Paul Eastwick says women prefer younger men just as much as men do, they just won't admit it

Women prefer younger men. They just won’t say so. That is the headline finding Paul Eastwick keeps pulling out of his matchmaking data, and it is the kind of result that makes two very different groups of people equally uncomfortable: the evolutionary psychology crowd who built a whole cathedral around male youth-seeking as a biological special case, and the women in the study who explicitly told the matchmakers not to set them up with younger guys, went on the dates anyway, and then asked to see those guys again.

Eastwick is not a contrarian gadfly. He is a careful, slightly nerdy UC Davis professor who has spent twenty years running speed-dating studies, partnering with matchmaking companies, and watching the gap between what people say they want and what they actually respond to widen into something embarrassing. His work does not just poke at one myth. It pokes at the whole architecture: the idea that attractiveness ratings are consensual, that similarity predicts compatibility, that men want younger and women want richer. The data, again and again, say otherwise.

The Number on Your Forehead

Eastwick opens with a classroom demo he runs on undergraduates, the one where students get a random number taped to their forehead and are told to pair up with the highest-value person they can find. People with low numbers get ignored. It is uncomfortable and clarifying. But here is where he pivots from the evolutionary psych script: the demo only works because the numbers are legible. Blur them, and the whole cruel market collapses. In real life, he argues, the numbers get blurry fast. Two strangers evaluating a third person for basic attractiveness agree only about two-thirds of the time, not the near-universal consensus the marketplace model assumes. That disagreement is, in his word, fortunate. It is what lets people who are not consensually desirable find each other anyway.

People might on average say that you’re a six, but if I’ve gotten to know you over time, it means there’s a chance I think you’re a nine. There’s also a chance I think you’re a three.

This is the mechanism underneath his most surprising finding. In the matchmaking dataset, the men searching for dates were older than the women by about four years on average. When Eastwick looked at who wanted a second date with whom, men were a little more interested in the younger women. Not dramatically, but consistently. Then he looked at the women. They were doing the same thing. The women who said do not send me younger men went on the dates, found them interesting, and requested follow-ups. The stated preference and the revealed preference ran in opposite directions.

Sometimes they are like kidding themselves a little bit that they actually do appreciate a younger guy who’s maybe fit and in shape or they don’t fully appreciate how exciting it would feel to be sitting across the table from a guy like that.

What People Say vs. What People Do

The age finding is the splashiest, but the same logic runs through his work on financial status. For decades, surveys showed women caring more about a partner’s earning potential than men did. Eastwick ran speed-dating studies where some women were ambitious, heading toward law or medicine, and some less so, then watched how the men responded. The men liked the ambitious women a little more. Then he flipped it: women liked the ambitious men a little more too. Same direction, same size effect. The gender difference that dominates the online discourse about hypergamy and female selectivity simply does not survive contact with an actual room full of actual humans. It shows up on paper surveys, in online profiles, in the kind of abstract preference-ranking that the apps basically are. It disappears when people meet.

This is Eastwick’s core argument and it is a genuinely uncomfortable one because it implicates the apps directly. Dating apps are, functionally, a preference-ranking machine. They produce exactly the kind of data in which gender differences look huge and compatibility looks like a matching algorithm problem to be solved. His research suggests they are optimizing for the wrong signal entirely. Similarity matching, he notes, lands you roughly at a coin flip for whether two people will actually click. Actual similarity, measured carefully, predicts almost nothing. The apps are also, he points out, the most unequal romantic market anyone has documented: a small number of people capture almost all the attention while regular acquaintance looks nothing like that.

When you look at who gets the right swipes and who receives messages on the apps, it’s the most popular people. I mean, folks have claimed that it’s one of the most unequal markets in the world. But regular acquaintance is not nearly so dramatic.

The Slow Burn Nobody Talks About

Eastwick’s alternative model is almost aggressively undramatic. The typical first impression, he says, is middling. Not a spark, not a gut-punch of attraction. Just: fine. And then you interact again and notice something specific, a way someone handled a tense moment, a piece of banter that lands differently than expected, the lab-world equivalent of watching someone aliquot antibodies with remarkable speed. Attraction, in his data, is usually a slow accumulation of idiosyncratic moments that cause your rating of a person to diverge from the consensus. The dating apps are structurally hostile to this process because they convert the whole thing into a resume review, optimizing for traits when the actual engine is stories.

His prescription, then, is almost retro: small groups, repeated contact, activities that are not structured as interviews. Church, improv class, beach volleyball, a cooking competition in someone’s apartment. Not because these venues filter for compatible values, he is skeptical they do, but because they put people in proximity without the escape hatch of not liking someone immediately. The algorithm gives you that escape hatch constantly. A pickup volleyball game does not. For a researcher whose job is studying how people fall for each other, his conclusion is almost defiantly low-tech: get in a room, stay in the room, and let the numbers get blurry.