89 Comments

I enjoy these data-driven posts, but I always end up a little disappointed that you only present the surface level statistics and don't do deeper analysis to understand what is really going on. Real-world sociological data is always rife with confounding factors and if you don't try to identify and control for them you might end up drawing the wrong conclusions.

For example, you note that women in their 40s only earn ~10% less than women in their 30s, which sounds encouraging, but I suspect there is a survivor effect here, where the number of escorts in their 40s is just significantly lower. If an escort turns 40 and the offers she gets aren't as good as when she was younger, she might well quit, rather than lower her price, but your bar chart doesn't reflect the quitters. A line plotting the number of samples per age group would be helpful here.

I think the real “wall” occurs at the point where participation drops significantly. In any case, it would be a mistake to infer that the typical 35-year-old can expect to continue working ten years later with only a 10% drop in income.

A similar effect might explain why men earn “only” 40% less than women: those men probably represent a higher percentile within their gender, as supported by the fact that there are about 50 times as many female escorts in the data set. It doesn't imply the average man can earn around 60% of the income of the average woman.

Speaking of men, the age graph should be split by gender. It seems unlikely that the effect of age on attractiveness is the same for men and women.

You mention that both ethnicity and body type are highly impactful, but the two are also correlated. Stereotypically, Asians are more likely to be slender and Black women more likely to be fat, which is corroborated by obesity statistics. It would be nice to control for these factors to see how large they are individually. (My guess: BMI is more important than ethnicity, but I could be wrong.)

There are other correlations that you could try to control for. For example, age correlates positively with BMI (and to a lesser extent ethnicity). It could be of practical significance for an escort who is planning her “career” to know if she can realistically maintain her income level if she stays in shape, or if age itself will do her in.

Finally, the height graph should really be split by gender.

Expand full comment

I think what you want is a multivariate regression with income as the independent variable and age, ethnicity, and body type (as well as hair color etc.) as dependent variables. You can even add in interaction terms.

Any statistical package will do this easily; I think you can even do it in excel.

Expand full comment

Agree, there's always more nuances to every data-driven posts.

Expand full comment

Interesting article!

I think one thing that could account for the different preferences/prices for a SW compared to a romantic partner is that it's not so much about what men want *in general* but what they cannot get without paying for it.

In this regard, the preference for 'classy' women (restricted social circles) and taller women (women generally prefer dating taller men, so a shorter man would find it harder to date a taller women) might be reflective of the fact that these are categories that men don't have access to in the dating world.

Expand full comment

Great nuanced perspective, smorg.

What are your thoughts on how societal norms and dating dynamics influence these preferences?

Expand full comment

Oh that's a great point!

I think it could be somewhat complex. If there are some traits which are desired by men but which they would be embarassed to seek out openly then that could create more demand hence upward pressure on prices, but at the same time they might perceive these same traits as being less valuable than the more conventionally 'high status' ones and hence not be willing to pay as much.

Expand full comment

I guess there are some things men will only know 🗿

Expand full comment

Also feels like supply of women in each catagory would be very important. As well as their average reserve price, such as how much of the low price for black escorts is the result of their economic options vs their desirability?

Expand full comment

Agree, so many other hidden factors at play here.

Expand full comment

I can't get past that first factoid about the last six inches of 6' in height being worth $175k of a man's income. Reduced by an order of magnitude ($17.5k), that I would believe. I mean, trading in that amount of income for some additional height would plunge me far, far below zero income (which I don't think would be a good deal for a hypothetical mate!), and that's probably the case for most men I know honestly. I think of any income above $120k or so as far enough into the stratosphere that it's hard for me to imagine a woman substantially caring how much more than that a guy makes. Then again, in some geographic areas (like CA, I imagine) it's different.

Expand full comment

I think it really depends on what it'd consider the base rate income? Like I can imagine many women happily going from a 5'6" man with a 675 000 salary to a 6'0" man with a 500 000 salary. I can't see many women going from a 5'6" man with a 180 000 salary to a 6'0 man with a 5 000 salary

Expand full comment

Yes, it’s not a linear curve. Obviously a 6 ft guy who makes 1.8 million a year is preferred to a 5’6 one who makes who makes 2 million. In the 200-400k range I could see this being true.

I suspect many men don’t realize just how ugly short men are to women. It’s probably similar to the question, under what circumstances would a guy date woman with a BMI of 40.

Expand full comment

No circumstances? I don't know billionaire? I am terrible... :(

Expand full comment

I wouldn't interpret that result literally. I think it is useful for gaging the relative attractiveness of various things but would discount the stated magnitude. If real world outcomes matched women's stated preferences nobody would get married and the species would be going extinct. I think a lot of women have very high stated requirements when asked about a hypothetical partner (finance, trust fund, 6'5, blue eyes 🤣) and much more realistic revealed preferences when it comes to actual men that they know.

Expand full comment

I feel sorry for shorter men 🤷🏻

Expand full comment

I think you underestimate how rare a commodity 6'0" people actually are compared to demand. Like, the famous meme "6'0", 6 figures", if I recall correctly, covers less than a promille. So basically, that finding is women saying that the first part of that meme matters more than the second.

Expand full comment
Jul 22·edited Jul 22

It's from data about online dating, so men are probably lying about their height so women are adjusting for that, and I expect the effect of income is greatly reduced compared to long term relationships. Also, the baseline income for men in the study was 62k.

Expand full comment

This study was published in 2006 so these incomes would be even more crazy now.

Expand full comment

Any study that asks women to contemplate a hypothetical is worse than worthless.

Expand full comment

And to think I was wasting my time and the fatness of my titties at age 22. Oh, the things I didn't know then that I know now.

Expand full comment

What do you mean?

How did it affect you then?

Expand full comment

I never marketed them. I would be so much closer to retirement had I been privy to all these handy dandy charts and statistics about their value.

Expand full comment

I think price has an inverse relationship to establishing semi-permanence or a "regular John" relationship. Men seek novelty - and men become visually fixated on certain women to the point that they will pay a fairly large amount for one sexual encounter. An escort who is really supporting herself with sex work needs to immediately drop her price to "special price just for you" with any acceptable, well-behaved new client, or she risks having him disappear after 1 date and maybe call her once a year. Very interesting if you explore providers' "Regular Price" for guys she's already validated and vetted.

Expand full comment

So dang true.

Kirk gets it.

Expand full comment

True, but it still gives us an idea of market value.

Expand full comment

Something else that caught my eye:

> In my data, the ‘ethnicity’ category was a little sus. [..] e.g., ‘white’ had meaningfully different income from ‘caucasian.’

This doesn't surprise me at all, and it ties nicely into what I mentioned before about confounding factors. “White” and “Caucasian” may imply the same race but they are not used by the same people.

With the caveat that I am not an American, it's my understanding that Caucasian has become an obsolete term for white people in America. High-status white people today call themselves ”white” instead, just like high-status Black people increasingly call themselves ”Black” instead of “African American” (and if they're really with it they use that exact capitalization).

So a person who self-identifies as Caucasian rather than White is likely old, out of touch, or a foreigner who isn't acclimated to American culture. All good reasons to expect that “Caucasians” are materially different from “Whites” despite being technically the same race. I bet you'd see similar differences with “African American” vs “Black”.

Expand full comment

Maks, that's an interesting observation about the differences between "white" and "Caucasian."

It makes sense that these terms can indicate more than just race but also cultural and social nuances. It's fascinating how language evolves and reflects our identities and statuses. Insightful analysis, thanks!

Expand full comment
Jul 17·edited Jul 17

Are you aware that most of the data you're working with here is worthless? LIterally a lot of these ads if not most, especially the lower range ones are just scammers using stolen pictures from webcam models or instagram influencers. I also don't like how you sliglthy entertain the idea (made up from a moral panic intended to destroy sex worker rights, among other things) that some escort ads are a font for sex trafficking, which is bullshit unless perhaps you're using a very broad definition of "sex trafficking", most of those ads are just scams. Also price range, especially in the US, has a weak correlation with demand or desirability. The alleged desirability of fat titties is a cultural meme with little basis in reality, there is no biological basis to it and most men don't care much about large breast size. The most desirable and sought after prostitutes, as told by

many working in the industry, are usually those that look the youngest, not the ones that have the largest tits or amount of body modifications.

Expand full comment

Not sure I agree, what historic/current art and now AI created images come up with tells you a lot about men’s desires (titties).

And even in my own life I think about say my wife, who is super cute and capable, in great shape, has aged like wine, and has a great ass. Lucky me. But she also has like a A+/B- cup. And I think most people in day-to-day life would call her a 7 or 8, not a 10. (Depending on where you live).

She evens looks a lot, especially body wise, like a lot of famous models and actresses…what is the main difference? The small breasts.

She is born with natural full Cs and she is suddenly a 9 or a 10 in a lot of men’s eyes and in all honesty we probably never get married because she has snagged some higher quality partner before she ever ran into me.

A lot of women considered *very attractive* honestly are just kind of above average but with way above average breasts. I don’t think that is some strange coincidence.

Though I do agree youthful appearance is also at a huge premium. As is facial symmetry/femininity.

Expand full comment
Jul 17·edited Jul 17

I think you're exaggerating the effect of breast size on attractiveness, if that was true then petite women would be considered ugly and that's obviously not true, most people would actually consider a beautiful petite much more attractive than a average woman with large breasts. Also historic art actually points to large breast fetishization to be relatively recent western cultural meme as I mentioned, and scientific studies point to men having a preference towards moderate size as opposed to very large tits which would always sag, and firmness/perkyness is a lot more important than size alone.

Expand full comment

Oh I am 100% on board the anti-sag train. But people aren't long term planners generally, especially about 10-15 years out. Your average 22 year old isn't thinking a lot about what their partner will look like in 15 years, or even in 5 after a kid.

My HS sweetheart had one of those curvy bodies that looks like a goddess at 18 and totally pedestrian at 30. Extra fat in all the right places at 18 becomes way too much fat at 30 in a lot of cases.

But hopefully what people are looking for in a long term mate extends pretty far beyond just how hot they are. A lot of hot people are pretty incompetent and/or emotionally a lot to handle.

Expand full comment

> if that was true then petite women would be considered ugly and that's obviously not true

That doesn't follow from what he said. Breast size being a significant factor in attractiveness doesn't mean it's the only factor. That's like saying that all short men are unattractive because women find height important.

> to the point where they would always sag, and firmness/perkyness is a lot more important than size alone.

That's why this data indicates youth is also important. Large breast size on a young woman doesn't suffer from sag yet.

Also, don't mistake what men prefer in a long-term partner with what men prefer for a short-term tryst. We're talking about the latter here. It could be the case that large breasts are preferred for trysts (novelty, fun, etc.) but not long-term partners for various reasons.

Expand full comment
Jul 17·edited Jul 17

I should've worded that sentence better, what I meant is that petite women are generally not considered significantly less attractive. Compare that to height where men below a certain threshold (I imagine 5'7 in the US at least) are considered unattractive in general, if not undateable. Similar effects can be seen with deviations in symmetry, youthfulness, body weight, etc but not with breast size. Being even completely flat is not enough to make a woman undesirable, especially if they have a proportional body, which means that breast size doesn't that much of an effect on attractiveness.

Expand full comment
Jul 17·edited Jul 17

> Compare that to height where men below a certain threshold (I imagine 5'7 in the US at least) are considered unattractive in general, if not undateable

No, this is just not the case. It looks like the same kind of effect as with breast size: some percentage of men/women who find that particular feature very compelling are no longer interested in people with small breasts/short stature, but those that remain simply don't prize it as much.

I agree that breast size for women is not *as* important as height for men, but it's still a significant factor. Like, it's been documented extensively in many studies that women with larger breasts attract more sexual interest and Aella's data on this is not surprising at all.

Let's consider a question: if you take a woman that's ugly, thin with small breasts and give her breast implants, will she attract significantly more male interest?

If you agree that she would, as is common knowledge and has been documented extensively in literature, do you agree that the increase in male interest indicates an increase in desirability / attractiveness?

If so, I frankly don't see how you can conclude that breast size does not have much of an effect on attractiveness. If not, how are you separating obvious and measured increases in male interest from desirability and/or attractiveness?

Edit:

> Being even completely flat is not enough to make a woman undesirable

This is what I mean by confusing claims. It's true that a woman that is flat-chested but is otherwise attractive will not have trouble finding men that still find her attractive, but that doesn't mean she wouldn't have *even more* people interested in her (and even more attractive people) if she had large breasts. There's just so much clear evidence that this is true.

Expand full comment

There are many many female athletes who are flat as a pancake but are 9 or a 10 for rest of their figure particularly their pins.

Expand full comment

Men talk about women's breasts.

Me see.

Me like.

🗿

Expand full comment

The data strongly matches my priors regarding weight and race, and boob size itself for that matter. I see no reason to doubt it unless you have alternative strong evidence

Expand full comment

John, an excellent counterpoint, with a certain nuance truth.

What are your thoughts on how to improve the reliability of data in studies like this, given these certain challenges?

Expand full comment

Thanks. I would recommend not taking data from escort ads in general, they are extremely unreliable because when they are not scams, there's the fact that a lot of women lie on those ads (catfishing). Some escort ad sites are at least more reliable than tryst, having photo verification and client feedback. Generally using client feedback would greatly improve reliability in this case.

Expand full comment
author

I don't think tryst has many scams. You do have to do photo verification for tryst, and it's pretty annoying to actually get approved to get listed on tryst. I also have manually looked through lots of diff websites and tryst seems pretty scam-free to me.

Expand full comment

Yes John, but then again even clients can lie. Do you think it's best to rely on biology and history, as to what never change?

Expand full comment

It's interesting that the increase in prices from "average" to "super hot" isn't that dramatic on Tryst. A girl I could have sex with on any random weekend by opening Tinder would be around $300 on Tryst. The hottest girl I've ever slept with would be around $600. A drop-dead gorgeous girl that wouldn't even look me in the eye is something like $1200.

Expand full comment

Dude, what market are you in?

Expand full comment

Why does the word cloud include "alway"?

Expand full comment

Looks like someone tried to merge singulars with plurals xD

Expand full comment

What's interesting, despite all the hoo-ha about evo-psy and hypergamy etc., is that, if you look at marriages (presumably, the end point/purpose of a dating career), social stratification/associated mating swamp all other factors.

https://www.demographic-research.org/articles/volume/44/7

https://www.tandfonline.com/doi/full/10.1080/14616696.2023.2290238

So, all that physical stuff is interesting but it's really just "while waiting" to marry - where your class/educational attainment will be the major factor, apparently.

Expand full comment

People were talking in a different thread elsewhere about mismatched couples and I know well (our kids are good friends) a doctor husband and wife where he looks like a chubbier, jewisher Paul Giamatti, and she looks like Iliza Shlesinger. They make for a very mismatched couple, but like you said assortive mating.

They met in medical school, and I suspect since there were both young the difference wasn't as striking and she (as a woman) probably wouldn't care as much anyway generally. I doubt you would see the opposite matchup very often though.

Expand full comment

Well, assorted mating suggests that, for the man, faced with the choice of a prettier but lower class mate and a less attractive but socially in-line one, he'll go for the second option (most of the time) - at least when it comes to marriage.

A somewhat separate issue but I kind of noticed a certain reticence from working class pretty girls (in the UK) to dating higher class guys (unless they were good talker/"had game") - I suspect they can feel the guy is just looking for a nice shag but won't be willing to commit.

So IDK. Assorted mating is probably something both sexes do.

Expand full comment

Great stuff! Submit to all known sociology journals immediately!

Expand full comment

They will likely reject it as it doesn't serve their agendas.

Expand full comment

It'd be interesting to see this for other countries where the rates are way cheaper

Expand full comment

Those rates are crazy high compared to England. But escorts are legal there so perhaps more competition.

Expand full comment

The summary makes me picture natalee.007

Expand full comment

A bombshell for sure 🗿

Expand full comment

Nonlinear age impact (where taller = better up until she's as tall as men at which point height rapidly gets considered worse) squares well with what I've seen from other sources. It seems to me that men's preferences for women is "tall, but shorter than me".

Obviously it's not available in this escort data, but I sometimes wonder if partner height preferences correlates with other variables. (Sexual dominance/submissiveness? Attraction to androgyny? Gender progressivism? All of these seem stereotypically relevant for height preferences, though less clear whether they are relevant in practice.)

Expand full comment

I think there may be a difference here between what men prefer in a partner and what women are able to charge as an escort.

Other studies have shown men prefer to date women in the 5'0-5'8 date range. I think this is a case of power dynamics. If we accept the premise that men prefer to be in a position of dominance over their partner, then a shorter woman would be more desireable as they would have less physical presence. However, in the case of escorting, paying a woman for service is itself an act of dominance, hence a taller woman may give a greater sense of empowerment to a man.

Expand full comment

Which "other studies" are you referring to?

Expand full comment
Jul 17·edited Jul 17

The one cited in the post:

https://home.uchicago.edu/~hortacsu/onlinedating.pdf

This study also shows that men prefer women in the 5'4 range for short-term dating and 5'6 for long-term dating, with a significant drop off for 5'8 and above. Although for an individual man it also depends on their own height.

https://www.frontiersin.org/journals/psychology/articles/10.3389/fpsyg.2022.937146/full

Expand full comment

I'm exactly 5'9" and not surprised by the findings. There's two things: first, height is associated with class in most cultures, and rich people are generally taller...in some cases historically this has been an enormous discrepancy with rich people being a good 6inches or more taller than poor, on average. So it goes into the "classy" element and fits the higher prices.

More importantly, for whatever reason there's a clear sexual selection evolutionary preference for height and being tall (say 90-99th percentile) but not freakishly tall (99.9th percentile). If this is the case, it makes sense that both men and women would prefer tall partners bc subconsciously that means better chances of having tall babies. But there's also a cross cultural norm for the man to be taller. The average height difference between male and female children with the same parents is 4". This means the male preference should be for women just slightly shorter than they are (thus any babies would preserve his height genes if he's tall or ramp them up if he's not). And the preference for women would be for a man much taller. Too lazy to google it, but the studies I've seen are that women on average want a guy 7" taller and men on average want a girl 2" shorter. That creates a mismatch, obviously, but also fits what you would expect if you assume people have a subconscious drive to get tall babies.

The messed up part is that short girls are even MORE interested in very tall men, in my experience, probably bc they really want those tall genes for babies. I've known more 5'0" chicks who refuse to date someone under 6' than girls my height.

Expand full comment

I should definitely send this article to my ex, a New York-based 5'9", thin, green-eyed blonde in her 20s with an erudite vocabulary and big boobs.

(Awful relationship, by the way - wasted at least 4 years of my life with her out of pure inertia and I stagnated a ton).

Expand full comment