User blog comment:Damesukekun/Song Translating and Mistranslating/@comment-31807596-20170422210256/@comment-24013116-20170423235858

Hm, as far as I can tell, all of these are more issues of localization than actual errors.

1. Chess is probably the closest western equivalent of shogi, and there is vast precedent for replacing foreign cultural items with more familiar analogues when it would carry the same emotional effect without disturbing other parts of the content significantly. This is done to avoid breaking immersion, which could potentially tamper with the reader's impression of the work more harmfully than such replacements.

2. This actually was a choice on the part of the original lyricist. 多重人格, the term used in the song, translates to MPD; DID would be 解離性同一性障害.
 * (If you were aware that you don't know whether or not this was a choice of the original lyricist, you really shouldn't be trying to use it as evidence. In fact, consider the possibility: if you didn't actually do the research, what if there were only one term for the phenomenon in Japanese that didn't directly correspond to either MPD or DID in English? In that case, it would be a pure choice in English, which is entirely unrelated to translation.)

3.  I'm not sure what you mean here, because I just checked for let's plays on Youtube and found at least one (The Anime Man, in 2015, with 50k+ views) for a version of the game that uses 'they'. Even so, there is not really a good solution to handling genderless characters in English. The translator has to make a judgment call: do you just use one of the gendered pronouns to keep the text flowing naturally, knowing that it's not actually correct? or do you use a genderless pronoun that is not really used in common speech for people, or perhaps dance around the grammar to avoid using pronouns at all, knowing that you are making the text unnatural, to the point where you could potentially be alienating the audience in a way that the original work would not have? For both approaches, there are ways to make this better or worse, but I feel the issue is muddy enough that it's hard to consider this strictly an error.
 * (For example, if you default to a gendered pronoun, you could scriptwrite a bit and add a passage to clarify that the gendered pronoun is just shorthand. Whether or not this is good is another judgment call.)

4. The English names evoke the semantic meanings of the Japanese names: 'syake' means salmon, and 'adauchi' means vendetta. This is a classic case of localization - names are not exactly the same, but they are meant to carry the spirit of the original Japanese. This is done because if the readers need a translation, the original names would probably not convey any meaning to them, where they would have conveyed meaning to Japanese readers. Changing the names is another judgment call: you still can't know if the meaning will be conveyed in the same way, but there's a better chance than if you just leave the names in romaji.

5. Does the name appear romanized in the way you claim in any official capacity? If not, this is purely a misguided attempt to parse romaji in your own way - and this seems likely because, if the game is being sold in English on steam, 'Theobald Leonhart' effectively IS the official source. Logically, 'Theobald' is an established name and thus was probably intended; 'Teobaldo' is not a name unless the author specifically wants it to be.

Overall, it seems like you have made these criticisms with insufficient investigation and understanding of the decision-making that went into these translation choices. I'm not saying I agree with all of her choices (in particular, I really dislike things like point #1), but whether or not one agrees with the particular decisions or even the general localization practices employed, I feel it's at worst dishonest and at best misinformed to cite these examples as actual errors. I don't dispute the original evaluation of vgperson's work, but I believe this is not good evidence for it.