The Mathematics Of Beauty

January 10th, 2011 by Christian Rudder

This post investigates female attractiveness, but without the usual photo analysis stuff. Instead, we look past a woman's picture, into the reaction she creates in the reptile mind of the human male.

Among the remarkable things we'll show:

Fair warning: we're about to objectify women, big-time. The whole purpose of this blog is to analyze OkCupid's data, and without a little bit of objectification that's impossible. Men will get their turn under the microscope soon enough. As usual, none of this (with the exception of the celebrity examples) is my opinion. All data is collected from actual user activity.

Let's start at the beginning.

All people, but especially guys, spend a disproportionate amount of energy searching for, browsing, and messaging our hottest users. As I've noted before, a hot woman receives roughly the messages an average-looking woman gets, and 25× as many as an ugly one. Getting swamped with messages drives users, especially women, away. So we have to analyze and redirect this tendency, lest OkCupid become sausageparty.com.

Every so often we run diagnostic plots like the one below, showing how many messages a sampling of 5,000 women, sorted by attractiveness, received over the last month.

These graphs are adjusted for race, location, age, profile completeness, login activity, and so on—the only meaningful difference between the people plotted is their looks. After running a bunch of these, we began to ask ourselves: what else accounts for the wide spread of the x's, particularly on the "above-average" half of the graph? Is it just randomness?

What is it about her:

that gets more attention than her:

...even though according to our users, they're both good-looking?

Not all 7s are the same

It turns out that the first step to understanding this phenomenon is to go deeper into the mathematically different ways you can be attractive.

For example, using the classic 10-point 'looks' scale, let's say a person's a 7. It could be that everyone who sees her thinks exactly that: she's pretty cute.

But something extreme like this could just as easily be going on:

If all we know is that she is a 7, there's no way to tell. Maybe for some guys our hypothetical woman is the cat's pajamas and for the rest she's the cat Garfield. Who knows?

As it turns out, this distribution of opinions is very important.

Celebrity photos: to titillate and inform

Let's look at what the ratings distribution might be for a couple famous people. I imagine that for, say, the actress Kristen Bell it would be roughly like this:

Ms. Bell is universally considered good-looking, but it's not like she's a supermodel or anything. She would probably get a few votes in the 'super hot' range, lots around 'very attractive', and almost none at the 'unattractive' end of the graph.

Compare her to Megan Fox, who might rate like this:

On the far right, you have the many dudes who think she's the sexiest thing ever. On the far left, you have the small number of people who have seen her movies.

Unlike Ms. Bell, Ms. Fox produces a strong reaction, even if it's sometimes negative.

Real People

Now let's look back at the two real users from before, this time with their own graphs. OkCupid uses a 1 to 5 star system for rating people, so the rest of our discussion will be in those terms. All the users pictured were generous and confident enough to allow us to dissect their experience on our site, and we appreciate it. Okay, so we have:

As you can see, though the average attractiveness for the two women above is very close, their vote patterns differ. On the left you have consensus, and on the right you have split opinion.

To put a fine point on it:

  • Ms. Left is, in an absolute sense, considered slightly more attractive
  • Ms. Right was also given the lowest rating 142% more often
  • yet Ms. Right gets as many messages

When we began pairing other people of similar looks and profiles, but different message outcomes, this pattern presented itself again and again. The less-messaged woman was usually considered consistently attractive, while the more-messaged woman often created variation in male opinion. Here are a couple more examples:

We felt like were on to something, so, being math nerds, we put on sweatpants. Then we did some work.

Our first result was to compare the standard deviation of a woman's votes to the messages she gets. The more men disagree about a woman's looks, the more they like her. We found that the more men disagree about a woman's looks, the more they like her. I've plotted the deviation vs. messages curve below, again including some examples.

The women along the graph are near the 80th percentile in overall attractiveness. You can click the tiny thumbnails to expand them.

As you can see, a woman gets a better response from men as men become less consistent in their opinions of her.

Our next step was to analyze a woman's actual vote pattern of 1s, 2s, 3s, 4s, and 5s:

If You're Into Algebra

We did a regression on the votes for and messages to a sample of 43,000 women. To keep everything consistent, all the women were straight, between the ages of 20 and 27, and lived in the same city. The formula given in the body of the post was the best-fit we found on our second regression, after dropping the m3 term because its p-value was very near 1.

msgs are the number of messages the woman received during the observation period. The constant k reflects her overall level of site activity. For this equation, R2 = .28, which isn't great in a lab or on a problem set, but is actually very good in a real-world environment.

This required a bit more math and is harder to explain with a simple line-chart. Basically, we derived a formula to predict the amount of attention a woman gets, based on the curve of her votes. With this we can translate what guys think of a woman's looks into how much attention she actually gets.

The equation we arrived at might look opaque, but when we get into it, we'll see it says some funny things about guys and how they decide which women to hit on.

The most important thing to understand is that the ms are the men voting on her looks, making up her graph, like so:

And those ms with positive numbers in front contribute to messaging; the ones with negative numbers subtract from it. Here's what this formula is telling us:

The more men who say you're hot, the more messages you get.

How we know this—the .9 in front of m5 is the biggest positive number, meaning that the guys who think you're amazing (voting you a perfect '5') are the strongest contributors to your messaging income. This is certainly an expected result and gives us some indication our formula is making sense.

Men who think you're cute actually subtract from your message count.

How we know this—because the .1 coefficient in front of m4 is negative. This tells us that guys giving you a '4', who are actually rating you above average-looking, are taking away from the messages you get. Very surprising. In fact, when you combine this with the positive number in front of the m1 term, our formula says that, statistically speaking:

If someone doesn't think you're hot, the next best thing for them to think is that you're ugly.

This is a pretty crazy result, but every time we ran the numbers—changing the constraints, trying different data samples, and so on—it came back to stare us in the face.

What We Think Is Going On

So this is our paradox: when some men think you're ugly, other men are more likely to message you. And when some men think you're cute, other men become less interested. Why would this happen? Perhaps a little game theory can explain:

Suppose you're a man who's really into someone. If you suspect other men are uninterested, it means less competition. You therefore have an added incentive to send a message. You might start thinking: maybe she's lonely. . . maybe she's just waiting to find a guy who appreciates her. . . at least I won't get lost in the crowd. . . maybe these small thoughts, plus the fact that you really think she's hot, prod you to action. You send her the perfectly crafted opening message.

"sup"

On the other hand, a woman with a preponderance of '4' votes, someone conventionally cute, but not totally hot, might appear to be more in-demand than she actually is. To the typical man considering her, she's obviously attractive enough to create the impression that other guys are into her, too. But maybe she's not hot enough for him to throw caution (and grammar) to the wind and send her a message. It's the curse of being cute.

The overall picture looks something like this:

Finally: What This Could Mean To You

I don't assume every woman cares if guys notice her or not, but if you do, what does all the above analysis mean in practical terms?

Well, fundamentally, it's hard to change your overall attractiveness (the big single number we were talking about at the beginning). However, the variance you create is under your control, and it's simple to maximize:

Take whatever you think some guys don't like—and play it up.

As you've probably already noticed, women with tattoos and piercings seem to have an intuitive grasp of this principle. They show off what makes them different, and who cares if some people don't like it. And they get lots of attention from men.

But our advice can apply to anyone. Browsing OkCupid, I see so many photos that are clearly designed to minimize some supposedly unattractive trait—the close-cropped picture of a person who's probably overweight is the classic example. We now have mathematical evidence that minimizing your "flaws" is the opposite of what you should do. If you're a little chubby, play it up. If you have a big nose, play it up. If you have a weird snaggletooth, play it up: statistically, the guys who don't like it can only help you, and the ones who do like it will be all the more excited.

427 Responses to “The Mathematics Of Beauty”

  1. TimPhillips says:

    Very interesting read. But I don’t really know about advertising flaws…I can’t say I’m too terribly excited to see someone hanging out of a tanktop…LOL.

  2. Fred says:

    Post gender comparison stats or GTFO!

  3. frogomatic says:

    Have you considered the possibility that ratings of 4 might be affected by the fact that this is one of the two rating which can be used to see if someone has picked you on Quickmatch? The number has special semantics (along with 5, though, I doubt people would use 5 ratings in attempting to game the system to see who quickmatch-picked them).

  4. Melissa says:

    I’ve got to chime in and agree with many posters that:

    1. It looks to me like the women getting more responses look like “more likely to put out”.

    2. I never even thought of using the rating system to rate just looks. I rate “4” for if they seem alright but there is a red flag and “5” if there are no red flags and only good things in the content of the profile.

  5. Pops says:

    Just so you all know, but not that any of you are going to read this, these results are NOT based on your quickmatch rating, they were based on a voluntary, photos only survey taken a few months ago. Don’t confuse yourselves, they aren’t ignoring data, your just adding useless info.

  6. dave eslinger says:

    how about a beauty contest? you take willing female participants, have staff and clients vote, aiming for the top 100? let clients make suggestions for the top ten?

  7. eavil says:

    I wonder how much to do with the hot/not aspect is down to quality of the photo… I’m noticing a lot of the ‘hot’ girls have more portrait-style shots and less candid ones. Even a bit of photoshoppery, increased contrast, lightened skin, soft lighting etc etc.

  8. Kyle says:

    Why go for an 8 when you can have four 2s.

  9. titsieroll says:

    I agree with Garrett. guys go for ugly girls because they think better and easier chance to get laid.

    They go for the hot ones, because they think they are stupid or clueless enough to agree to sex.

    dont need a protractor and e=mc2 to figure that out. just need to have lived in a populated area through puberty to understand this concept

  10. Tae says:

    Great article, guys! All these dummy nay-sayers who are trying to rationalize the results are obviously ignorant of what econometrics is. Personally, I would have liked to see a more serious/academic version of this study (maybe just a little link to a PDF), including an exploration into the possible problems with your model. But considering the nature and scope of this blog, I think you guys did a great job. R^2 = .28 should be a good enough reason for most people to give it some thought and at least be entertained.

    Also, I can understand that you guys need to make an effort to cater to a broad audience, but times-are-a-changin’, and for every 10 dummies you alienate with your baby-drowning jokes and objectification of women, you gain the respect and loyalty of at least one pretentious pseudo-intellectual such as myself! :D

  11. Makbuth says:

    titsieroll, i think that’s just what you a few guys do, so don’t lay that blanket statement on all people. Not everyone is as shallow as you.

    Also, this entire article is really crappy. It leaves literally NO room for the possibility that a man might ACTUALLY rate girls not on their looks. Believe it or not, men are not the shallow buckets America makes them out to be. They have emotions too, and the only way to make a relationship work, is not looks, but compatibility. Sure, being good looking is a perk and helps, but christ, it’s not an end all be all. Next time maybe you should actually as the people why they rated the girl/boy the way they did instead of throwing rocks on the ground and guessing what they mean.

  12. interweb_persona says:

    Guys go for hot women way out of their league because this service caters to voyeurism and fantasy, and the men have nothing to lose. The reason men do it is because it’s safe to- the sting of being rejected or ignored by someone who wouldn’t give you the time of day anyway is dulled by the wall of the computer screen and the ability to hit on hundreds more within seconds.

    There’s a tinge of Nazi in this plan, but what if the site filtered ALL searches on the basis of a mutual level of attractiveness? It’s flawed from the get go, but it would create an incentive for men to not use this site as a place to whack off their ego, and would curb the tendency for hot women to use this site as a vanity mirror. Hot girls don’t USE internet dating. They use it to absorb attention wholesale- they’re like the farmers of the internet dating world, harvesting the men for messages.

    I think if you were to come up with an algorithm that would only allow for 10% of more attractive women to appear in the radar of the average male, 10% less attractive, and 70% roughly within the same measure, (this goes for both immediate matches AND personal searches) you’d have a more fluid service. The top percentile of attractive women would cease to congest ALL male attention, and thus the men would pursue more realistic matches feeling as though they have encountered the most attractive person available to them.

    Perhaps this kind of social engineering is unethical, but I bet it would yield very interesting results. You’d have to implement it slowly and systemically, otherwise men would go WTF all over the place, but even then, it would filter out the men who only use the site to whack it.

    I’d say you should do the same for women too, but the odds of them reaching out anyway would probably still be very low.

  13. Jasmeene says:

    Interesting article… even more interesting replies ;-)
    All in good spirit… I hope :-)

  14. Bryan says:

    reptile minds. i definitely dont like the tone of that.

    aside from that, game theory works because of what you know about others. she is a smoker and you are ok with that? well you know thats an exclusion principle for other people.

    it’s not about being hot. it’s about percieved availablitiy. if a guy thinks you are more available and approx. similar attractiveness guess who will get more messages?

  15. Guy in SF says:

    I’m surprised at how hetero-normative this study is. There’s no discussion whatsoever of the behavioral differences of men who like women versus women who like women. I suspect there *are* differences, but if there aren’t I would at least like them to say that.

    I’m curious to see the results of the “men’s turn,” but it will (sadly) be purely academic for me as a gay man. Unless you report on how men who like men differ from (or are the same as) women who like men, I won’t get anything valuable to me personally out of the study.

  16. Darwin says:

    Maybe the causal relationship is purely sociobiological?

    The women I find attractive are not classically beautiful because I’m being driven (against my better judgment and that of my peers) into the arms of those for whom I have a better chance at a viable contribution to the gene pool? Whether ‘I’ personally want children or not! Maybe they just ‘seem’ more fertile.

    The low hanging fruit for those of us cruelly snubbed by genetics and look more like Crispen Glover than Tom Selleck (and those who’re vertically challenged).

    But then where does that leave me and my super-fantasy girlfriend Megan Fox?

    Maybe you can provide me with a formula to determine the one’s ‘I’ find ‘beautiful’ from those whom ‘I’ find merely ‘attractive’?

    Perhaps that formula is BEER.
    Yes. send me beer!

  17. Deb says:

    Hahaha…. any volunteers for a study like this will be like those typical survey respondents who wither have something really good or really bad to say. These stats and numbers seem to be made up, as well as the graphs. I’d like to see the actual database for this “research.” Also, trying to explain the statistics behind this makes it seem almost like you’re trying too hard to empirically support an opinion. Different types of girls will appeal to different types of guys. Sure, there are those who will go for the less attractive women because they’re primarily looking for an easy in, which make results like these possible, but I’d like to think that this formula can’t necessarily be applied to the entire population of male OkC users.

  18. Cate says:

    “1. It looks to me like the women getting more responses look like “more likely to put out”.”
    —–
    This!

  19. K says:

    Second what eavil said. I can’t believe a potential factor as obvious as this wasn’t controlled for. All the more frequently messaged users have exponentially better photos (in the various aspects detailed by eavil)

  20. stormy says:

    I’m not so sure about your analysis of this data. First, when I was on here I was using the stars as about the whole profile, more about their potential datability, than just looks. But even if we think you’re analyzing the attractiveness of their profiles rather than their pictures, I don’t buy your analysis. I don’t think guys are looking at a girl and thinking ‘oh, some people think she’s ugly so there’s less competition’. Presumably its the very guys who rated her a 5 that like her. So its they actually find her really attractive. I don’t think the that some people thinking she’s ugly make them like her more. Certainly, when I look at Megan Fox, I don’t think “some people think she’s ugly (this is first i’ve heard of that phenomenon, btw) and therefore i think she’s hot.” I think the variance in ratings is because you want someone unique and distinctive, and those unique and distinct things are very polarizing, people either love them or hate them. But they don’t love them because some people hate them.

  21. a_pineapple says:

    A couple things I saw in this:

    – people who vote with a 2, 3 or 4 are really just voting neutral, and probably not motivated to follow through with a message. (The behavior seems similar to what the concept of Net Promoter Score makes use of)

    – people who get 1’s and 5’s are highly attractive for some particular niche trait. Within a niche, people are probably much more likely to message each other. If someone doesn’t fit in your niche you’re probably much more turned off.

  22. Fergal Daly says:

    If you had look at how men behaved and correlated messages sent to how they voted on a particular woman then you could decide whether men send messages to women they find unattractive or whether they just find different women attractive and then message them.

    Showing off your snaggle tooth will probably lower your avg score and yes, maybe you will get more messages than average for someone at that lower score but will you get more messages overall?

  23. grnanrkist says:

    i wonder if there will be any uncertainty phenomena as a result of this. perturbing the observables. Hf=Ef

  24. Kevin says:

    @Pops wrote:
    Just so you all know, but not that any of you are going to read this, these results are NOT based on your quickmatch rating, they were based on a voluntary, photos only survey taken a few months ago. Don’t confuse yourselves, they aren’t ignoring data, your just adding useless info.

    Thanks for the information, I’m wondering where you got it…? And I disagree with the last part: users did not confuse themselves. The blog post created the confusion by not clearly stating where the looks ratings came from.

  25. BB says:

    I wanna see my stats in those graphs. Is there an option for that? (new to okcupid)

  26. Scott says:

    When it comes to women, they can very much control their own beauty.

    It comes from simply finding their niche.

    As mentioned women with tattoos are a great example. What it all really comes down to is confidence and being a woman.

    Men like women, and sure that sounds like “duh” but a lot of women I believe forget this this simple little concept. Women are the source of beauty in human society. Beauty is derived from organization. Organization meaning proper body weight and proportion, hair styled, make-up done, proper fitting clothes.

    Women who do not care about these attributes are the same as men who do not care about their jobs, lifestyle, and purpose.

  27. vibe says:

    A more correct interpretation is that people tend to exaggerate their dislike for the appearance of those other people think are hot but they don’t.

  28. Zanderilicous says:

    Well I always read some of the profile before voting in quick match…. but it doesn’t matter, really. I never, ever get a response back from the people I message. Either this site is fake and no one really gets the email or the ratio on this site is like 100 guys to every 1 girl… I have had ZERO luck with this site and I am starting to think that it’s FAKE…. I have done and paid for all the rest of the dating sites out there and I am just about done with it all. I am starting to think that it’s like santa claus, you will never ever meet someone in real life from a dating site, but thanks for keeping this one running. Maybe one day my luck will change?

  29. athensfemme says:

    I thought the statistics were solid, and the use of game theory was excellent! This is one of the more solid essays I’ve seen in regard to the math behind the rhetoric.

    I also agree with others, (though I am a woman and therefore my opinion is biased). It appears that OKC users prefer edgier/risk seeking women. The average looking women have higher rankings but the people who rank them as attractive are likely to move on to match.com and get a better variety of that same kind of woman. Meanwhile, the edgier looking women have the OKC frequent messengers all to themselves…

    I didn’t like the use of only one race/ethnicity.

  30. Thornbob says:

    I didn’t know the “Rate Her” thing was just for appearance. Your research may be somewhat thrown. I saw a girl once and rated her 4 because she was Perfect with a capital P in my opinion, though I know for a fact other guys wouldn’t be into her looks. I did know that many of her values were in direct conflict with mine. I’ve rated a few other girls “5” who were not nearly as attractive (how can you compare with perfect, anyway?) because I thought they were a perfect fit for me.
    I suppose most men do use it just for an appearance rating, though.

  31. Normal says:

    Can you guys write a book or something? I want to read this all the time, everywhere.
    This is fascinating on so many levels. I’ve always just instinctively chosen my weirdest photos, since I figure anyone who would write me after looking at my actual-life-photos (not some idea of what they want me to look like) has already passed the boiling oil test and I’ll never have to justify why I’m at a dive bar in a chicken suit sucking white russians through a bendy straw.
    If I’m lucky, I might even find someone who wants to wear a powder blue leisure suit and hang out with me.

  32. Max says:

    I am definitely looking forward to the male equivalent of this analysis.
    Also looking forward to some similar analysis of the text of the profile for men!

  33. porpoise says:

    The rating system is flawed. The point of it for the user is to try to match with someone. I’ve already said this before in public forums. The only numbers that matter are 4/5 and not 4/5. That’s how it works for the user. yes or no. I’m not sitting there attempting to numerically measure someone’s looks. Do I like her? if yes, then she gets a 4. If not then a 1. Obviously 1 2 and 3 are the same as is 4 and 5. I’m sure people have their own systems. But I’m also sure it boils down to yes and no. Which means using that data to then discern attractiveness is unwise. The purpose of the system should explicitly be rating people for the sake of rating if that’s what okcupid researchers want data for.

  34. Ryan says:

    Hahaha. You’re normalizing for overall attractiveness, right? One more straight explanation of the results is that the more 1 votes you got the lower the girl’s score turned out. Imagine a world in which only those men rating the woman 5 would message her. Then the only valuable rating to predict messaging is the % of 5s. However, since you normalize for score, the more 4s you get, the more you compete against girls with % of 5s. This gives the false impression that there is meaning in the results even though it is an artifact of the normalization.

  35. Cat says:

    OMG! That explains everything!! I’m considered cute and often-times guys don’t give me a second glance. Hence why I’m on online dating sites. Ughhh. Thank you for this. I should start flaunting even things I’m insecure about.

  36. Michael says:

    So basically…

    Guys are more likely to message women when they feel they’ve got a shot? AND they tend to have resentment for “genuine hotties” they know would drop them in two seconds or never respond?

    Hmm… Duh!

    How about you figure out a way to stop referring to me: single moms, hipster losers, and women that live clear across the country?! I need the perfect woman in my area, not across state lines!

  37. redroom26 says:

    Have the contents of the msgs been analyzed? This study doesn’t make any sense if you don’t know what guys are writing to girls. More msgs is not necessarily the proper metric to be tracking. For instance, shouldn’t more msgs from interested men that are better matches for you be the metric for success here?

  38. ravenfairydust says:

    All this does is explain why this many people are single and on a dating site. Clearly VALUES are what separate those inclined to have secure, successful relationships, from those who can’t figure out why you can’t seem to find “the right” person.

    Thanks OKCupid, for so vividly demonstrating this.

  39. OrionEridanus says:

    Looking at your “master plot” of Number of Messages in the last month vs. Attractiveness, this is not a strong trend. your simple regression fit best fit, does have a pretty low confidence value (Chi-squared or whatever you choose to use.). You could try something fancier with polynomials, but more likely what this means is there is some other relationship governing the number of messages besides the standard deviation of “hotness” that happens to weakly correlate with your linear regression derived metric but is not it exactly.

    Meanwhile, I am guessing the male users who read this blog will start sending more messages to the more standard rated “4”s and fewer to the HOT or NOT or women, which perhaps was your intent.

  40. Hildegardo says:

    No way this can represent a mens true reactions because of this being online. There picture don’t tell the full story. All these men thoghts and actions wil change when face to face with a woman. Those results would totally be different if it was conducted not in the virtual manner.

  41. Joolz says:

    This was fun to read! Great concept! I love me some snaggle tooths (snaggle teeth just doesn’t have the same appeal) and big noses!

  42. Matt says:

    I think OKtrends did the wrong kind of statistical analysis to find their formula. When doing multi-model inference you want to guard against lots of parameters. Of course a complicated formula fits the data better than a simple one, but this does not make the complicated formula more correct. For example a quadratic model will fit better than a linear model every single time. That does not mean the complicated quadratic formula is better than the linear one.

    Now lets look at Okcupid’s formula, it is of the form y = a*m1 + b*m2 + c*m4 + d*m5+k. This is a very complicated linear model with 5 factors. How do we know this model performs better than the model y = d*m5 + k, well the only way to test this would be to look at how well this model fit the data and then compare that to OKtrends model. Now remember OKtrends model is complicated so you must punish it for having so many parameters. Therefore, we use an AIC calculation to determine which is the best model (http://en.wikipedia.org/wiki/Akaike_information_criterion). It would be interesting to see which model wins. My guess is that the model of the form y = d*m5 + k would win, but thats just a guess, we would have to do real statistics to find out.

  43. Fred says:

    “So this is our paradox: when some men think you’re ugly, other men are more likely to message you. And when some men think you’re cute, other men become less interested. Why would this happen? Perhaps a little game theory can explain:”

    What if the answer were more about people being genuine individuals, and understanding that not everyone like everything, maybe being okay with less people liking them.

    And who would date an A-lister? PROFESSIONALLY SINGLE…

  44. Lord Walrus says:

    The natural thing to model here would be probability that man M will message woman W as a function of M’s rating of W. Instead, you have modeled the number of messages received by W as a function of her rating distribution. This seems to assume that the generators of responses–individual guys–are sensitive to these ratings distributions (that is, that each man knows what all the others are thinking). This is strange and, I think, unnecessary.

    Suppose, by analogy, we have two basketball teams, both with players of the same average height. Team A has has five guys who are all between 6’2″ and 6’4″. Team B has 4 guys who are under 5’8″ and one giant. Team B, it turns out, gets many more dunks. Is this because having highly variable height increases dunking success? No, it’s just that only very tall guys can do it at all. Does this explain the patterns of your aggregate data? It certainly gets the first part–the variance effect: the probability of messaging gets a big boost from “5” ratings (analogous to extreme tallness). What about the reverse “4” effect, in the counts regression? Well, it’s possible that the conservation of probability mass acts like a fixed average team height, and so makes 4 and 5 ratings anti-correlated. More 4 probability predicts less 5, and therefore, fewer messages.

  45. Dude says:

    Ugly chicks do way more stuff.

  46. afroblanco says:

    What you’re missing here is her profile. Consistently the “women men disagree about” are all women who look more interesting — maybe a little gothy, more of an individual style — and that’s probably reflected in her profile. The ones who get more messages are probably more specific in their appeal, and more likely to put interesting things in their profiles. Whereas the pretty nondescript ones are probably more likely to say shit like “likes to go out dancing, likes to go out to eat, likes to have fun” that you can’t really base an email on. If you really wanted to do something interesting, you’d cross-correlate this with some sort of profile word frequency analysis.

  47. Chris says:

    I think it’s about vulnerability. Men like a woman who has SOME chink in her armor somewhere or they risk wasting their time on an impossible diva. When a picture has something divergent like that, it says a woman isn’t perfect, even if she’s almost there. Men like vulnerable women. It’s in the genes.

  48. Jimmy Soul says:

    if you wanna be happy for the rest of your life never make a pretty woman your wife
    go for my personal point of view, get an ugly girl to marry you

  49. mgv99 says:

    I’m not going to read 400 comments to see if this has been brought up, but I see two flaws with the author’s conclusion:

    The biggest is people are not told to rate JUST on appearance. But even if we rewrite the conclusion to say “profiles that generate more diversity in rating are more desirable than those that are more middle of the road”, there is another very likely possible inference from these data (I’m a researcher too can you tell?): it all comes down to the # of 5s.

    There’s only 3 (plus the fictional celeb rating) examples to go on here, but it’s entirely plausible that the presence or absence of 1s and 2s are irrelevant, and that what matters is comparing quantity of 5s. One set of examples were tied but I think in the other pairs the one with more 5s got more messages. This fits the standard “the more desirable you are (pic only or whole profile), the more messages you will get” interpretation.

    At first I thought this alternate explanation wasn’t valid because adding in the 4s complicates things. I thought more messaged users would have more total 4s and 5s, but in the example pairs, this wasn’t true. However, I would venture the vast majority of profiles receive ratings of 3 and 4, so it may be that having a rating of 4 doesn’t really mean you stand out to someone and thus receive more messages, because the site is full of 4s. I would like to see the breakdown of % of people on OKC that have received ratings of 1, % who have been rated a 2, etc. If my assumption is correct about 4s, this furthers the notion that what matters in terms of getting more messages is getting more 5s. Diversity of ratings is irrelevant.

    Mary

  50. Robert says:

    Putting up a scattergram without an R2 value for correlation is misleading to all of your readers. Visually, you think you see a pattern, but it must have a statistical relevance. I think ethnic background is a large variable myself. Italian find other Italians the hottest and Scots find Scots the hottest.