Do Men Know How Attractive They Are? 478 Self-Ratings vs the First-Glance Read
Our data: men who called themselves 'below most guys' read 102, 'about average' 107, 'above most' 109. One in five of the first group read clearly appealing.
Since mid-October we have asked one question before the photos go in: where do you think you land? Below most guys, about average, above most guys, near the top. Four taps, no number. Then the read comes back and the report puts the two side by side.
Four hundred and seventy-eight men have answered it so far, and the comparison is the most human thing in our data. It is also the least flattering to the mirror.
Barely. Men were right about direction and wrong about size. Those who rated themselves "below most guys" read an average of 102; "about average" 107; "above most guys" 109 — a seven-point spread across three self-ratings on a 70–155 axis. Only "near the top" separated from the pack, at a median of 127, and even there 30% read under 105. Meanwhile 22% of the men who said "below most" read 115 or above, the band where a first glance is clearly appealing.
The usual honesty first. These are aggregated outputs of our read — a model calibrated on first-impression research — across men who chose to be read. Not women's ratings, not a representative sample, no personal data; no group under 30 men publishes an average. What we measure explains the axis. This page is about the gap between what men expected and what the read said.
Analysis first · $5 for the full report · about a minute
The table shows how 478 men's guesses compared with the read. Make your own guess, then let the box check it: one candid photo and a few quick picks, and the report holds your guess against the read side by side.
- Your read on the 70–155 first-impression axis, the six-part face read (jawline, cheekbones, eyes, chin, skin, harmony), your ceiling, how many women notice you, and a dated plan
- The analysis runs before you pay. The full report is $5 one-time — no subscription, nothing renews
- A few quick picks (height, weight, build, how you carry yourself) and one more photo next — the read lands about 40 seconds after you submit
- Photos are encrypted, never public, never used to train models, deletable any time
Key numbers
- Three self-ratings, seven points: "below most" 102, "about average" 107, "above most" 109 (averages). The sample median is 109.
- "Near the top" is the only group that separates: median 127, 44% read 130+. But 19% read under 95.
- 22% of "below most guys" read 115+. One man in five who expected to be below most was clearly appealing.
- 25% of "about average" read under 95, and 7% read 130+ — the widest spread of any group.
- The ceiling barely depends on self-image: 131 for "below most", 133 "about average", 134 "above most", 139 "near the top".
How did men rate themselves, and what did the read say?
The read moved far less than the self-ratings did. Here are the four groups, with the share of each that read clearly appealing (115+) and the share that read below the threshold (under 105):
<!-- data:self-guess -->| How he rated himself | Men | Avg read | Median read | Read 115+ | Read under 105 |
|---|---|---|---|---|---|
| Below most guys | 109 | 101.9 | 102 | 22% | 59% |
| About average | 215 | 107 | 107 | 38% | 47% |
| Above most guys | 111 | 109.3 | 112 | 44% | 39% |
| Near the top | 43 | 119.7 | 127 | 60% | 30% |
Look at the middle two rows. Two hundred and fifteen men said "about average" and 111 said "above most guys"; they read 107 and 109. The men who believed they were above most were, on average, the median man. Not below it — the self-image was not delusional — but not above it either. The read did not find the thing they thought set them apart.
Now the top row. The men who said "below most guys" read 102, seven points under the median. They were right that they sat below the middle. They were wrong about how far: 41% of them read 105 or above, and 22% read 115 or above.
Which way do men get it wrong?
Downward at the bottom, upward at the top, and all over the place in the middle. The cross-tab shows where each self-rating group actually landed:
<!-- data:self-guess-crosstab -->| He rated himself… | Men | …and read <95 | 95–104 | 105–114 | 115–129 | 130+ |
|---|---|---|---|---|---|---|
| Below most guys | 109 | 36% | 23% | 19% | 17% | 6% |
| About average | 215 | 25% | 22% | 16% | 31% | 7% |
| Above most guys | 111 | 21% | 18% | 17% | 32% | 12% |
| Near the top | 43 | 19% | 12% | 9% | 16% | 44% |
Three cells deserve a long look. "Below most guys" → 115–129 or 130+: 22% of these men were clearly appealing, and they walked in expecting to be filtered out. "Near the top" → under 95: 19% of the men most confident in their first glance produced a read that barely registers — the single largest miss in either direction. "About average" → under 95: a quarter of the men who thought they were ordinary were well below the threshold, which is the group most likely to be surprised by the report and least likely to have expected the surprise.
Caveat: the self-rating is a four-option tap made in a few seconds, not a validated scale, and the read is a model estimate. We are comparing one rough instrument with another. The pattern is strong enough to survive that; the decimals are not.
Why is self-rating so weak?
Because men rate their hardware and strangers read their presentation — and those are different objects. The research says the correlation is small; our data says why.
Greitemeyer ran six studies with 1,180 participants and found that self-ratings of attractiveness correlate with strangers' ratings at only r = 0.13–0.32, with the least attractive people the least aware of it and the attractive accurate or slightly underestimating themselves (Greitemeyer 2020). Hönekopp found that private taste accounts for nearly as much of a facial-attractiveness judgment as shared taste — the consensus is real, but it is about half the story (Hönekopp 2006). A man grading himself is grading against his own private taste and a lifetime of mirror angles. The stranger has neither.
There is a more specific mechanism in our data. In 70% of all 1,563 reports the physique read sits five or more points below the face read; in only 20% is it the other way round. A mirror shows a face. A room sees a frame, a posture, a fit, a bearing — and that is where most of the drag is. A man who rates himself "above most" because his face is good is often right about the face and wrong about the glance, which is the whole argument of face or physique: which holds men back. It is also why the camera and the mirror disagree so often, which we took apart in mirror vs photos.
Are men who think they're "below most" actually below most?
A little, and by less than they think — and their ceiling is the same as everyone else's. This is the finding we would most like a man in that group to read twice.
The "below most guys" group read 102 against a median of 109. That is below the middle, yes. It is also seven points, on an axis where the threshold band is ten points wide, and 59% of this group read under 105 — which means 41% did not. One in five read 115 or above.
The number that matters more is the ceiling. The average natural ceiling for men who rated themselves "below most" is 131. For "about average" it is 133, for "above most" 134, for "near the top" 139. Eight points separate the most pessimistic group from the most confident one on where they could land; twenty-five points separate them on where they think they land. Self-image is a poor guide to the read and a worse guide to the headroom — the subject of how much more attractive a man can realistically get.
Caveat: the 478 men who answered are the men who took the test after mid-October, and men who doubt themselves are over-represented everywhere in our sample. The self-ratings are not a survey of men in general.
What about the men who said "near the top"?
They were right more often than any other group — 44% read 130 or above — and wrong more badly when they missed. Nineteen percent read under 95.
This matches Greitemeyer's asymmetry from the other side: the men most confident in their first glance split into those who had earned the confidence and a smaller group whose confidence was coming from somewhere other than the room. We cannot see which is which from the aggregate, and we would not publish it if we could. What the report does, for each man, is name the blind spot: the thing he weights that the stranger doesn't, or the thing he has stopped seeing that the stranger sees first.
What should you do with your own guess?
Make it before the photos go in, write it down, and treat it as a hypothesis the read gets to test. Then read the gap, not the number.
If the read lands above your guess, the information is that you have been filtering yourself out of rooms that would have let you in — and the report will usually show that your face carried more than you credited. If it lands below, the information is almost never "your face is worse than you thought." In our data it is overwhelmingly that the presentation is dragging: frame, fit, posture, hair. Those are the levers in what moves a man's first glance, and they are the cheap ones.
And if the exercise itself is making you anxious — if you are reading this page to find out whether you are allowed to feel okay — stop here. A self-rating is a mood with a number attached. The read is one stranger's first second, modelled. Neither is a verdict on you, and the point of comparing them is to find the lever, not the sentence.
The bottom line
Men know roughly which direction they lean and almost nothing about how far. Three of four self-ratings landed within seven points of each other; one man in five who expected to be below most was clearly appealing; one in five who expected to be near the top barely registered. Self-image predicts the read weakly and the ceiling hardly at all.
Your guess is the mirror's opinion. The read is the room's. The analysis runs before you pay anything; the full report holds your guess against the read, names the blind spot behind the gap, and orders the levers that close it — $5 one-time. Start the test, or see a sample report first.
How this snapshot was built
The figures come from a script that reads the production database and writes only group statistics — counts, means, medians, shares — over scored, completed reports that carried a self-rating. No identifier, photo, free text or per-person value leaves the database; groups under 30 men publish no average. Snapshot dated 2026-11-07, refreshed monthly; the tables regenerate from it. Quote freely with a link (CC BY 4.0).
Studies referenced
- Greitemeyer, T. (2020). Unattractive people are unaware of their (un)attractiveness. Scandinavian Journal of Psychology, 61(4), 471–483. doi:10.1111/sjop.12631
- Hönekopp, J. (2006). Once more: Is beauty in the eye of the beholder? Relative contributions of private and shared taste to judgments of facial attractiveness. Journal of Experimental Psychology: Human Perception and Performance, 32(2), 199–209. doi:10.1037/0096-1523.32.2.199
- Willis, J., & Todorov, A. (2006). First impressions: Making up your mind after a 100-ms exposure to a face. Psychological Science, 17(7), 592–598. PubMed 16866745
Frequently asked questions
Do men overestimate or underestimate their attractiveness?
Both, in different places. In our data 22% of men who rated themselves 'below most guys' read 115 or above — clearly appealing — while 19% of men who rated themselves 'near the top' read under 95. The middle two groups were almost indistinguishable: 'about average' read 107 and 'above most' read 109. The research pattern is similar: see Greitemeyer 2020.
How accurate is self-rated attractiveness?
Weakly accurate about direction, poor about size. Across six studies with 1,180 participants, self-ratings and strangers' ratings correlated only r = 0.13–0.32 (Greitemeyer 2020). In our 478 men, three of the four self-ratings landed within seven points of each other on a 70–155 axis. What we measure explains the axis.
Why do men misjudge how attractive they are?
Mostly because they rate the wrong thing. A man in a mirror judges his face; a stranger reads the whole presentation — and in 70% of our reports the physique read sits five or more points below the face read. Private taste also makes up nearly half of any attractiveness judgment (Hönekopp 2006). See face or physique.
If I think I'm below average, am I?
Probably a little below the median, by less than you think. Men who chose 'below most guys' read an average of 102 against a sample median of 109; 41% of them read 105 or above and 22% read 115 or above. Their average ceiling, 131, was within three points of the men who rated themselves higher — see how much more attractive a man can get.
Does Real World Appeal compare my guess with my read?
Yes. One of the quick picks asks where you think you land; the report then holds that guess against the read and names the blind spot behind the gap. It is one of the most useful parts of the report precisely because the gap is so common. The analysis runs before you pay; the full report is $5 one-time — start the test.