The Measure of a Man

I · The Forest and the Spark

Some years ago, three psychologists—Samantha Joel, Paul Eastwick, and Eli Finkel—asked a computer to solve romance. The raw material was generous: hundreds of young adults had completed what amounted to more than a hundred questionnaires’ worth of self-description—their traits, their values, their politics, their bodies, their stated ideals in a partner, down to preferences most people are never asked to articulate—before being sent on rounds of four-minute speed dates with one another. The researchers then trained random forests, a workhorse of machine learning, to predict who would want whom. If desire is an arithmetic of attributes, as both the poets’ enemies and the dating industry insist, the machine held every variable anyone had thought to measure.

The results, published in Psychological Science in 2017, were a beautiful failure. The models did respectably at two of the three things asked of them. They could predict which individuals would be desired in general—the consensus targets, the people nearly everyone rated highly. They could predict which individuals would do the desiring—the romantically eager, who rated nearly everyone highly. But the third thing, the only thing anyone actually wants forecast—whether this man would want that woman, and she him, over and above their general appeal—the models could not predict at all. The unique spark between two particular people, the authors reported, was essentially unforecastable from everything measurable about them in advance.

This essay is an attempt to take that result seriously, in both directions. Because the first two predictions did work: desirability in general—the property of being widely wanted—is real, measurable, and surprisingly well understood. A half century of evolutionary psychology, and a quieter tradition of sociology, have mapped it in some detail for the case this essay takes up: what makes a man desirable to women, in bed and in life. The findings arrange themselves into three stories, told at three altitudes. There is the story of the species—the old preferences, written slowly, that travel with the body. There is the story of the market—the local economy of status, money, scarcity, and fashion that decides what a given city will pay for. And there is the story of the room—the dyad, the encounter, the two people actually present—where, as the random forests discovered, the first two stories surrender jurisdiction.

A note on the word. “Desirable” is not one property but a verdict returned by different juries on different timescales—a night, a season, a life—and the juries weight the evidence differently. The science is at its best when it keeps the timescales apart, and so will we.

II · The Body Remembers

Begin with the oldest story. Robert Trivers observed, in 1972, that the sex investing more in offspring—in humans, conspicuously the one that gestates them—evolves to be choosier, and the sex investing less evolves to compete for the choosy. From this single asymmetry, evolutionary psychology derived its catalogue of predictions about what women, on average and especially for short acquaintance, should find compelling in men: signs of health, formidability, and genetic quality; later, and more interestingly, signs of the willingness to stay and provision. The catalogue has since passed through the fire of the replication crisis, and it is worth saying plainly which entries survived.

Strength survived, embarrassingly well. In 2017, Aaron Sell and colleagues had women rate photographs of male torsos and found that cues of upper-body strength accounted for roughly seventy per cent of the variance in bodily attractiveness—and that once strength, leanness, and height were counted, there was almost nothing left to explain. They went looking, in their data, for a contingent of women who preferred physically weaker men, and reported finding none. Height survived, too, with an asterisk: the preference is relative, not absolute—taller than her, comfortably, with returns diminishing at the extremes. The economists Günter Hitsch, Ali Hortaçsu, and Dan Ariely, mining an American online-dating site, put an exchange rate on it: for a man of five feet six to attract the same attention as a six-footer, his stated income needed to be higher by roughly a hundred and seventy-five thousand dollars a year. Musculature helps most in moderation—David Frederick and Martie Haselton found women’s ideal noticeably more muscular than the average man but well short of the bodybuilder, a physique best summarized as built, not bulging; the very muscular reported more brief liaisons and inspired less confidence in their fidelity.

The face is a quieter instrument than advertised. Symmetry and averageness carry modest, real effects; the strong-jawed, testosterone-cast face turns out to be a contingent taste rather than a law, preferred in some contexts and penalized as a fidelity risk in others. Facial hair obeys fashion’s arithmetic: heavy stubble tends to win ratings, and beards gain value as they grow rare—a negative frequency dependence that Zinnia Janif and colleagues demonstrated with the cheerful title of “peak beard.” The voice matters, and reveals something: David Puts and others have shown that a lower pitch impresses other men—as a signal of dominance—rather more reliably than it seduces women, a pattern repeated across much of the male ornament. A good deal of what men assume women want was evolved, it appears, for the benefit of an audience of rivals. The baritone is partly for the other baritones.

Now the entries that did not survive. The elegant idea that women’s preferences swing toward rugged masculinity in the fertile window of the cycle—for two decades a staple of magazine science—failed its large preregistered tests; Benedict Jones and colleagues, tracking hormones directly, found no compelling evidence of the shift. The famous sweaty-T-shirt studies, in which women preferred the scent of immunologically dissimilar men, have proved fragile under meta-analysis. The replication crisis came for desire, too, and it mostly took the tabloid parts.

One more finding belongs here, because it corrects the whole genre. On paper, men say looks matter more and women say they matter less. In the flesh, the difference evaporates: Eastwick and Finkel showed that in live speed dating, physical attractiveness predicted romantic interest about equally for both sexes. Women’s stated discount of male beauty is largely a paper phenomenon. The practical reading of the body’s evidence is thus less deterministic than either its fans or its critics suppose: the body sets a floor more than a ceiling, and the floor—strength, leanness, grooming, bearing, the general broadcast of vitality—is, of all the tiers in the architecture, the one most tractable to effort.

III · The Ledger

Money is the finding everyone has heard of, usually in caricature. The caricature has a distinguished source. In 1989, David Buss published the largest mate-preference survey then attempted—ten thousand and forty-seven people across thirty-seven cultures—and found that women rated “good financial prospect” higher than men did in virtually every one of them, and preferred partners a few years older besides. Thirty years on, a preregistered replication led by Kathryn Walter, spanning forty-five countries and some fourteen thousand participants, found the same sex differences everywhere it looked. Whether they shrink as societies grow more equal is the field’s liveliest argument—one prominent study found the gap narrowing with national gender parity, the larger replication mostly did not—and the honest summary is that the preference is robust and its ultimate cause still contested: an evolved bias, say the evolutionists; the rational residue of women’s restricted access to wealth, say the social-role theorists; in all likelihood some of each, in proportions that shift with the society.

But the caricature—woman as gold-digger, man as wallet—misreads the ledger twice. First, money has never topped the list. In Buss’s data and nearly every survey since, both sexes put kindness and intelligence above wealth, and traits like dependability and emotional stability crowd the next places. Second, the preference behaves less like greed than like a threshold. In a clever series of experiments, Norman Li gave people constrained “mate dollars” to spend on a partner’s attributes. On tight budgets the sexes diverged sharply: women spent first on intelligence and status, treating them as necessities, while men treated physical attractiveness as the necessity. On lavish budgets, everyone converged on the same warm, clever, well-made person. The sexes differ at the floor, not at the ceiling: resources function for women the way beauty functions for men—a qualification to be met, after which other virtues decide the outcome.

Why should wealth attract at all? Not, mostly, for its own spending. Money arrives bundled with the things it certifies: competence and discipline (somebody paid this man for something, repeatedly), security against the world’s bad weather, access to its rooms, and a forecast—the reasonable expectation that the future will resemble the balance sheet. It is a signal dense with other signals, which is why its close cousins work too: ambition, industriousness, education, and what the psychologists call earning capacity, all of which women in the surveys weight as heavily as present holdings. And the signal has a grammar. Displayed wealth and possessed wealth read differently: in a series of studies wryly titled “Peacocks, Porsches, and Thorstein Veblen,” Jill Sundie and colleagues found that conspicuous consumption by men was read by women—accurately—as mating effort, and boosted a man’s appeal for a fling while doing nothing for his appeal as a husband. The Porsche is legible, but what it spells is short-term.

Status behaves the same way, and the science has usefully split it in two. Joseph Henrich’s distinction between dominance—deference extracted by force or fear—and prestige—deference granted freely, to competence—turns out to organize the attraction data almost perfectly. Prestige is desirable across every timescale. Dominance attracts, when it attracts at all, in bounded arenas: the athlete in the contest, the man assertive against rivals but never against her. Jeffrey Snyder and colleagues found women drawn to dominance in tournaments and repelled by it in interpersonal life. The desirable form of status, in other words, is the kind other people volunteer. Meanwhile the sociologists add a colder ledger note: as women’s education and earnings have risen, the marriage market has grown more assortative, not less—Christine Schwartz and Robert Mare documented earners increasingly marrying earners, degrees marrying degrees. In the educated urban market, his money no longer substitutes for the rest of him; it merely keeps him in a tournament where everyone has some.

IV · Character

The most replicated finding in the whole literature is also the least reported, presumably because it makes a poor headline: the trait women rank first, nearly everywhere it has been measured, is kindness. In Buss’s thirty-seven cultures, “kind and understanding” led the list for both sexes, with dependable character and emotional stability close behind—ahead of money, ahead of looks, ahead of everything the industry sells. In Li’s budget experiments, warmth was a necessity purchase for men and women alike. The evolutionary reading is unsentimental: a long-term mate is the single largest cooperative venture most humans undertake, and the trait that best forecasts a partner’s behavior across decades of unforeseeable circumstances is not his jawline or his salary but whether he is, reliably, kind. Desirability over a life is mostly a solvency judgment about character.

Around that core, the character findings arrange themselves into a small portfolio. Humor, first. The sexes trade it asymmetrically—Eric Bressler and Sigal Balshine found that women want a man who produces humor, while men want a woman who appreciates theirs—and the evolutionists’ explanation is that a joke is an intelligence test administered painlessly: timing, theory of mind, verbal reach, and confidence, all audited in a sentence, impossible to fake for long. Intelligence itself attracts almost monotonically, with some evidence of a mild penalty at the very top of the scale—the ninetieth percentile, in one study, outscoring the ninety-ninth—perhaps because brilliance without warmth reads as distance. Emotional stability does its work quietly: neuroticism is among the best prospective predictors of a relationship’s misery, and people seem to price it in advance. Generosity attracts when it is habitual rather than performed—altruists report more romantic success, and the costly-signal logic requires the cost to be real—while bravery, inconveniently for the etiquette books, outranks mere niceness: in vignette studies, the heroic risk-taker beat the equally pleasant non-hero. Confidence attracts as a forecast of competence and repels the moment it curdles into entitlement, which is a forecast of expense.

Two registers, again. For brief acquaintance, the data are cheerfully cynical: flash, formidability, immediacy, and a certain unbotheredness all rent well, and the cad has his weekend. But “desirable overall”—the phrasing of the assignment—is decided by the long jury, and the long jury reads character the way an underwriter reads a balance sheet: kindness as collateral, stability as credit history, humor as liquidity, integrity as the absence of undisclosed liabilities. The short game and the long game share a single genuinely transferable asset, and it is warmth attached to competence—the man who is plainly good at something and plainly good to people. Nearly everything else is rented plumage.

V · The Room

In the early nineteen-seventies, on a footbridge of cedar planks and wire cable swaying two hundred and thirty feet above the Capilano River in British Columbia, a young woman with a clipboard stopped men halfway across and asked them to invent a brief story about an ambiguous picture. She offered her phone number in case they had questions. Men interviewed on the wobbling bridge wrote stories with more sexual imagery, and called her afterward at several times the rate of men interviewed on a low, solid bridge upstream. Donald Dutton and Arthur Aron’s experiment became the most charming demonstration of a large truth: attraction is not only an appraisal of the other person; it is an interpretation of one’s own racing pulse, and the interpretation can borrow arousal from anywhere—fear, exercise, laughter, a near-miss. Part of desirability is simply being present, plausibly, when the heart is already loud.

The room has other laws, all of them offstage in the trait catalogues. Judgment is fast—Janine Willis and Alexander Todorov showed that a tenth of a second of a face yields verdicts that longer exposure mostly just firms up—but fast judgment is thin judgment, and it decays on contact with information. Familiarity compounds: in a classic study, Richard Moreland and Scott Beach had women silently attend a lecture course five, ten, or fifteen times, speaking to no one, and the students rated the most frequently seen faces most attractive. Proximity does the quiet work Leon Festinger documented in postwar married-student housing, where friendships tracked stairwells. Reciprocity is close to a reflex—discovering that someone likes us is among the most reliable causes of liking them—with one refinement out of Erin Whitchurch, Timothy Wilson, and Daniel Gilbert’s lab: women were most attracted to men whose interest was uncertain, the mind returning helplessly to the unresolved. And underneath it all runs the machinery Helen Fisher and Aron saw on the scanner: early-stage attraction lights the dopaminergic wanting system, the circuitry of reward pursuit—which is why desire behaves less like a verdict than like an appetite with a narrative attached.

The deepest of the room’s laws is the one that dissolves the market entirely. Eastwick, with Lucy Hunt, tracked how much observers agree about who is desirable. Among strangers, consensus is high: everyone sees the same handsome few. With acquaintance, consensus collapses. Month by month, as people accumulate history with one another—the joke that landed, the kindness witnessed, the shared ordeal—their rankings diverge until desirability is mostly idiosyncratic: a property of the pair, not the person. Mate value stops being a number and becomes a biography. This is the room’s answer to the random forests, and it is why the failure to predict the spark is not a gap in the science but its most secure result: the spark is manufactured in situ, out of materials—timing, attention, misattributed adrenaline, accumulating private history—that do not exist until the two people are in each other’s presence.

VI · The Market

Sociology enters where evolutionary psychology stops: at the observation that a man’s desirability is not carried entirely on his person. It is priced, and prices require a market. Pierre Bourdieu supplied the accounting: individuals hold capitals—economic (money), cultural (taste, education, accent), social (who will vouch for you)—and fields in which those capitals are worth more or less. Catherine Hakim pressed the point by naming a fourth, erotic capital—beauty, charm, self-presentation, social grace—and arguing, controversially, that it is a genuine asset class, undervalued only because acknowledging it embarrasses everyone. The decisive correction came from Adam Isaiah Green, whose “sexual fields” framework observed that there is no exchange rate valid everywhere: every scene—a hotel bar, a climbing gym, a gallery opening, an app—is a field with its own hegemonic currency, its own structure of desire. The same man is a different man three blocks away. This is the single most practical idea in the sociology of attraction, and it is almost entirely absent from the self-improvement literature, which speaks of raising one’s value as if there were one bourse.

The great structural event of the era is disintermediation. Michael Rosenfeld’s surveys of how American couples meet show every traditional broker—family, church, colleagues, friends—in decline for decades, and the graph of “met online” rising through them all: roughly two-fifths of new heterosexual couples by 2017, a share that has only risen since. The apps did not merely move courtship; they re-weighted it—toward the photographable and the legible (height, job title, altitude of the selfie), and toward abundance, which has a psychology of its own. Dutch researchers documented a “rejection mindset”: as options accumulate, the default answer drifts toward no. Eli Finkel’s review of the industry’s matchmaking claims reached the same verdict as the random forests—no algorithm has demonstrated it can predict compatibility from profiles—which means the apps sell access, not prophecy. And access has a geometry. Elizabeth Bruch and Mark Newman, ranking desirability on an American dating site the way Google ranks pages, found a steep hierarchy in four cities, with most users messaging hopefully upward, toward partners about twenty-five per cent more desirable than themselves, replies growing rare in proportion to the reach.

Markets also have demographics. Marcia Guttentag and Paul Secord argued in 1983 that the ratio of the sexes sets a market’s moral climate—scarce women raise the price of commitment; scarce men lower it—though modern field data have complicated the folk version, sometimes finding men more committal where they are the surplus sex, competing on fidelity itself. Ashley Mears’s ethnography of VIP nightlife described the purest market of all, in which beauty circulates as a currency men spend money to sit beside, the spending itself being the display. And hovering over every field is what R. W. Connell called hegemonic masculinity: each era’s locally winning script of manhood. The script is historical, and it moves. In the educated urban fields of the moment, the provider script has not been retired so much as merged—Eva Illouz’s emotional capitalism come true—with a fluency requirement: the desirable man is solvent and articulate about the interior weather, competent in the market’s language and in the room’s. Which brings us to New York, where the same man can enter several different rooms.

VII · New York’s Many Rooms

Consider Manhattan first, because the city is large enough to make the market metaphor tempting. In the New York data of Bruch and Newman, desirability is inferred from who writes to whom online. It is a measure of attention at first contact, not of attachment after acquaintance. Ashley Mears’s account of VIP nightlife describes another market, where money buys proximity to beauty. Neither scene is the whole city. A person can cross from a professional reception to a music venue and discover that the same display purchases quite different kinds of notice.

That variation is the point. A title, an address, an expensive evening, and a polished profile are all legible before a stranger knows much else. They can secure an introduction; the research earlier in this essay gives them no power to guarantee affection. Status may buy attention while making a prospective partner wonder whether the display was designed for a short encounter. New York’s scale gives people many chances to begin again, and also many chances to mistake the ease of beginning for the work of continuing.

The city has smaller rooms inside its large one: colleagues, friends, neighbors, people who will see one another again. In those rooms, conduct can be checked. Attention and follow-through become visible over time, and responsiveness ceases to be a line in a profile. This is an editorial reading of the studies, not a measured comparison of New York neighborhoods or scenes. The useful distinction is between being selected for an initial message and becoming someone another person trusts after the message has been answered.

VIII · The Measure

What, then, does the evidence license a man to conclude? Five things, in ascending order of neglect. Keep the animal well: strength, leanness, sleep, grooming, bearing—the floor of vitality is the most tractable tier in the whole architecture, and the only one no field discounts. Earn prestige rather than dominance, in an arena where his particular talents make deference voluntary—and choose the arena with the care the self-improvement books reserve for the self, because the sociology’s plainest finding is that value is field-specific: a man unhappy with his price should first suspect his exchange. Let competence arrive warm—displayed through humor and attention rather than announcement, since the audit for warmth cannot be passed by assertion. Let character compound: kindness, stability, and the kept word are the only assets in the portfolio that acquaintance appreciates rather than discounts. And learn the currencies of New York’s different rooms—legibility at first contact, patience and verifiability among people who meet again.

Then the humility. The trait catalogue predicts the market, and the market is real: it decides who gets the first conversation. But the finding this essay began with has not been overturned—it has been reinforced by every attempt since: no model, given everything measurable about two strangers, can say whether the spark will arrive. Consensus desirability governs exactly that portion of romantic life in which nobody knows anybody, and it dissolves on acquaintance into something pair-shaped and unrankable, assembled in the room out of timing, reciprocity, borrowed adrenaline, and accumulating private history. This is the science’s parting kindness, and it is considerable. The lists can be tended, and should be; they get a man into rooms. But no ranking survives the room itself. Somewhere above the Capilano River the planks are still swaying, the clipboard is still out, and the oldest finding in the field still holds: the heart, loud for any reason at all, will go looking for someone to blame.

Notes on Sources

Joel, S., Eastwick, P. W., & Finkel, E. J. (2017). Is romantic desire predictable? Machine learning applied to initial romantic attraction. Psychological Science.

Trivers, R. (1972). Parental investment and sexual selection. In Sexual Selection and the Descent of Man.

Buss, D. M. (1989). Sex differences in human mate preferences: evolutionary hypotheses tested in 37 cultures. Behavioral and Brain Sciences. N = 10,047.

Walter, K. V., et al. (2020). Sex differences in mate preferences across 45 countries: a large-scale replication. Psychological Science. Cf. Zentner & Mitura (2012), whose gender-parity attenuation the replication largely did not find.†

Li, N. P., Bailey, J. M., Kenrick, D. T., & Linsenmeier, J. A. (2002). The necessities and luxuries of mate preferences. Journal of Personality and Social Psychology.

Sell, A., Lukazsweski, A. W., & Townsley, M. (2017). Cues of upper body strength account for most of the variance in men’s bodily attractiveness. Proceedings of the Royal Society B.

Hitsch, G. J., Hortaçsu, A., & Ariely, D. (2010). What makes you click? Mate preferences in online dating. Quantitative Marketing and Economics. Source of the height–income exchange estimate.

Frederick, D. A., & Haselton, M. G. (2007). Why is muscularity sexy? Personality and Social Psychology Bulletin.

Dixson, B. J., & Brooks, R. C. (2013), on facial hair; Janif, Z. J., Brooks, R. C., & Dixson, B. J. (2014), negative frequency-dependent preferences for beards. Evolution & Human Behavior; Biology Letters.

Puts, D. A. (2010). Beauty and the beast: mechanisms of sexual selection in humans. Evolution & Human Behavior.

Jones, B. C., et al. (2018). No compelling evidence that preferences for facial masculinity track changes in women’s hormonal status. Psychological Science.† (Overturning the ovulatory-shift literature.)

Wedekind, C., et al. (1995). MHC-dependent mate preferences in humans. Proceedings of the Royal Society B.† (Fragile under later meta-analysis.)

Eastwick, P. W., & Finkel, E. J. (2008). Sex differences in mate preferences revisited. JPSP. Eastwick, P. W., & Hunt, L. L. (2014). Relational mate value: consensus and uniqueness in romantic evaluations. JPSP.

Sundie, J. M., et al. (2011). Peacocks, Porsches, and Thorstein Veblen: conspicuous consumption as a sexual signaling system. JPSP.

Snyder, J. K., Kirkpatrick, L. A., & Barrett, H. C. (2008), on dominance vs. prestige in attraction; Cheng, J. T., et al. (2013), Two ways to the top. JPSP; Henrich & Gil-White (2001) for the prestige–dominance distinction.

Schwartz, C. R., & Mare, R. D. (2005). Trends in educational assortative marriage. Demography.

Bressler, E. R., Martin, R. A., & Balshine, S. (2006). Production and appreciation of humor as sexually selected traits. Evolution & Human Behavior. Miller, G. (2000). The Mating Mind.

Gignac, G. E., Darbyshire, J., & Ooi, M. (2018), on the mild attractiveness penalty above the ninetieth percentile of intelligence. Intelligence.

Karney, B. R., & Bradbury, T. N. (1995), neuroticism and marital outcomes. Psychological Bulletin. Arnocky, S., et al. (2017), altruism and mating success. British Journal of Psychology. Kelly, S., & Dunbar, R. (2001), bravery vs. altruism. Human Nature.

Dutton, D. G., & Aron, A. P. (1974). Some evidence for heightened sexual attraction under conditions of high anxiety. JPSP. Willis, J., & Todorov, A. (2006), first impressions in 100 ms. Moreland, R. L., & Beach, S. R. (1992), classroom mere exposure. Festinger, Schachter & Back (1950), propinquity. Whitchurch, Wilson & Gilbert (2011), uncertainty. Aron, A., Fisher, H., et al. (2005), reward systems in early-stage romantic love. Journal of Neurophysiology.

Bourdieu, P. (1986). The forms of capital. Hakim, C. (2010). Erotic capital. European Sociological Review.† (Concept contested, chiefly on field-specificity and gendered convertibility.) Green, A. I. (2008, 2014), the sexual-fields framework. Sociological Theory.

Illouz, E. (2007, 2012, 2019). Cold Intimacies; Why Love Hurts; The End of Love. Connell, R. W. (1995). Masculinities.

Rosenfeld, M. J., Thomas, R. J., & Hausen, S. (2019). Disintermediating your friends. PNAS. Finkel, E. J., et al. (2012). Online dating: a critical analysis. Psychological Science in the Public Interest. Pronk, T. M., & Denissen, J. J. A. (2020), the rejection mindset. Bruch, E. E., & Newman, M. E. J. (2018). Aspirational pursuit of mates in online dating markets. Science Advances.

Guttentag, M., & Secord, P. (1983). Too Many Women? Modern complications: Schacht & Kramer (2016) and successors. Mears, A. (2020). Very Important People. Birger, J. (2015). Date-onomics, for the New York graduate sex ratio. Kenrick & Keefe (1992), age preferences. Behavioral and Brain Sciences.

† marks findings that are contested, weakened by replication attempts, or debated at the level of interpretation. Figures are stylized schematics after the cited studies, not replotted data. Where a number appears in the text—the seventy per cent and the $175,000—it is the cited authors’ estimate, reported as such. Part VII applies the evidence to New York as an editorial argument, not a measured comparison of the city’s social scenes.

Drafted with AI (Claude, Anthropic) as a private research synthesis, August 15, 2026; sources above are real publications, and contested science is flagged rather than smoothed. 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. ◆ ◆ ◆