The "Is This Cheating?" Sheet's 40 Questions, One by One: How 33 Streamers Answered

First, where the data comes from

Streamers make their sheet answers public in three ways: filling it in live on stream, posting the completed sheet on X, or registering their answers in a summary page. This site's method: use the public summary matrix as the base text, transcribe every streamer's answers question by question, and attach each person's original source link (stream archive or post) so everything can be checked. Of 34 streamers with public answers, 33 completed all 40 questions — 1,320 valid choices in total.

Two caveats, stated plainly. First, the transcription uses the summary matrix as its base text, so it's second-hand; we have not yet re-watched each streamer's original footage to verify every question, and a few answers were inferred from circumstantial evidence where the original image was too blurry to read — those inferred values are flagged as "cannot compare precisely" in the comparison tool. Second, the original sheets were filled in by hand, and a few streamers' answers sit between two options (a question mark next to "off-putting," for example); there are 15 such in-between values, counted separately rather than forced into one of the three options.

So the precise claim behind every percentage below: among the 33 publicly available sets of answers this site had compiled by the end of October 2026, this many people picked this option on this question. It's a snapshot of this community, not rigorous survey research — but its job was never to survey; it's to give you something to talk about.

The strongest consensus: theme parks are ground zero

The question streamers judge "cheating" most unanimously is Q30, "going to a theme park one-on-one with a friend of the opposite sex": 28 of 33 call it straight-up cheating (85%), with only 2 picking totally fine. Sukoya Kana's note on this one is admirably concise: "わたしと行け" ("go with ME" — if you want a day out, ask me, not someone else). Close behind is Q27, "going to meet a friend of the opposite sex who says 'can you meet right now?'" (70% cheating); her note there reads "会うわい!!!" (Kansai dialect, literally "oh, I'd GO!!!").

The common thread is easy to see: a one-on-one outing that could have been politely declined, and wasn't. Compare Q22, "driving a friend of the opposite sex home when they've missed the last train" — only 30% call that cheating, because giving someone a ride carries a "what else could I do?" justification, and a theme park doesn't. What streamers actually mind is rarely the act itself; it's the "I could have said no" choice buried inside it.

The other extreme: feeling happy at a compliment, 82% say fine

The question everyone's most forgiving of is Q11, "simply feeling happy when someone of the opposite sex calls you 'cute' or 'good-looking'": 27 people (82%) pick totally fine, with only 3 calling it cheating. Similar ones: Q38, "exchanging ordinary pleasantries with an ex at a mutual friend's wedding" (12% cheating), and Q39, "hearing an ex has someone new and thinking 'good for them'" (73% totally fine).

Put those two tables side by side and the collective boundary line of 33 streamers comes into focus: feeling happy is an internal state and can't be cheating; a one-on-one outing that had a ready alternative is, for most people, the red line. Carry that line into the next chat argument you watch, and you can tell whether people are actually fighting about the behavior or the motive.

The most split question: 33 people, perfect thirds

Q36, "not deleting photos of an ex," landed at exactly 11 cheating, 11 off-putting, 11 totally fine — a perfect three-way split. Q1, "having lunch one-on-one with someone of the opposite sex," nearly replicates it: 10, 11, 12. When a single question divides one community this evenly, it means the question sits precisely in the middle of everyone's boundary line.

The notes on Q1 show where the disagreement comes from. Sukoya Kana wrote "ヤラバートはヤシ、仕事の合間ならしゃーない" ("the reckless kind is a no; if it's between work things, can't be helped"), and 渚トラウト wrote "事前に／偏ってきたら構って" ("fine if agreed in advance / a problem once it gets frequent") — same lunch, but one person judges the occasion, one judges the frequency, one judges who it's with. Three fixed options can't hold all those conditions, which is why the sheets are covered in small handwritten notes.

By section: close friends of the opposite sex judged hardest, ex-related questions most lenient

The 40 questions fall into four sections, and streamers' attitudes differ sharply between them. The ex section is the most lenient overall: continuing to use a wallet an ex gave you reads as cheating for only 18%, wedding pleasantries 12%, and 73% pick totally fine for genuinely being glad an ex has moved on — most people accept that a past relationship can stay respectfully in the past.

The strictest is the close friends of the opposite sex section: pet nicknames, daily chats, late-night calls — intimacy signals that a majority reads as cheating overall. One-on-one outings come right behind: theme parks 85%, shopping and picking clothes 64%, movies 61%, karaoke 52% — yet driving someone home after a missed last train is only 30%, meaning the recreational outing you could have declined is what stands out. The opposite-sex distance section lands in the middle, and one distinctly modern-life question jumps out: daily chatting and messaging (55% cheating) outranks late-night calls (42%) — frequency bothers people more than the hour of the night.

The famous notes

The 342 handwritten notes are the best part of this dataset. A few, to get a feel for the room: on Q17, "using a pet nickname with one particular friend of the opposite sex that only the two of you understand," るる・らら・りりぃ wrote flat-out "別れた方がいい" ("you should probably break up"); on the same question, Sukoya Kana's note reads "キモい" ("gross").

早瀬走, next to Q36 ("hasn't deleted photos of an ex"), wrote "8,278枚あるけど どこにあるかも わからん" ("I have 8,278 of them, and I don't even know where they are"). Sister Cleaire (シスター・クレア), on Q23 ("looking after a drunk friend of the opposite sex"), wrote "クレアがまってるからさ♡って笑って断ってほしい♡" ("I'd want them to smile and refuse with 'Cleaire is waiting for me ♡'") — casting themselves as the partner at home. And 渚トラウト, next to the bonus question at Q40, wrote "相手が嫌なら直しますって" ("if it bothers my partner, I'll change") — turning the whole sheet into a personal statement about how they do relationships.

These notes are exactly why the sheet went viral: the options are fixed, but the way everyone turns in their paper is completely their own.

What this data can and can't tell you

What it can tell you: where, within the VTuber streamer community, consensus on relationship boundaries is strong and where it falls apart. If your own answer distribution lands close to the streamers' percentages, your boundary line is fairly mainstream; if you split from the majority on a few specific questions, those are your best conversation starters with friends.

What it can't tell you: it doesn't represent the general population, and it is in no sense a set of correct answers. The sample is 33 streamers whose job is entertainment, and the compilation is based on second-hand transcription. To compare yourself question by question with any given streamer, use the answer comparison tool — all 33 sets of answers are in it. To fill in your own, the online sheet on the homepage is one tap away.