The Kindest Refusal

I · A Diagnosis Delivered in a Blackout

On the third of November, 1994, NBC broadcast the seventh episode of a situation comedy then in its first season. New York City has lost power. The characters are stranded in an apartment burning through its last candle; one of them is shut in a cash-machine vestibule with a lingerie model, trying to be charming. One of them, Ross, has been quietly in love with another, Rachel, for a very long time, and has said nothing about it for just as long. His friend Joey supplies the diagnosis. Ross waited too long to make his move. There is a zone for men who do that, and Ross is in it. Ross resolves to speak; a stray cat belonging to the new neighbor attacks him before he can.

The joke escaped the soundstage and became a category. The Oxford English Dictionary now records the friend zone as a situation in which a friendship exists between two people, one of whom has an unrequited romantic or sexual interest in the other. That is a fair description of a state of affairs, and it is not what the phrase does in conversation. In conversation the phrase explains. It says, in five syllables, that a man is somewhere; that he got there by moving too slowly; and that a woman put him there. The state, the cause, and the agent arrive together, which is why the expression is so satisfying to say and so difficult to examine.

The satisfaction is not evenly distributed. A man who says he was friend-zoned is reporting a real and unpleasant experience, and he is also, without necessarily intending to, filing a small complaint about somebody else’s preferences. A woman on the receiving end of the phrase hears an ordinary decision—she did not want to—described as an act she performed upon him. Very little in the vocabulary of modern courtship compresses so much into so little.

This magazine is interested in the parts of the sentence that can be checked. The reader’s question comes in three pieces, and they are worth keeping apart, because two of them have confident folk answers and the evidence declines to supply either. How does it happen: what is the mechanism by which two people arrive at incompatible views of what they are doing? When does it happen: is there really a window, and does closing it explain anything? And why: what is the friendship for, on each side, that neither party can afford to name it?

Some of the answers are unflattering to the man, some to the phrase, and one to a widely believed theory about women which does not survive its own experiments. The most useful of them is unflattering to nobody. It is that the friend zone is not a place, and that describing it as one has kept a generation of disappointed people from noticing the plainer thing that happened, which is that a question was never asked.

A caution about the evidence before it starts arriving. The research on cross-sex friendship is overwhelmingly American, heterosexual, and collegiate; where a study reached beyond undergraduates it will be said so. The findings below describe averages inside samples, and an average is a weak instrument for adjudicating anybody’s Tuesday. They are useful for the same reason a map is useful, and unreliable for the same reason.

II · Eighty-Eight Pairs

Nearly all of what is written about attraction between friends suffers from a design flaw so obvious that it is easy to miss: it asks one person about two people. In 2012, the psychologist April Bleske-Rechek and seven colleagues at the University of Wisconsin–Eau Claire avoided it. They recruited eighty-eight pairs of cross-sex friends—both halves, separately, under a promise that neither would ever see the other’s answers—and asked each of them, on nine-point scales, how attracted they were to their friend and how attracted they believed their friend was to them.

The first result is the one that gets quoted. Young men reported more attraction to their female friends, on average 4.94, than young women reported to their male friends, on average 3.97. About a point, on a nine-point scale, and both numbers sit near the middle of it: the average young man is not consumed, and the average young woman is not indifferent.

The second result is more interesting and almost never quoted. The two friends’ reported attractions barely tracked each other at all. Across the eighty-eight pairs the correlation was .19, which the authors report as not statistically significant. Knowing how much he wants her tells you close to nothing about how much she wants him. This is a stranger fact than it looks. Two people can spend two years in the same conversation, the same cafeteria, the same running joke, and end up at unrelated positions on the only question that would change what the friendship is.

The third result explains why nobody notices. Each person’s own attraction predicted their estimate of their friend’s attraction strongly—.61 for the men, .49 for the women. Both sexes were reading their own instrument and calling it the weather. The consequences were asymmetrical only because the starting positions were. Men overestimated how attracted their friend was to them; women underestimated it. He thought she was at 4.54 when she reported 3.97. She thought he was at 4.25 when he reported 4.94. Neither is a wild error. Both are errors in the direction that makes an undeclared courtship last.

The same group then asked a different question of a different sample, including adults between twenty-seven and fifty-two, recruited by post rather than from a lecture hall. Rather than rating scales, participants were invited to name the benefits and the costs of opposite-sex friendship in their own words. Attraction was volunteered as a cost or a complication by thirty-two per cent of them and as a benefit by six—five times more often a problem than a prize. Among the younger group the split by sex was wide: twenty-two per cent of men named attraction as a cost, against forty-seven per cent of women. Twelve per cent of men named it as a benefit, against three per cent of women.

Read those two lines together and the friend zone stops looking like a room with a door. The men are more attracted, more likely to think the feeling is returned, and more likely to regard the whole business as an upside. The women are less attracted, likelier to underestimate the interest coming at them, and roughly twice as likely to describe attraction as the thing that spoils an otherwise good friendship. Nobody has put anybody anywhere. Two people are standing in different weather and both are checking their own instrument.

III · The Window That Was Never There

Now the timing, which is the load-bearing claim in the folk theory and the one that fails most cleanly. If the friend zone were a place entered by delay, then the longer two people had been friends, the cooler the romantic possibility should be. Bleske-Rechek’s group had the data to check this, since they knew how long each pair had been friends. Friendship duration was unrelated to attraction, and unrelated to perceived attraction, for either sex. Not weakly related. Unrelated, with every test comfortably short of significance.

Something much larger points the same way. In 2022, the psychologists Danu Anthony Stinson, Jessica Cameron, and Lisa Hoplock pooled seven samples of university students and crowdsourced adults—1,897 people in all—and asked how their current or most recent romantic relationship had begun. About two-thirds of them had been friends first. Among the students, friends-first was not merely the commonest route; asked what they would prefer, they preferred it. The friendships that turned had lasted, on the authors’ reckoning, on the order of two years before anything romantic happened, and most participants said they had not entered them with romantic intentions at all.

The authors’ second finding is a rebuke to their own discipline. The relationship-initiation literature, they observe, is built almost entirely on strangers meeting—the laboratory introduction, the speed-dating event, the first-encounter vignette—and has largely overlooked the pathway that most people actually take. Science studied the exception because the exception is easier to arrange in a room.

These two findings dismantle the window. There is no evidence of a closing aperture, and there is good evidence that the corridor most couples walked down was a friendship of about two years in which, at the outset, nobody was pursuing anybody. If long friendship extinguished romance, two-thirds of the couples in that meta-analysis would not exist.

And yet the experience the phrase names is obviously real. Something does happen, to a lot of people, and it feels exactly like being placed. Both things can be true if the variable is not time. The friendships that turned romantic and the friendships that produced a disappointed man are not distinguished by their duration; they are distinguished by whether anyone ever said what they wanted, and by whether the other person, when asked, said yes.

That reframing costs the phrase most of its content. “I waited too long” describes a race against a clock that the evidence cannot find. “I never asked, and I was afraid to, because the friendship was worth something and the question would have put it at risk” describes a real dilemma, one with genuine stakes on both sides, and it has the additional merit of being a sentence about a decision the speaker made.

IV · The Light He Is Reading By

In 1982, the psychologist Antonia Abbey sat a man and a woman down for a five-minute conversation while another man and another woman watched from a hidden position. Afterward everyone rated the female participant. The male participant and the male observer both rated her as more seductive and more promiscuous than the women rated her. She had done nothing differently for the four of them. The study is small, forty-four years old, and has been argued over ever since, and it opened one of the most durable questions in the field: whether men systematically read sexual interest into behavior that does not contain it.

For two decades the leading explanation was an evolutionary one. In 2000, the psychologists Martie Haselton and David Buss proposed that inference systems should be biased wherever the two possible mistakes carry unequal costs—error management, in their term. A man who wrongly infers interest loses a little time and some dignity; a man who wrongly misses it loses a reproductive opportunity. Bias the dial toward the cheaper error and you get overperception. The same logic, run for women, predicts that they will underestimate a man’s commitment, which is what Haselton and Buss found in the first of their studies, on two hundred and seventeen participants.

The prediction has repeatedly been confirmed in the shape the theory expects. In a 2012 speed-meeting study, the psychologists Carin Perilloux, Judith Easton, and David Buss put ninety-six men and a hundred and three women through five three-minute conversations each. Men overestimated women’s interest; women underestimated men’s. The overestimating was concentrated in men oriented toward short-term relationships, in men who rated themselves attractive, and in encounters with more attractive women.

Then the foundation moved. In 2015, Perilloux herself, writing with the psychologist Robert Kurzban, pointed out that the whole literature rests on a comparison it cannot make. To say that a man overestimates a woman’s sexual intent, one must know the woman’s sexual intent, and the only available measure is her own report. Women, the two authors propose, may understate it. If they do, men may be reading accurately and the field has been scoring the answer key wrong for thirty years. The objection is not a refutation. It is worse: it is an unresolved problem with the criterion.

A second correction came from scale. In 2020, the psychologist Anthony Lee and four colleagues ran 1,226 people through a speed-dating study and found the familiar pattern—men over, women under—and then found that it very largely dissolved once they accounted for two ordinary things. One was sociosexual orientation, how open a person is to uncommitted sex. The other was projection: a person’s own level of interest predicted what they saw in the other chair. Overperception is not a male instrument. It is what desire does to inference, in whoever is doing the desiring.

Which returns us, usefully, to the friendship. In 2016 the psychologists Edward Lemay and Noah Wolf studied opposite-sex friends over time and found projection operating exactly where this essay needs it: friends who wanted more believed the wanting was more mutual than it was. Two details make the finding sting. Projection was stronger among those who considered themselves high in mate value—confidence does not correct the error, it feeds it. And the projection produced behavior, and the behavior sometimes worked: perceivers who acted on the belief changed their friend’s desire over time, particularly when the friend regarded them as high in mate value. Wishful thinking, in that subset, was a self-fulfilling prophecy.

This is why the bias survives, and why no essay can honestly tell a man that his reading of the situation is a delusion. Sometimes he is right. Sometimes he is right because he acted. What the evidence will not support is the confident version—that he can tell which case he is in from the inside, using an instrument that has been shown to report his own state back to him.

V · The Grammar of No

There remains the question of what she said, and here the most useful evidence comes from an unexpected direction. In 1999, the social psychologists Celia Kitzinger and Hannah Frith applied conversation analysis—the study of how ordinary talk is actually built, turn by turn—to the business of refusing. Their subject was sexual refusal and their target was the advice, then standard in prevention programs, that a woman should simply say no. What their material showed is that almost nobody, refusing almost anything, simply says no.

A refusal in ordinary English is an elaborate object. It arrives late rather than promptly, because promptness is the signature of acceptance. It is prefaced, usually with appreciation. It is padded with a palliative—something warm placed next to the refusal so that the refusal is not the only thing in the room. And it is accompanied by an account, a reason external to the refuser’s preferences: the timing, the circumstances, a general condition of life. Kitzinger and Frith’s other finding is the one that matters here: a bare, unhedged no is heard as rude, arrogant, or foolish. Directness is not the neutral option. It is the socially expensive one.

The scope of that work must travel with it. It was written for refusal-skills training and date-rape prevention, and it reaches conclusions about sexual coercion that are not the subject of this essay; a later study by the psychologists Rachael O’Byrne, Mark Rapley, and Susan Hansen extended it by showing that young men, in focus groups, describe the indirect grammar of refusal with complete fluency. What transfers to a disappointed courtship is only the description of the form. But that description is devastating enough on its own.

Assemble the pieces. She likes him and does not want him. She must convey the second without withdrawing the first, because the first is true and because the friendship is not a fiction. The language available for this has evolved to be gentle, and gentleness is achieved by delay, warmth, padding, and an external reason. Now hand that utterance to a listener whose own attraction is the light he is reading by, who has been shown to project it onto the person in front of him, and who is being told, in the same breath, that she thinks he is wonderful.

The unrequited-love research records the result precisely. In 1993, the psychologists Roy Baumeister, Sara Wotman, and Arlene Stillwell collected a hundred and thirty-four accounts of unreciprocated attraction, written from both roles. The would-be lovers reported that the attraction had seemed mutual, that they had been led on, and—the phrase worth keeping—that the refusal had never been definitely communicated. The rejectors reported guilt, and scriptlessness, and the peculiar difficulty of having hurt somebody while doing nothing wrong. Both sides are giving honest testimony. She refused in the only grammar available to her. He did not receive it, in part because it was built not to wound.

Two limits keep this from becoming an accusation in either direction. The first is that indirectness is sometimes not a decision at all: a person can be genuinely unsure, and a hedged answer is then simply an accurate one. The second is that none of this applies to the man who asked plainly and was told plainly. That is ordinary rejection, it is not what the phrase describes, and it has the considerable advantage of being over.

VI · What the Friendship Was For

A friendship that neither party can afford to define is usually doing something for both of them. In 2000, the psychologists April Bleske and David Buss asked several hundred people to name what opposite-sex friendship supplied, and then to rate how valuable each item would be. The sexes diverged in ways that are uncomfortable in both directions. Men rated sex with an opposite-sex friend as a more substantial benefit than women did. Women more often named protection as something a male friend provided, and rated it highly when they did.

Neither of those findings describes a scheme. They describe average valuations, drawn from self-report, in American samples. But they do establish that the phrase “just friends” can cover arrangements that are not symmetrical in what they deliver, and that the asymmetry runs in more than one direction. Some friendships are courtships that were never announced. Some are arrangements in which reliable attention and a measure of safety are received without any romantic price attached, which is not a crime, and is also not nothing.

The communication scholar Heidi Reeder made the useful distinction. Interviewing pairs of friends, she separated four kinds of attraction that can occur between them: finding the other person physically attractive oneself, recognizing that they are attractive without feeling it, wanting a romance, and simple attraction to the friendship itself. In her larger questionnaire study, friendship attraction was the most common and romantic attraction the least. The friend zone, in Reeder’s terms, is not a zone. It is one of these four dials being turned up in one person and not the other.

Which brings us to the folk theory that has done the most damage, and which is straightforwardly false. The nice-guy account holds that women announce a preference for kind men and then reliably choose otherwise, so that kindness is not merely unrewarded but actively penalized. In 2003, the psychologists Geoffrey Urbaniak and Peter Kilmann tested it. College women read scripts in which two men competed for a date and the niceness of one was manipulated; in a second study, with a hundred and ninety-four participants, both niceness and physical attractiveness were varied. Niceness was a positive factor in the women’s choices and in their ratings of desirability. It was not a penalty. Attractiveness weighed more heavily when the man was being evaluated as a casual sexual partner, and less when he was not.

So the man who concludes that he was too kind has drawn the wrong lesson from real pain. Nothing in the evidence says kindness is punished. What fails is something else, and it is worth naming precisely: kindness extended on account, against a debt the other party was never asked to agree to. Lifts to the airport, two-hour phone calls, the standing offer to help move a sofa—all freely given, all genuinely appreciated, and all quietly entered in a ledger the other person cannot read. When the invoice is presented, in the form of a declaration made after two years of accumulated service, it lands as a surprise and a claim at the same time. She is being asked to settle an account she did not open.

The reverse error deserves the same treatment. The complaint packed into “she friend-zoned me” requires that an ordinary preference be reclassified as a withholding—that not wanting somebody become a thing done to him. It is a small injustice, done mostly by grammar rather than by malice, and it is the reason the phrase deserves less affection than it gets.

VII · Who Does the Introducing

An undeclared courtship needs a channel in which it can happen. The best available evidence about channels is national rather than New York-specific: in Rosenfeld, Thomas, and Hausen’s American survey, the share of couples who met online rose from two per cent in 1995 to thirty-nine in 2017, while the share who met through friends fell from thirty-three to twenty. The categories can overlap, and the figures describe couples who formed, not every friendship in which someone hoped for more.

New York offers both routes at once. An app can introduce strangers who are explicit about looking; a friend can introduce people whose intentions remain unknown; a workplace or shared activity can sustain acquaintance before either person declares an interest. None of these channels supplies a moral ranking of the people who use it. Each gives a different amount of time for attraction, friendship, and uncertainty to coexist.

That is why the phrase “friend zone” explains so little. It assumes that friendship was a waiting room and that somebody failed to call the next number. A friendship may instead have been the relationship one person actually wanted. The national meeting figures do not tell us how often that happens in New York or how many undeclared courtships failed. The practical question is smaller: can both people name what they want without making the other responsible for having wanted something else?

VIII · The Kindest Refusal

It is worth noticing what the 1994 episode does with its own joke. Joey issues the diagnosis—you waited too long, there is a zone, you are in it—and then the script declines to test it. Ross resolves to speak, and a cat prevents him, in the dark. A phrase about two people failing to establish what they are to each other was coined, as it happens, during a blackout, and the man it was coined about never got the sentence out. The writers presumably did not intend the joke to reach that far.

The answers, then, to the three questions. How: two people arrive at incompatible readings because each is estimating the other from their own level of interest, and because the refusal, when it is issued, is issued in a grammar built to be gentle, which is the same as saying built to be deniable. When: not at the closing of a window, because no window can be found—friendship duration predicts nothing, and two-thirds of romances began exactly here—but at the moment a declaration is either made or, more often, not made. Why: because the friendship is supplying something to both parties, because declaring puts that something at risk, and because the man’s most reliable instrument for reading the situation is the one that reports his own state back to him with the label of hers.

A final absence should be reported rather than papered over. This essay found no peer-reviewed estimate of how often anyone is friend-zoned, or of how that experience is distributed. The one study located that addressed the phrase directly is an unpublished poster presented at Binghamton University in 2017 by Lillian Harrington and seven colleagues, and its public abstract does not carry the percentages that circulate under its name; they are therefore not repeated here. The most-used word in the modern vocabulary of romantic disappointment has essentially no prevalence literature behind it. That is itself a finding, and not a flattering one for the field.

What the evidence can say, it says quietly. It can say who is likelier to be more attracted, and who is likelier to misread whom, and in which direction, and by roughly how much, in samples of American undergraduates who were promised their friend would never see the page. It cannot say who will be wanted by whom, and it has never been able to. It can, however, say something about the sentence itself. “She friend-zoned me” and “he never actually said anything” are, in the great majority of cases, one situation described from two chairs. Only one of those sentences contains a decision the speaker could have made differently, which is presumably why it is the less popular of the two.

Notes on Sources

Friends, season one, episode seven, “The One with the Blackout,” NBC, first broadcast November 3, 1994. Source of the phrase’s popular entry into English; the blackout, the last candle, the cash-machine vestibule, and the cat are from the episode’s published synopsis. Accounts of the exact dialogue differ slightly in whether Ross is warned about the friend zone or diagnosed as already in it, so the essay paraphrases rather than quotes.† (An origin for the words, not evidence about the phenomenon, which predates the term.)

Oxford English Dictionary, “friend zone.” Source of the definition quoted.† (The wording was read in secondary reproduction of the entry rather than in the subscription database.)

Bleske-Rechek, A., Somers, E., Micke, C., Erickson, L., Matteson, L., Stocco, C., Schumacher, B., & Ritchie, L. (2012). Benefit or burden? Attraction in cross-sex friendship. Journal of Social and Personal Relationships, 29(5), 569–596. Study 1: eighty-eight pairs; means of 4.94 and 3.97 and the estimates of 4.54 and 4.25 are Table 1 values on nine-point scales; friends’ attractions correlated r(87) = .19, p = .09; each sex’s own attraction correlated with its estimate of the friend’s at .61 and .49; friendship duration was unrelated to attraction or perceived attraction, all ps > .40. Study 2: postal sample including adults aged twenty-seven to fifty-two; attraction nominated as a cost by thirty-two per cent and as a benefit by six; among emerging adults, twenty-two per cent of men against forty-seven per cent of women named it a cost, and twelve against three per cent a benefit.† (Small Midwestern undergraduate sample in Study 1; self-report throughout. The paper’s reported effect sizes are not all reconcilable with its reported test statistics, so no d from it is cited here.)

Stinson, D. A., Cameron, J. J., & Hoplock, L. B. (2022). The friends-to-lovers pathway to romance: Prevalent, preferred, and overlooked by science. Social Psychological and Personality Science, 13(2), 562–571. Meta-analysis of seven samples, N = 1,897; source of the two-thirds figure, the students’ stated preference, the approximate two-year friendship, the authors’ report that most friends-first participants began their friendships without romantic intent, and the observation that initiation research has concentrated on strangers.† (Retrospective self-report from student and crowdsourced samples; it describes couples that formed and cannot estimate how often the same pathway fails.)

Abbey, A. (1982). Sex differences in attributions for friendly behavior: Do males misperceive females’ friendliness? Journal of Personality and Social Psychology, 42, 830–838.† (A single small laboratory interaction, four decades old, and the origin of a still-unsettled debate.) Haselton, M. G., & Buss, D. M. (2000). Error management theory: A new perspective on biases in cross-sex mind reading. Journal of Personality and Social Psychology, 78(1), 81–91; Study 1 N = 217. La France, B. H., Henningsen, D. D., Oates, A., & Shaw, C. M. (2009). Social–sexual interactions? Meta-analyses of sex differences in perceptions of flirtatiousness, seductiveness, and promiscuousness. Communication Monographs, 76(3), 263–285.

Perilloux, C., Easton, J. A., & Buss, D. M. (2012). The misperception of sexual interest. Psychological Science, 23(2), 146–151. N = 199, ninety-six men and a hundred and three women, five three-minute conversations each. Perilloux, C., & Kurzban, R. (2015). Do men overperceive women’s sexual interest? Psychological Science, 26(1), 70–77.† (Source of the criterion objection: the comparison requires a measure of the woman’s true intent that does not exist independently of her report.) Lee, A. J., Sidari, M. J., Murphy, S. C., Sherlock, J. M., & Zietsch, B. P. (2020). Sex differences in misperceptions of sexual interest can be explained by sociosexual orientation and men projecting their own interest onto women. Psychological Science, 31(2), 184–192. N = 1,226.

Lemay, E. P., & Wolf, N. R. (2016). Projection of romantic and sexual desire in opposite-sex friendships: How wishful thinking creates a self-fulfilling prophecy. Personality and Social Psychology Bulletin, 42(7), 864–878. Source of the projection, mate-value, and self-fulfilling findings.† (The self-fulfilling result is correlational across time in young samples; it explains a pattern and licenses no tactic.)

Kitzinger, C., & Frith, H. (1999). Just say no? The use of conversation analysis in developing a feminist perspective on sexual refusal. Discourse & Society, 10(3), 293–316. Source of the delay, preface, palliative, and account structure and of the finding that bare refusals are judged rude. O’Byrne, R., Rapley, M., & Hansen, S. (2006). “You couldn’t say ‘no,’ could you?”: Young men’s understandings of sexual refusal. Feminism & Psychology, 16(2), 133–154.† (Both papers were written for refusal-skills training and date-rape prevention and reach conclusions about sexual coercion. Only their description of the conversational form of refusal is borrowed here; nothing else in them transfers to courtship disappointment.)

Baumeister, R. F., Wotman, S. R., & Stillwell, A. M. (1993). Unrequited love: On heartbreak, anger, guilt, scriptlessness, and humiliation. Journal of Personality and Social Psychology, 64(3), 377–394. A hundred and thirty-four accounts from seventy-one students; cited here only for the would-be lovers’ report that no definite refusal was issued, and the rejectors’ guilt and scriptlessness.† (Unpaired retrospective narratives from a young student sample.)

Bleske, A. L., & Buss, D. M. (2000). Can men and women be just friends? Personal Relationships, 7, 131–151. Act nominations from four hundred participants; Study 1 N = 231, Study 2 N = 229. Reeder, H. M. (2000). “I like you . . . as a friend”: The role of attraction in cross-sex friendship. Journal of Social and Personal Relationships, 17(3), 329–348. Twenty interviewed pairs and a questionnaire study of 231; source of the four attraction types and of friendship attraction being commonest and romantic attraction least.† (Self-reported nominations of recalled and hypothetical benefits in predominantly American samples; averages, not accounts of any particular friendship.)

Urbaniak, G. C., & Kilmann, P. R. (2003). Physical attractiveness and the “nice guy paradox”: Do nice guys really finish last? Sex Roles, 49, 413–426. Study 1 N = 48; Study 2 N = 194 college women.† (Scripted vignettes measuring stated choice. They refute the strong form of the stereotype; they do not predict any individual’s behavior.)

Rosenfeld, M. J., Thomas, R. J., & Hausen, S. (2019). Disintermediating your friends: How online dating in the United States displaces other ways of meeting. Proceedings of the National Academy of Sciences, 116, 17753–17758. Table 1 point values for 1995 and 2017 across 5,421 heterosexual couples from the 2009 and 2017 waves of How Couples Meet and Stay Together; the paper states that categories are not mutually exclusive.

Harrington, L., Butler, A., Boyle, M., Goodman, T., Katz, S., Cinnamo, M., Merriwether, A. M., & Massey, S. (2017). The Friend zone: An Infinite Place Where Nothing Good Happens: A Comprehensive Study of Friend zoning Across Varying Genders and Sexual Orientations. Research Days poster, Binghamton University, April 21, 2017.† (An unpublished conference poster, and the only located study addressing the term itself. Its abstract of record does not contain the percentages circulating in secondary summaries, so none are quoted.)

† marks findings limited by design, sample, publication status, contested criteria, or transfer beyond their tested setting. Figures 2 and 3 are qualitative editorial diagrams; Figure 1 plots published means. Part VII uses national American meeting-channel data and does not claim a New York estimate. No study comparing the prevalence of the “friend zone” across the city’s meeting channels was located.

Drafted with AI (Claude Opus 5) as a private research synthesis, September 1, 2026; edited, and answered for, by humans. Every study named above was read at its source or its published record before it was cited, the two figures carrying numbers carry only published ones, the field’s missing prevalence data is reported rather than substituted, and one tempting argument was cut for want of a dictionary. Set, with affection and without affiliation, in the manner of a certain Manhattan weekly.New York revision drafted with AI on October 12, 2026; editorial review pending. ◆ ◆ ◆