Sasaki Kojirou (Sasaki Kojirō), Shuten-Douji (Shuten-dōji), Tawara Touta (Fujiwara no Hidesato), Ushiwakamaru (Minamoto no Yoshitsune), Xuanzang Sanzang (Tang Sanzang).
A few people were better known by names other than the ones used for the servants. As such, searches using the less-usual version of their names would result almost exclusively in FGO entries. Thus, we also had to exclude Avicebron (a medieval Latinization of his name, Solomon ibn Gabirol) and Nagao Kagetora (his birth name, but he was later known as Uesugi Kenshin). Another case would be Iskandar, which is the eastern/Persian rendition of the name; he’s better known as Alexander the Great or Alexander III. But that one has the benefit of having a kid version in the game with the name Alexander. Yet another case would be Ozymandias, which is the Greek name for Ramesses II. However, we have not excluded him because the name Ozymandias is equally famous in za wārudo, from art (poetry and painting) to art? (Watchmen).
All that clean-up left us with “only” 172 servants: 79 female, 85 male, and 8 servants that FGO classifies as ‘Other’. We used the game’s classification for the sake of simplicity in our statistical analysis (see below), even though we consider it a rather poor and debatable one, as it includes gender-diverse servants, as well the genderless construct Enkidu, and the Dioscuri, which are a brother-sister duo (i.e., one male and one female). This category includes Astolfo, Caenis, Chevalier d’Eon, Dioscuri, Enkidu, Kiichi Hogen, Qin Shi Huang, and Taira no Kagekiyo.
Number of hits
Three servants dominated the searches, with all results out of the first 50 being related to FGO or the Fateverse. They are: Nitocris, Sitonai, and Yu Mei-ren. (For Yu Mei-ren, the game’s spelling with a hyphen doesn’t make a difference, as the result is the same if we use the more common spelling, Yu Meiren.) Yu Mei-ren and Sitonai are obscure characters and don’t seem to have any easily findable representations. Thus, the search results are easily flooded by Fate entries.
Nitocris is a special case because she was neither a real person nor a legendary/literary character. Rather, the idea of a female Pharaoh being the last ruler of the 6th Dynasty was a complete misinterpretation of the archaeological record, as we have explained elsewhere (Salvador, 2020). The actual Pharaoh was a man named Netjerkare Siptah (sometimes spelled as Neitiqerty Siptah). As such, Fate’s Nitocris dominates any Google search.
Next (see Table 1), we have seven servants with 49 out 50 results being Fate-related (the other hit is typically the entry from Wikipedia). They are: Astolfo, Li Shuwen, Mandricardo, Nero Claudius, Osakabehime, Scathach, Yan Qing. All of them are rather obscure characters, though their names might be well-known in their countries of origin. However, there is one glaring exception: Emperor Nero is really (in)famous for being implicated in the Great Fire of Rome. However, that freakish red Saber seems to be even more famous.
Next in line are: Bedivere, Caenis, Diarmuid Ua Duibhne, and Okada Izo with 48 hits; Mori Nagayoshi with 47; Ereshkigal and Oda Nobukatsu with 46. All equally obscure characters with the possible exception of Ereshkigal, whose representations are well-known from archaeological sources. Though confusion of her names and identity with other Mesopotamian goddesses could cloud the search results. Either way, Ereshkigal has a legion of very devoted fans.
Among the other characters with over half the results being related to Fate (Table 1), there are a few that warrant further comment. First, the high number of Fate-related hits for Jekyll & Hyde is due to the use of the ampersand. A search using ‘and’ returns no Fate-related entries whatsoever. Bradamante had 5 non-FGO hits, which is more than we expected. She seems to have been more commonly featured in art than her cousin Astolfo.
Conversely, Kiyohime and Edmond Dantès are characters from very famous stories, so the number of Fate entries in the results is surprising. The lack of recent movie adaptations of The Count of Montecristo (the last one is from 2002) might explain this in the case of Dantès (besides, he is popular and one of the most grailed servants). In the case of Kiyohime, we thought it was a bias due to searching in English. However, when we did a search using a Japanese IP address and her original name 清姫, we got even more Fate-related hits! So, her servant incarnation seems to be quite popular. The same is more or less valid for Tamamo-no-Mae. FGO renders her name as 玉藻の前, and a search using this rendition resulted in a flood of Fate entries. However, her name can be rendered in additional forms in Japanese, which resulted in almost no Fate-related hits in the searches.
Nearly all the servants with no hits are all extremely famous (like Caesar and Cleopatra), have very common names (like David and Gareth), or are simultaneously famous and also a bird (Nightingale). As such, there was no space for Fate-related imagery on the top 50 search results. The exceptions are Nemo, Europa, and Paris, who are not such well-known characters, and Chacha, who is a very obscure figure. Our search for Nemo resulted in us finding plenty of fish. Europa and Paris are also names of places, which complicates the search. The word ‘chacha’ means several other things in others languages and the search results were an assorted mix of random stuff. The historical figure Chacha is better known in English by the names Yodo-dono or Yodogimi; searching for Yodo-dono resulted in two entries from FGO.
And what about Fate’s face? The poster girl Altria Pendragon, the original Seiba. Surprisingly enough, when searching for Arthur Pendragon, 9 of the 50 hits were from Fate, though 8 of them were of the male Prototype counterpart. Though there are many representations of the character in other media, including numerous films and series (such as BBC’s Merlin), this is still a surprisingly high number of Fate-related hits. For the full ranking of servants by search hits, please see the Appendix at the end of this article.
Finally, considering that FGO is blatantly a waifu-bait gacha game, we were curious to see if there was any correlation between our results and the servants’ gender. We tested whether the number of hits differed in relation to the gender of the servant. To that end, we did a Zero-inflated Poisson regression using the package ‘pscl’ (Zileis et al., 2008) and used likelihood-ratio tests to obtain p-values in R (R Core Team, 2020). Servants classified by FGO as ‘Other’ had significantly more hits on average, compared to male and female servants (p<0.01). When only male and female servants were considered, there was no significant difference between their number of hits (p=0.48). The probability of having zero hits was the same for all categories (p=0.99).
We also noted the position in our search results of the first hit that was a Fate-related image. The idea was to find out which characters are more famous for their Fateverse incarnation than their original one. Obviously, for those three servants for which all the 50 results were from Fate (Nitocris, Sitonai, and Yu Mei-ren), the first hit was always the 1st search result. Thus, the discussion below only considers the other servants that had 49 or fewer Fate-related hits.
For thirty-seven characters, the very first search result was from Fate, even superseding entries about their real-world counterparts from sources like Wikipedia. Most results were consistent with our expectations. For instance, Fate’s Astolfo and Bradamante are more famous than their Matter of France’s counterparts. Some of the surprising results from above (number of hits) repeated themselves here, like Dantès, Ereshkigal, Kiyohime, Nero, and Tamamo-no-Mae.
There were, however, a few additional surprises. Lu Bu, for instance, of whom we were expecting a Dynasty Warriors entry to top the search instead. Others were famous mythological characters for whom there are many artworks available, such as Stheno, Semiramis, and the Dioscuri. Though the latter are better known by their individual names, Castor and Pollux, rather than by the term Dioscuri. Sei Shonagon was also surprising, as she is a rather well-known writer in the West.
At the other end, 32 servants were nowhere to be seen in the first 300 search results. All of them were famous people, real or fictional, that have multiple and/or well-known representations in photographs, art, films, series, and other games or manga/anime. Some examples are Robin Hood, Sherlock Holmes, Medusa, Shakespeare. For the full ranking of servants by “first hit”, please see the Appendix in the end of this article.
We also did a statistical analysis to compare the first hit on our searchers. Because in several instances the count stopped at 300 (those 32 servants from the previous paragraph), we did not have the exact value for the first hit for those servants. Thus, we first checked if different genders had a different probability of not appearing after 300 searches, which was not the case (p=0.10). Because of that, we focused only on the servants that we had an exact search value and used Poisson regression to check if the servant’s appearance was related to gender. We saw that servants classified as ‘Other’ have their first hit way earlier than male or female servants (p<0.01), supporting our hypothesis regarding the ‘Astolfo Effect’. Male and female servants, however, did not differ on average in their first hit (p=0.99).
In essence, we confirmed that the ‘Astolfo Effect’ is real. Consequentially, many historians and other scholars will continue to curse Fate and to wonder why their Google search was suddenly flooded by big-eyed anime characters – and, oftentimes, NSFW pics.
However, we once said (Salvador, 2019) that FGO could be a starting point for fans to go after information about History, Mythology, etc., and learn more about the real-world figures related to their favorite servants. We not only stand by that assertion here but also say that the other way around can work as well. Maybe some historians and literary researchers will become interested in FGO and otaku culture through this first contact with a servant on their Google search.
Chapple, C. (2020) Fate/Grand order surpasses $4 billion after becoming Japan’s top grossing mobile game of 2019. Sensor Tower. Available from: https://sensortower.com/blog/fate-grand-order-revenue-4-billion (Date of access: 19/Nov/2021).
Drummond, A. & Sauer, J.D. (2018) Video game loot boxes are psychologically akin to gambling. Nature Human Behaviour 2: 530–532.
Hood, V. (2017) What the UK can learn from the Far East’s battle with loot boxes. Eurogamer. Available from: https://www.eurogamer.net/articles/2017-10-19-what-the-uk-can-learn-from-the-far-easts-battle-with-loot-boxes (Date of access: 19/Nov/2021).
Orr, A. (2021) Top 10 highest-grossing mobile games of all time. Pocket Gamer. Available from: https://www.pocketgamer.biz/feature/77017/top-10-highest-grossing-mobile-games-of-all-time/ (Date of access: 19/Nov/2021).
R Core Team. (2020) R: A Language and Environment for Statistical Computing. R Foundation for Statistical Computing, Vienna.
Salvador, R.B. (2020) Ancient Egyptian royalty in Fate/Grand Order. Journal of Geek Studies 7(2): 131–148.
Wiltshire, A. (2017) Behind the addictive psychology and seductive art of loot boxes. PC Gamer. Available from: https://www.pcgamer.com/behind-the-addictive-psychology-and-seductive-art-of-loot-boxes/ (Date of access: 19/Nov/2021).
Zeileis, A.; Kleiber, C.; Jackman, S. (2008) Regression models for count data in R. Journal of Statistical Software 27(8): 1–25.
We are very grateful to Barbara M. Tomotani (NIOO-KNAW) for the help with the statistical analysis and to Daniel C. Cavallari (FFLCRP) for reviewing our manuscript.
About the authors
João Tomotani, MSc., is an engineer and Rin simp since the 2006 FSN anime, who likes to make huge Excel spreadsheets to study stupid phenomena. He is currently saving for the Ereshkigal banner in February 2022 after failing last year, but he is ready for all the salt.
Dr. Rodrigo Salvador is a curator in the Museum of New Zealand Te Papa Tongarewa. He has always been interested in mythology, ancient history, and literary works, not to mention manga and anime, so he was easily baited by Fate. Actually, no, that’s a lie. He found the original FSN meh. But he came back to the series full force with FGO, as his growing Nendoroid collection can attest.
Number of search hits (# out of 50)
50: Nitocris, Sitonai, Yu Mei-ren.
49: Astolfo, Li Shuwen, Mandricardo, Nero Claudius, Osakabehime, Scathach, Yan Qing.
48: Bedivere, Caenis, Diarmuid Ua Duibhne, Okada Izo.
47: Mori Nagayoshi.
46: Oda Nobukatsu.
43: Fergus mac Róich, Qin Liangyu, Suzuka Gozen.
42: Jing Ke.
40: Edmond Dantès, Mordred.
39: Taira no Kagekiyo.
37: Sakata Kintoki.
35: Watanabe no Tsuna.
34: Henry Jekyll & Hyde.
33: Queen Medb.
31: Kiichi Hogen, Yagyu Munenori
30: Paracelsus von Hohenheim, Sei Shonagon.
27: Chen Gong.
25: Mochizuki Chiyome.
23: Musashibou Benkei.
22: First Hassan.
21: Charles-Henri Sanson, Dioscuri.
20: Lu Bu Fengxian.
19: Atalante, Gawain.
18: Cu Chulainn.
17: Ishtar, Semiramis, Xiang Yu.
15: Abigail Williams, Ashwatthama, Yang Guifei.
13: Anastasia Nikolaevna Romanova.
12: Red Hare.
10: Fionn mac Cumhaill, Jeanne d’Arc, Tomoe Gozen.
9: Altria Pendragon.
8: Prince of Lan Ling.
7: Antonio Salieri, Chevalier d’Eon, Eric Bloodaxe.
6: Astraea, Gilgamesh, James Moriarty, Saito Hajime.
5: Anne Bonny & Mary Read, Darius III, Euryale, Murasaki Shikibu, Oda Nobunaga, Qin Shi Huang, Vritra.
3: Hijikata Toshizo, Kama, Lakshmibai, Saint Martha.
2: Charlotte Corday, Helena Blavatsky, Nezha, Quetzalcoatl, Scathach-Skadi, Sigurd.
1: Arjuna, Asclepius, Boudica, Edward Teach, Hektor, Katsushika Hokusai, Lancelot, Medea, Miyamoto Musashi, Ozymandias, Penthesilea, Siegfried, Sima Yi (Reines), Solomon, Vlad III, William Tell, Zhuge Liang (Lord El-Melloi II).
0: Achilles, Alexander, Altera, Arash, Bartholomew Roberts, Beowulf, Billy the Kid, Calamity Jane, Caligula, Carmilla, Chacha, Charles Babbage, Chiron, Christopher Columbus, Circe, Cleopatra, David, Elizabeth Bathory, Europa, Francis Drake, Frankenstein, Gaius Julius Caesar, Galatea, Ganesha (Jinako), Gareth, Georgios, Geronimo, Gorgon, Hans Christian Andersen, Heracles, Himiko, Ivan the Terrible, Jack the Ripper, Jaguar Warrior, Jason, Leonardo Da Vinci, Leonidas, Marie Antoinette, Mata Hari, Medusa, Mephistopheles, Merlin, Miss Crane, Napoleon, Nemo, Nightingale, Nikola Tesla, Odysseus, Orion, Paris, Parvati, Paul Bunyan, Phantom of the Opera, Queen of Sheba, Rama, Robin Hood, Romulus, Salome, Scheherazade, Sherlock Holmes, Spartacus, Thomas Edison; Tiamat; Tristan; Valkyrie; Van Gogh; William Shakespeare; Wolfgang Amadeus Mozart; Wu Zetian.
Position of 1st Fate-related hit
1st: Astolfo, Bedivere, Bradamante, Caenis, Chen Gong, Diarmuid Ua Duibhne, Dioscuri, Edmond Dantès, Ereshkigal, Fergus mac Róich, Henry Jekyll & Hyde, Jing Ke, Kiichi Hogen, Kiyohime, Li Shuwen, Lu Bu Fengxian, Mandricardo, Mochizuki Chiyome, Mori Nagayoshi, Musashibou Benkei, Nero Claudius, Nitocris, Oda Nobukatsu, Okada Izo, Osakabehime, Qin Liangyu, Sakata Kintoki, Scathach, Sei Shonagon, Semiramis, Sitonai, Stheno, Suzuka Gozen, Taira no Kagekiyo, Tamamo-no-Mae, Watanabe no Tsuna, Xiang Yu, Yagyu Munenori, Yan Qing, Yu Mei-ren.
2nd: Atalante, Ishtar, Mordred, Paracelsus von Hohenheim, Queen Medb, Red Hare.
3rd: Ashwatthama, Fionn mac Cumhaill, Karna, Saito Hajime, Tomoe Gozen.
4th: Charles-Henri Sanson, Enkidu, First Hassan, Gawain.
5th: Eric Bloodaxe, Yang Guifei.
6th: Abigail Williams, Cu Chulainn, Gilgamesh.
9th: Murasaki Shikibu.
10th: Hijikata Toshizo, Vritra.
11th: Qin Shi Huang.
12th: Altria Pendragon.
13th: Anastasia Nikolaevna Romanova, Astraea, Darius III, James Moriarty, Oda Nobunaga.
16th: Prince of Lan Ling, Sigurd.
17th: Chevalier d’Eon.
18th: Penthesilea, Quetzalcoatl, Siegfried.
19th: Antonio Salieri.
20th: Charlotte Corday.
21st: Anne Bonny & Mary Read, Miyamoto Musashi.
22nd: Euryale, Jeanne d’Arc, Saint Martha.
31st: Lakshmibai, Medea, William Tell.
36th: Zhuge Liang (Lord El-Melloi II).
38th: Kama, Ozymandias, Vlad III.
43rd: Helena Blavatsky, Sima Yi (Reines).
47th: Boudica, Katsushika Hokusai.
49th: Edward Teach, Hektor.
60th: Miss Crane.
67th: Bartholomew Roberts, Galatea.
76th: Wu Zetian.
105th: Jaguar Warrior.
112th: Ivan the Terrible.
114th: Wolfgang Amadeus Mozart.
124th: Elizabeth Bathory.
183rd: Phantom of the Opera.
185th: Queen of Sheba.
187th: Jack the Ripper.
193rd: Mata Hari.
201st: Francis Drake.
223rd: Gaius Julius Caesar.
259th: Hans Christian Andersen.
262nd: Charles Babbage.
263rd: Billy the Kid.
300th: Achilles, Alexander, Altera, Beowulf, Calamity Jane, Chiron, Christopher Columbus, Circe, Cleopatra, David, Europa, Frankenstein, Ganesha (Jinako), Gareth, Jason, Leonardo Da Vinci, Marie Antoinette, Medusa, Napoleon, Nemo, Nikola Tesla, Odysseus, Orion, Paris, Paul Bunyan, Robin Hood, Salome, Sherlock Holmes, Thomas Edison, Tristan, Van Gogh, William Shakespeare.
 FGO is based on gacha mechanics. To learn more about the global academic and legal discussions on the evils of gacha, see the articles by Hood (2017), Wiltshire (2017), and Drummond & Sauer (2018).
 This is the opinion of RBS and he owns it. He blames British imperialism for the prevalence of the Round Table over the Paladins. Now, can we have Charlemagne in FGO?
 Sorry, Charlemagne.
 Note that Tsuru no Ongaeshi is actually the name of the tale, not the character. It translates to ‘The crane’s return of a favor’.
 A surprising number of hits (seven) were dakimakura. Actually, maybe that’s not that surprising.
 And constantly disappointed.
 It should probably be best referred to as the ‘Nitocris-Sitonai Effect’. We’re keeping Astolfo in the name, though.
 Orlando is another spelling of Roland, who appears in Astolfo’s interlude in FGO but who we are yet to see as a Servant.