> Except for the twin study, all of these focus on initial attraction, not on which relationships work and survive. They don’t rule out a situation where the “initial spark” of romantic attraction is random, but people with similar interests and personalities are more likely to stay together. Maybe everyone in these studies is very stupid (cf. they’re mostly undergrads), and they all feel attraction to random unsuitable people at speed dating events, but in real life, selecting people who are long-term compatible with you is the way to go.
Something along these lines, minus the 'very stupid' and most of the 'random', seems really obvious to me. *Of course* short-term attraction is based more on hotness + charisma + other 'shallow' instinctive preferences than on a checklist of deeper qualities or life-plan compatibility! And of course that doesn't imply those deeper qualities and preferences are unimportant for long-term relationship success!
I think the main issue there is that all of those women sound like insufferable cliches. I mean, I know you exaggerated them into insufferable cliches to make a point, but 1) I don't think you're far off, and 2) if I saw any of those profiles I wouldn't read it as "I am this kind of person." I would read it as "I am choosing to represent myself as this kind of person, because I've found that this kind of person gets dates with people who represent themselves as the kind of person I want to date." Which is different. And a dystopian hellscape you can't opt out of if you want to participate meaningfully in the dating market. I'm mostly just glad I'm married.
Lots of obvious counterarguments ("What? People in bars don't lie?") but I think this is a different case. The issue is the same as all social media - the incentives are to go broad and shallow. My friends who are in the dating scene feel big pressure to present themselves as a type. There's no room to be a driven lawyer who is also a Taylor Swift fan.
The thing that rubs people wrong about these dating docs isn’t so much that people express preferences or even that people try to make themselves sound good (this is bog standard for a dating profile) but that there is an implied “look at how high status I am” embedded in the assumption that lots of people are going to open your particular document and read through your personal essay to determine whether you would think they are a good match for you. On a normal dating site (or in real life) there is a certain symmetry: you see their (short) profile and say “yes”, they look at your (similarly short) profile and say “yes” back, then there’s a conversation that eats up time equally for both of you. It’s arguably true that requiring potential suitors to sign off on your 26 points of agreement before even starting to talk is more efficient since there are fewer false positives, but putting the burden of figuring that out on the other person just feels wrong to fairness-obsessed humans.
I think the docs are a great idea. Swiping apps are horrible at connecting people. I am sure some company will streamline it soon, too. As your samples show, people are trying very hard to sound exactly the same as one another. The docs enable people to highlight their differences.
Shouldn’t dating “success” be based on marriage length, lower divorce rates, happiness level after long periods of time, etc. and not just whether your individual preferences are being temporarily met? This seems like evaluating a diet based on whether you enjoy the immediate taste of the food and not on whether you are actually losing weight.
Am I the only person who would go on a first date with all of the fake dating profiles mentioned, aside from the first? I’ve been partnered for a while but #2-#5 remind me of people I actually dated at some point, and enjoyed the time with. Although with #5 it could be really hard to keep a conversation going unless you happened to both love the same obscure anime.
I am very likely an anomaly, but my dating life has been WILDLY variable. My dates/partners have had about a 40 year age span, different classes, education levels, drastically different Big 5 spreads, different races, nationalities. I don't feel very surprised when I hear about these things not mattering in practice.
I obviously do have preferences, but they are subtle and flexible enough that they can be satisfied in ways that look superficially dissimilar.
Maybe I'm just a mutant, but I'm a mutant that aligns with the science here, so?
Also, I think you may be missing that people sort of merge when they are together. Like I would love god to tell me how many of those 96% matching political people shifted their politics specifically in contact with their partner. It doesn't surprise me that 96% end up converging, but that doesn't tell me how they started, and I bet that matters for politics and a lot of other dimensions that can change over time. A different example is that my first wife was a working class high school dropout, but also blazingly smart and hard working, and in contact with my upper middle class intelligencia sensibilities (and a fair bit of cultural metis), she got an advanced degree in math and now is a math professor at a local university. That sort of thing is not captured in the data.
I would not have preferred to know my wife’s preferences before our first date, and I’m pretty confident she feels likewise. That was the late 90s, but I don’t think that dynamic would be different for us now.
Moderately liberal democrat and irreligious NY+Miami, conservative evangelical old Yankee family. The only real thing we both had in common when we met was science as an avocation. We grew as adults, and became way more compatible, then got married. The change from who we each were ~25 years ago to now is stark. The same dynamic was even more true of my parents, this month is their 50th anniversary.
The only information that would have been a dealbreaker before or during a first date? Smoking. Fortunately for both of us, she quit cold turkey about a month before I asked her out. Re: a describable dating preference, for me it would be fine to learn this during a first or second date, rather than published in writing.
Perhaps these are too anecdotal to have any value for reasoned debate on the principle of dating preferences. But within our spheres of friends, most of whom have been married once and still are, the degree of political, religious, and cultural change each person has traversed between 20-40 years old is between “a large degree” and “a difference in type.” I hold to my mom’s advice to me, with my daughter: date different types of people, to explore and learn what you are looking for and esp what you can’t abide.
I suspect a lot of the objection is just coming from the tell-don't-show style that many of these are written in, and foregrounding the boring-but-important stuff instead of putting it off in a corner somewhere. I recall Aella complaining about how may of these are tell-don't-show and should be more show-don't-tell, and claimed hers was, but on reading hers, I still found it to be tell-don't-show! So, yeah.
I am reminded of a novel, The Rosie Project, by Graeme Simsion, about a man who's tired of deal-breakers turning up among his dates (e.g. his dates habitually don't tell him that they're vegetarian until after arriving at his house for dinner), so he compiles a questionnaire that all prospects must answer. His friends futilely try to explain that this is not how you go about it.
I guess sometimes you might not know that you would like someone with a given trait until you meet a person with that trait. Also, people change their preferences, sometimes influenced by people they love. The problem with prefiltering on a set of criteria is that you might miss an opportunity with a person which fails on some of your preferences, but scores very high on preferences you did not know are also important to you.
The problem here isn't that profile description is a bad filter - it's just that mutual physical attraction is a much more reliable one.
Text profiles are hard to verify, reflect self-image more than personality, and it's still a cointoss whether you two actually vibe IRL. Meanwhile the photos are pretty foolproof - if someone shows up on a date not looking like the photos, the jig is up straight away. And it takes a split-second to judge a photo, when reading a dating doc takes a minute or more - making it faster to find your 1/500 person.
With all the dating apps switching to a swiping-based system, it seems like using apps as a first-pass attractiveness filter is a much more workable model than expecting them to sort people by compatibility - which was way easier 20 years ago where online dating was a niche for a specific kind of person.
In fact, from what I see, the vast majority of dating profiles you see on apps like Tinder don't resemble any of five examples in this post. They're simply empty or have a pithy one-liner. It's very easy to discard these profiles as "not taking dating seriously enough" or "what do I even talk with her about".
I strongly recommend in favour of attempting to match with these profiles anyway! Empty bio does not equal empty brain. Some of my best dates came out of matching with these wildcard profiles - purely on physical attraction alone. If there's no long-term compatibility, that's okay, spending an evening in the company of a stranger who finds you attractive is still better than staying at home. And anything more than that comes as an unexpected bonus.
The long-form dating profiles found on sites like OKCupid and PlentyOfFish were great back in the 00s, but they lost out to swiping apps and their ease-of-use because at the end of the day the only truly key issues are looks, money, and status (and also partly industry consolidation). Unless you are looking for something really specific, like a religion or someone into polyamory, nobody much cares about reading your carefully written profile--especially all the cliched ones.
1. Should you have expressible dating preferences, and
2. Should you actually express them?
I would say that of course yes, it's reasonable to have dating preferences, but you should be very careful about actually expressing them lest you alienate or repel even the people who satisfy them.
I think I'm in love with Hana. Or possibly the internal simulations of Aaron Smith Teller and the Comet King have hijacked by brain's attraction centers to make me think I'm in love with Hana. God dammit.
OK Cupid used to have a blog called OKTrends where they did statistics with data from their userbase. One post I remember in particular looked at the match questions that predicted resulting relationships (which they knew from a survey when people disabled their profiles and answered, I met someone via OKCupid). They suggested that the questions people thought were very important (religion, smoking, etc) were less predictive of relationship-forming that some others ("Do you like scary movies?"). Here's the link: http://web.archive.org/web/20110315063632/blog.okcupid.com/index.php/the-best-questions-for-first-dates/
Jane has remarkably narrow tastes in anime for someone who likes anime. Someone should introduce her to Monster and Nodame Cantabile. But she sounds like a nice person.
You should see the matrimonial section of Indian newspapers. However, I hope you won't mock them.
They're very earnest and absolutely open about things Americans would find non PC.
Often, the parents are placing ads. I only saw about 10 of these and one was an ad for a gay son - the mom was looking for a man for him.
Westernized folks there wouldn't deign to use ads to find spouses for their offspring and while I think they'd have very similar preferences, they'd pretend to be very open minded. Therefore I find these ads charming.
I live in America and every time I visit India I browse a few ads in the papers because they tell me so much about things that have changed (or not) - since I grew up there.
Hi, I'm an ultra-orthodox Jew, and I would like to point out a few things about "shidduch resumes."
1) The single biggest thing that's looked at are the schools the person attended. Ultra-orthodox high schools are generally selective, and most schools have a defined stereotype and perceived rank. This becomes even clearer after high school, when the boys go to yeshivas that are very often smaller than most high schools (my options for yeshiva were ~25 guys, ~30 guys, & ~160 guys), leading to even more sorting. Girls go to 'seminary' after high school and similar dynamics are at play for them. In addition, prospective in-laws are now able to ask the staff of the school about the person. This works pretty well only because ultra-orthodox jews generally get married less than one year out of the religious school system.
2) Another useful feature of these resumes is to provide references who are willing to talk about the person, which allows prospective in-laws the chance to see who their friends are, and how their friends describe them without meeting the prospective partner.
3) The basic outline of the reference will also tell you if the person has deviated in any significant way from the standard ultra-orthodox educational path, which is then interpreted in different ways according to personal preference. (If someone is currently in Yeshivas Brisk in Jerusalem, the path they took to get there may show that they are not a typical 'Brisk Bochur,' for better or for worse.)
4) Ultra-orthodox parents often support their sons-in-law in Torah study for several years after marriage, the resume may indicate how long such support is being sought or offered.
5) It also bears mentioning that in the orthodox jewish world, parents are heavily involved in their childrens' dating lives, and the possibility of comparing potential partners is antithetical to all of the social & religious norms surrounding the dating process. People in the shidduch process are not supposed to see one profile that says "I'm in pre-med" alongside one that says "I'm a secretary in an elementary school," instead, resumes are given to parents. Parents reject any that are extremely obviously incompatible, kind of as in Scott's example, and then if they receive one that seems to make sense, they investigate. This includes general fact-finding (personal history, looks, family, sub-subsectarian religious affiliation, etc.), reputation checking (are they a 'catch'?), and trying to get a sense of whether they would be likely to hit it off. The potential partners are then supposed to consider the combination of resumes, 'backround check', and any other information and decide if they are interested in dating each other. If they agree to meet, the partners are understood to be interested in the possibility of marriage; at this point the potential partners are now running things, mostly. This extremely high degree of parental involvement may make the high level of filtering easier than it would be for the person themselves, who may have a bias toward saying yes.
6) It is also worth noting that the general sentiment of the community is that resumes are a necessary evil, and we commit a gross injustice when we attempt to pin someone down to what can be stated on a resume.
7) Finally, it is worth noting that resumes are a relatively recent innovation, and while I am not certain when they became common, they definitely were not a feature of jewish life in prewar European communities.
TL, DR:
The orthodox Jewish process is a lot more complicated than resumes, and resumes are not a longstanding tradition.
This is absolute insanity. I have never found a girlfriend on a dating app who was in the same country as me, let alone one who was in a similar age range or had remotely similar views as me. Or maybe it's just that guys with Muslim names either have to be super-successful, or date Muslim girls by pretending to be super-religious.
My experience having a dating doc: with one or two exceptions, everyone whose emailed me about it has been unusually interesting and compatible with me (I think less based on specific listed preferences than on being a good match for my writing tone). The exceptions were also alright in that at least we had a more interesting initial conversation than I usually have with a random app match, because they were working off knowing my interests.
I've also had one or two people I matched with on a dating app that clicked through to it, saw I want kids, and cancelled the date on me (which wasn't fun, but probably a net positive).
For what it's worth, my marriage is a statistical anomaly. I'm an atheist liberal Democrat with an undergraduate engineering degree, and my wife talks to ghosts, voted for Trump, hasn't finished community college, and, most astoundingly of all, does read books but does not like Terry Pratchett novels.
I started dating her around nine years ago. If you had asked me back then what I wanted in a girlfriend, she would have checked almost none of the boxes and probably should have set off a lot of alarm bells, but when you're 32, not employed, living with your parents, and have never had a girlfriend before, when someone shows interest in you you're inclined to give them a chance; it only took one date to decide she was a keeper, and ever since, in spite of problems that would probably make a sane man or woman run for the hills, I've never been happier or felt like my life has been more meaningful.
Different take on why people react negatively to dating docs:
For some people, writing a dating doc is socially similar to saying "I'm smart".
People don't think you're smart because you say you're smart, they think you're smart because either
1. You convincingly "perform smartness"
2. You demonstrate social proof of smartness (e.g. you have a PhD in Mathematics from
Failure to do either of the above make people think you're either too stupid to understand the rules, or trying to deceive people too stupid to understand the rules.
I see two main (overlapping) groups of people for whom a dating doc will trigger a similar reaction:
1. If the reader lives in a low trust environment (maybe due to their filter bubble), the idea of telling people who you are will trigger the trust issue. They believe you know that only a gullible person would believe this doc, and thus you are manipulative.
2. Many people are not comfortable writing structured documents - to them this sort of guide is likely to come off as trying *way* too hard, which signals either desperation or deceptiveness. The idea that this is something someone could casually type out never cross their mind. (perhaps adjacently ,the idea of solving problems through structured documents may be alien to them)
Now here's the twist I find interesting - even if you're able to understand why these documents may be a good idea in principle, if you have the emotional reaction that sharing this doc was a rather large social faux pas that is generally going to short-circuit any "pro dating doc" argument because people filter *hard* against people who do things they consider large social faux pas.
I get all those arguments, but my personal experience with any kind of online dating was not so good, mostly because the subconscious aspects of attraction were missing. Most of the time, I knew after 5 minutes into a date that this is not it.
After my divorce, going to events where like minded women could be expected worked much better. For me it was tantra, and the criteria were interest in self development, a certain openness in relationship, willingness to grow together and an interest in new experiences. I found my second wife and later most of my lovers there, people that I probably would have neglected in online dating
Since we're doing this about the dating profile stereotypes (I'm married in real life, but whatever):
1) I'd totally give Cindy a chance. I don't go to bars, or at least I don't drink at them, but I did enjoy dancing back in high school and she sounds like a lot of fun. It could go wrong, and I imagine that the most likely outcome is that she rejects me, but I'd still like to see what happens. And I do happen to like Taylor Swift's music.
2) This girl is out of my league. I don't have much in the way of career ambitions and tend to be lazy and procrastinate a lot. I could offer to be a househusband or assistant-type the way a wife might have been in The Bad Old Days to a high-achieving husband, but I couldn't be her equal. I'd probably leave her to other people who are Going Places and just not respond to the profile.
3) Astrology? Eww, red flag. She sounds like a totally sweet and caring person, but I'd ruin it with my intolerance toward woo. Pass. (I don't like my actual wife's belief in the supernatural either, but that's another story.)
4) Yeah, I'm totally interested in this kind of person; I took one economics class in college and the textbook totally changed my way of thinking. Definitely contact.
5) On paper, this is my dream girl. I would totally be down to hang out and let her show me her favorite animes and play video games together. I'd be afraid of coming on too strong, though, or that I'd end up with yet another one-sided attraction to someone who only wants to be friends. Once upon a time in college, I ended up in a conversation with a girl I while we were waiting for a campus bus; when I asked her what her favorite Final Fantasy was and she said Six, my response was a only half-joking "Will you marry me?"
"98% of whites marry other whites; 90-95% of blacks marry other blacks."
Those figures are too high, at least for American newlyweds. According to the Wikipedia article you link to, in 2010:
"Among all newlyweds, 9.4% of whites, 17.1% of blacks, 25.7% of Hispanics and 27.7% of Asians married someone whose race or ethnicity was different from their own."
> [after speed dating] realistically probably people just chose whoever was hottest and maybe most personable
If true, this would be a pretty damning indictment of the usefulness of speed-dating. Instead of going to a speed dating event to talk to people for 3-4 minutes each and then asking to be put in touch with the hottest of them, you can open tinder and swipe right on the hottest profile pictures in a smaller number of seconds each without leaving your couch.
Knowing 'Hana' is just a Scooby-Doo-style mask and if you ripped it off you'd find either Scott Alexander or GPT-4.. doesn't exactly seem to diminish this longing.
Suspect this sensation is not unique amongst ACX readers.
I am currently building a dating app that is similar to Tinder but WITHOUT any chatting. It will launch in about three months.
I agree that certain "basic" info like what the person is looking for (gender, long or short term relationship) is obviously significant. But the rest is largely bullshit. You only really get to know a person in real life.
I'm still reading, but wanted to add to first part, where Scott describes critical incompatibilities. In my reality (country close to Russia), there is most important point, probably. I totally won't be able to date someone who supports Russia in the war with Ukraine. It is just not imaginable.
"For a while, the person in the entire US with the highest match percent with me was my actual girlfriend (who I had met separately, not using the site)."
I'm a bit confused about how this could happen, isn't it extremely improbable that the person you're most compatible with in the entire country just happens to be someone you've also met irl? Seems a bit like when people claim to have met their one soul mate out of the 8 billion people in the world and it just happens to be someone that lives in the same building or works at the same place, rather than say... a village in Mozambique.
On this topic, maybe a bit of an odd question, I've been wondering for a while reading the blog, how Scott manages to have so much success dating people selected on extremely niche psychological traits. The readership for ACX skews about 90% male, right? You'd think there'd be 10 guys similar to Scott for every one woman similar to his partner. I'd guess the competition for guys following this strategy is extremely fierce. Is this something most guys could feasibly pull off, or would you need to be some kind of rationalist chad?
>For a while, the person in the entire US with the highest match percent with me was my actual girlfriend (who I had met separately, not using the site). She told me I was her second-highest match percent; her ex was #1.
Surely this demonstrates that these questions only tease out short-term surface-level compatibility rather than long-term compatibility? After all, your then-girlfriend and her ex broke up, as did you and your then-girlfriend.
IRL you have three seconds to get someone's attention and it takes ten seconds max for them to decide if they even want to talk to you, which is why your first words to them must always be a question. The swipe-right apps are little different. You have only seconds to make an impression. On first contact a salesperson doesn't try to sell the sale. The salesperson tries to sell the appointment. Perhaps there are intellectuals who set aside time and sit down purposely to read through dating resumés, dull though they may be with little attempt to capture the reader from the first sentence. I think that a dating doc should only be communicated after initial contact and a primary expression of interest. At the very least it's more graceful than going straight to intimate pix.
Whatever happened to frustration, rejection, not lying about yourself and dignity? All character building experiences. I guess I'm showing my age, but I'm so glad I don't have to try to date anyone (married for 47 years). I rather think online dating is like social media: you lie about yourself and your life for 'likes' of a more direct kind.
My wife and I were married in 1985. We were 23 and 22 when we got married. My parents were Republicans, hers were Democrats. It wasn't something we ever discussed or felt worthy of discussion.
So the 4% number of cross-party marriages really stood out to me. That's a huge decline in the available pool of potential spouses. Which is a real shame.
Who wants to marry someone like themselves? Many great relationships have a yin and yang that a dating doc would actually filter against. Larisa and her ultra ambitious husband would end up at each others' throats. She'd probably be better with a laid back, good looking, funny guy, who could also look after the kids' emotions while Larisa grew her multinational shoe empire. Maybe I'm old fashioned, but I never felt like I needed a partner to share too many of my interests. She does have to be smart (good breeding genetics), good looking, and from the same social class or higher. But if I am into anime and reptiles then i actively don't want my partner to be into that stuff. It is just too much.
The 5 women described here are hobbies, not people. I can't get any judgement out of them other than "their hobby seems fun/boring". #1 is the only one I strongly dislike, I don't have much preference between 2-5.
I myself find dating docs useful. Anybody who writes one is a hostile tentacular alien to which I am anathema to, as is anybody who thinks they are a good idea. This is because they signal wildly divergent cultural norms to which I view as monstrous, being that they are almost entirely represented in the Rationalist/EA sphere. One could say they identify low match percentage. Online dating profiles more broadly signal only mildly divergent cultural norms from my own, to the extent that it's theoretically possible for somebody I find normal to be found by sheer accident.
I could then assume that the monstrous aliens would love them as a tool for finding each other, if only the population wasn't skewed heavily towards male.
I recently did a quick study where I asked people for their personality-adjective partner preferences, and then asked them what they meant by those. I think the answers were pretty revealing: https://twitter.com/tailcalled/status/1680269225418522628
For instance, one guy who preferred an honest partner brought up the importnace of her not cheating on him. And one guy who preferred a nice partner brought up that he wanted her to take on a housewife-like role.
I think part of the issue is in using abstract terms like "nice", when really everyone wants something more specific. Like the conservative guy wants a nice housewife, you probably want a nice girlfriend who doesn't mock nerds and doesn't aggressively push woke stuff, and a woke guy probably wants a nice woman who cares about the oppressed. You all agree that you want someone nice, but you disagree about what "nice" means.
”How do you translate these to single-dimensional scores on a psychological exam?"
A lot of the things you bring up in your profiles could totally be measured psychometrically by the way, they'd just be in the interests domain rather than in the temperament domain. There are correlations between interests and personality, but they are relatively modest, so practically they are pretty different.
Usually the way you measure interests is by asking people whether they are interested in the relevant things. So for instance, "Are you interested in business?", "Would you be fine with combining vacations and work?", "Would you be interested in going skydiving?", "Would you be interested in hiking in the Alps?", etc..
Probably this isn't gonna be entirely unidimensional (I imagine this profile is combining nearly-orthogonal factors of thrill-seeking/outdoorsy interests and business interests), but that's not really a problem psychometrically (you're probably familiar with the Big five, which is a multidimensional measure of personality). As long as it's relatively low-rank, it's still gonna work fine, and I think you'd be surprised about how much variation you can stuff into surprisingly low-rank models.
If pairings ultimately end up being sorted without the docs then who needs them.? Without them you'll date people from the broader spectrum of humanity and maybe learn. These docs strike me as an effort to sanitize human contact. A contraction of human possibility.
How much of the problem is the fact that the set of single men and the set of single women are incompatible with each other? Single men skew conservative, and single women skew liberal. Small group differences overall compound into large differences at the margin. I think most counterintuitive dating market conclusions can be explained by adverse selection. The most compatible people are already matched up.
I'm surprised at the negativity. For me, someone having a dating doc signals some degree of confidence, intelligence, good communication, awareness of their wants, commitment to taking a potential relationship seriously, and an interest in organising and optimising their life. That's before I've read the contents.
Those are all attractive qualities to me and I would happily filter based on them were I still in the dating pool.
I’ve been married for 38 years (married at 19). This is all so baffling to me - life and love can just happen if you let it - if you enmesh yourself in other humans. The heart wants what the heart wants...
Serious question for all the lonely nerd guys here: which personality type did work for you? (One of Scott's or another.) We all know there aren't enough Hanas to go around, so who did you find happiness with?
I have a close straight male friend where part of our friendship is acting as one another's dating coach and commiseration buddy. We've both noticed how OKCupid and all its sibling websites and apps across Match have been degraded to the point of uselessness. Someone whose profile I haven't seen yet sends me a message? Maybe I receive it, maybe I don't, maybe it hits my inbox THREE YEARS after they sent it (true story). Someone who I've matched with sends a message? Again, maybe it shows in the inbox right away, and maybe it shows up a month after it was sent, and probably it shows up never.
Dealbreaker preferences, like smoking or parenting status? Frequently ignored.
Premium membership does not fix these issues, either. While I've never paid for membership, my friend has, and he likewise often receives messages weeks/months/years after they were originally sent and also has his dealbreakers ignored. Plus, as a straight man he is inundated by messages from laughably transparent scammer profiles that OKC is absolutely capable of shutting down but does not.
It's so clear that OKC is constantly being tweaked to thwart most daters' ultimate goal: To find a partner and no longer need the dating platform. I don't necessarily mind a business giving users the choice between a shitty free product and a terrific premium product, but OKC can't even do *that,* likely because (duh), once they've identified that someone is willing to pay $40 a month for their profile, then they REALLY don't want that person to be successful enough in dating to leave.
It just sucks. How do you solve the fundamental problem of dating companies and their customers having perfectly opposed goals?
(Use Fetlife, I suppose, but that's a very tough putt for those with weak stomachs.)
Okay, I was dubious about "dating docs" but the comments section has converted me on them. The amount of people going "Oooh Hana my dream girl" while I'm going "she has to make up some silly jokey name for the rooms in her house in order to remind her to buy groceries???? how does she cross the street unaccompanied?" 😦 Being interested in your pet subject or hobby is one thing, being economics-obsessed to that degree is.... worrying. I think you all know I like the works of Tolkien but dang it, I'm not naming the rooms in my house out of his writings in order to remember "I need to change the bed sheets and put the washing on".
Definitely need dating docs to keep the likes of that away from me! (And me away from them, because I'm sure I'm every bit as much not their cup of tea).
(Also I am so glad I have never had the faintest interest in dating. So, so glad. I now see why the sexual instinct has to be so strong, otherwise the rest of you would never manage to reproduce the species if left to your own devices).
On the other hand people are quite capable of rapidly filtering through basic profile information themselves. The big problem with matchmaking isn't the matchmaking part, it's getting enough people in and helping everyone filter through everyone else as efficiently as possible. So helping people put together profiles containing relevant information including, most importantly, their picture, is a good thing, but telling them who they should pair with is not.
This is all making me wonder - what's the male version of Hana? Stereotypes suggest it's "I like Taylor Swift concerts, long chats about feelings, and bringing you flowers", but does anyone know a real organic example that looks like it?
Partner trait preference is all well and good, but the instant you turn dating into a game of app algorithm minmaxing or find yourself writing a goddamned resume you've type-A'd yourself and the whole scene you're in into a region where I consider you more an animate object than a human being, for better or worse.
Apparently some subconscious attraction between individuals in physical proximity is determined by distinct smells caused by variations in their combinations of MHC (Major Histocompatibility Complex) genes:
So if these could ever be classified and expressed in a fairly compact set of letters, analogous to blood groups, then perhaps there is a marketing opportunity for a new dating app, or a supplement to existing apps:
Clients registering would mail a cheek swab, from which their relevant DNA could be analyzed. Then the app would pop up possible matches whose MHC differs widely (which it seems is the positive factor, as opposed to similarities). A bonus is that presumably the offspring of a successful match on that basis, in combining differing MHC from both parents, would have an unusually healthy immune system!
But it is questionable whether these instinctive chemical indicators could trump incompatibilities of the kind being discussed here, such as different interests and goals or conflicting opinions on politics for example.
I had no idea how statistically rare my relationship with my spouse is. I'm a Christian Republican with a JD and they are a Agnostic Democrat that dropped out of college. I never would have guessed those were strong sorters. However, both of us have the same energy as Jane in that last profile. I think it might be really hard to use survey's to best pin down what makes people have a good marriage, but it would be great to have more research.
An unexplored angle: That’s also some “is/ought” confusion (and a bit of naturalistic fallacy) in the objection that Scott dive into.
Things like dating docs are an (attempt at) innovation. So, by definition, objecting with “but that’s not how people do it” is more of a non sequitur than it first appears. If Henry Ford showed someone his Model T plans and someone said “but, statistically, most people travel by horse or foot, not by machine,” I wouldn’t begrudge Henry saying “no sh*t, what’s your point?”
Along a similar vein, when divorce rates are 45% in the US, an objection that’s basically “we can’t innovate long-term partner selection in X way because people tend to do Y instead” is pretty weak. Clearly Y has plenty of room to improve, and innovation necessarily assumes a future that looks different than the past.
Here is my take on this: you can infer a ranking of people from their decisions and preferences (People’s choices determine a partial ordering over people’s desirability: https://nunosempere.com/blog/2023/06/17/ordering-romance/). Then, it's not implausible to think that someone using a dating doc reveals that they are lower down that ranking--because they couldn't find romantic success without the dating doc. Therefore, from an evidential decision theory perspective, it's a bad move to have a dating doc**.
** caveats may apply. other decision theories may provide other recommendations.
As an aside: I’d never heard of the ‘dating doc’ before this article, but they remind me of the Personal User Manual idea that I learned about ~5 years ago, which I love and always try to proselytize.
As someone currently in the muck of dating apps after the end of a long relationship, I haven't encountered the 'dating doc'. But I think the issue with listing preferences generally is that lists of traits and profiles are bad a capturing personality - and personality determines a lot of the 'spark' of compatibility.
That, and there's a tendency to treat the date like a dry job interview, and it can be tough to break out of that format. People ask formulaic questions, and are guarded about potential flaws and their pasts in ways that don't allow them to act like the interesting and idiosyncratic people they are.
Anyway, I've met a lot of people in my preferred demographics who I don't otherwise seem to have much to converse about, whereas most of those demographics besides gender (age, cultural background, education, politics interests...) aren't rigidly defined and can be flexed.
Done well, a dating doc is an asynchronous speed dating event. Hundreds or thousands of potential matches can spend 4 minutes with "you," and you do the same, without having to coordinate a specific time.
Looked at that way, a video where you say all the stuff you would put in a doc should also be an option. Less crunchy data transmitted per minute, but 1000x more "vibes"
Hmm. I feel like the negative views of date-me docs don't really have much to do with doubts about the efficacy of describable dating preferences, but rather a concern that writing a date-me doc is a yellow flag for potential narcissism and entitlement. That doesn't mean that this is necessarily a reasonable view, but it at least explains some of the revulsion.
I wanna write a longer reply but I wanna preliminarily say I have no idea why it’d be possible to pick one significantly more interesting person out of those hypothetical girls based on just that, and I really really don’t understand why I so obviously should be. The idea is really strange to me!
I feel similarly about Myers-Briggs as you do about dating compatability docs. It's so obviously useful once you understand it. Existing studies that don't show value tend to be poorly designed.
This quote from Reddit is a sentiment I’ve seen around a lot:
“Jesus. I'm not trying to be insulting at all but even just the existence of a Date Me Doc, or an online dating profile is just so antithetical to how I grew up. (I'm 38) I see people addicted to these apps and now "burning out" from them, it all seems kinda crazy and gamified. The over analysis and "sciencification" of attraction really take the fun out of life, and it happens in every game where people min max to suck the joy out of it. Just talk to people...it is still easy for people from my generation.”
But what it misses — one thing it misses, at least — is consent. If you join a dating website, you can reasonably assume that the people you find there want to find dates and will not be offended if you ask them. Compare that with asking someone out that you’ve met at a local board game meetup or folk dance meetup or whatever: someone very well could be offended by that, since they’re just there to do $HOBBY and now they feel like they’re being treated like a piece of meat. I often see this expressed through a feminist lens — “women go to $HOBBY meetups to do $HOBBY, not to get hit on!” — but it seems more broadly applicable.
"only 4% of marriages are between Democrats and Republicans". I think this means if you get one partner who's willing to state "Democrat" on a survey and the other willing to state "Republican". As opposed to say all those "independents" who still often vote the straight ticket as being easier. Before you object to that characterization, other surveys have found more than 30% of people answering "independent".
"The most popular dating apps, like Tinder, almost push you into that mode." I think it's more that the companies that can monetize their apps successfully have found that the users who go for attractiveness are the ones who pay (or are susceptible to ads or however else they actually get money).
I marveled at the profiles, but then thought better. It's a signaling device - you care about x, and if your potential partner is repulsed by that it's better not to waste both your time. On the other hand, I've had a couple of relatives who ended up divorcing because (probably) not discussing these signals enough - but not before producing kids. Are we better off for not having these now? Unsure.
Whenever the topic of relationship choice comes up, I’m stuck by the question, ‘how do people KNOW this person will be a good match? Dating docs try to solve the problem, but people (or at least some people) seem surprisingly good at it without help. (Finding the person is a separate problem.)
Example, my husbands best friend in grade school now lives in another country though they’ve remained in touch over the years. Their initial attraction was largely based on a mutual love of fart jokes and their accomplishments were of the video game variety. They’ve grown into very different people, both from who they were then and from each other. So it was surprising to eventually meet him and discover how strikingly similar he is to me. We took different paths, but ended up with the same career. We have read many of the same books, blogs, etc. - notably ones of no interest to my husband. I think it is true that this friend is the sort of person my husband is very compatible with, but how on earth did he know that based on information available when they were in kindergarten?
Having thought more about this I have a more coherent objection than my "I don't like Scott's made-up profiles and want to make that a complaint about actual profiles" comment earlier.
Determining dating preference is a multi-armed bandit problem.
For those unfamiliar, this is a problem in computer science/machine learning. Imagine you have two slot machines - one of which has a 40% chance of giving you a dollar and one of which has a 60% chance of giving you a dollar. You want to make as much money as possible but you don't know which is which. Assuming you have limited resources to play these machines you want to spend as much time as possible at the 60% machine, yes. But you also want to spend some amount of time confirming that you have guessed the right machine. Since you can never be completely sure which machine is the 60% machine, you employ a strategy that balances exploration to increase your confidence that you have the right machine with exploitation of the one you suspect is the 60% machine.
In dating you should also optimize for both exploration of your preferences and optimization of those preferences. Some folks like Scott may have a strong understanding of their preferences, plus be extremely high status and able to whittle down a large dating pool by several orders of magnitude and still find a match. Such a person should spend a lot of time exploiting those preferences because they're optimized and possible to exploit. But for most folks, preferences are a guess at what would make them happy in a partner, not a 100% certain formula for partner perfection. You may have dimensions along which you're very certain ("I want to date a woman"), dimensions along which you're not so certain ("I think sharing my political affiliation is important in a partner but I'd be open to meeting someone who thinks differently") and worst, dimensions along which you think you're certain but you're not ("Oh turns out I didn't want to date a woman, something it took me 10 years to learn because I only went on dates with women."). Plus for most folks, some amount of preference compromise is necessary, so knowing what is and is not a dealbreaker is another exploration/exploitation puzzle.
It's not necessarily true that dating docs/expansive profiles/checklists/whatever have to silo you, preventing exploration and putting you all in on your guessed preferences. Someone who thinks critically about their own assumptions about their preferences might use these tools in a different way that actually helps optimize. But in practice, I suspect they do way more often than not.
When I wrote a dating profile, I updated it after every unsuccessful date in such a manner as to be less attractive to (the sort of person) who I went on a date with, targeting in particular the ways in which we were incompatible.
It worked; my spouse messaged me because my profile came across as a likeable asshole. (Fair cop.)
Where dating profiles fail to be useful (for the specific purpose of finding a compatible life partner), I think they tend to fail because the users are trying to attract as many people as possible (complicated social reasons for this, but this isn't what you actually want to do, even if you are struggling to find matches now*).
And dating docs are basically just dating profiles that have been divorced from the specific user interface of a specific dating app. All the same failure modes apply.
Maybe somebody should create an industry standard for dating profile documents, so they can be imported into any dating app that supports the standard, using whatever data the dating app supports. I don't think it's a secret that people using data apps are likely to have accounts on several (although probably their actual usage is mostly limited to one, as regional variations in the popularity of specific applications means that a given dating app may be locally dominant).
*I know some of the responses I'll get to this. "I struggle to find any partners / get any responses now, why would I try to be deliberately unattractive to people?"
Well, first, you aren't trying to be deliberately unattractive, exactly, so much as being honestly unattractive; posting a picture of yourself sneezing ... okay, that might work, as I think about it, but maybe not the way you might hope. Maybe.
Trying to attract as many people as possible is what everybody else is doing, second. Your goal is to stand out, yes? If you can get somebody to go "Wait what?" and keep reading, you've won your first battle already, getting somebody to spend more than a second or two to say "Nope" and move on.
Third, trying to attract everybody is, unless your social skills are pretty on-point, is obvious, and comes across as desperate. Don't come across as desperate!
You can come across as desperate if you go the honest unattractive route as well, note. Don't snivel. Don't apologize. The attitude you should take: These are the things you may not like about me, and if you don't like them about me, move on to somebody else, so we can both save our time, because I'm looking for somebody for whom these things are not a problem / are a positive. The attitude you should not take: These are the things nobody likes about me, and I'm sorry I'm such a terrible person, but maybe you can take pity on me and date me anyways. With the caveat that maybe the "attitude you should not take" is the attitude you should take if that's who you are; there are people out there who will like that. But there aren't a whole lot of them, and you'll have more success if that isn't your entire dating pool, so only go that route if that is exactly who you are, and that is exactly who you need to date.
If you're going into dating for the purposes of finding a life partner - then take that purpose seriously. You are trying to find a life partner. Quit messing around with trying to date people who aren't that, and quit trying to date people who are your ideal life partner, but for whom you are not their ideal life partner; even if that succeeds, do you -want- to spend the rest of your life with somebody who is unhappy in their relationship with you?
Mind, there are other reasons to date which are perfectly valid.
Even assuming that humans actually know what they really want in a partner and can/will formulate their desires into words, I get the idea behind describable dating preferences, but I cannot count the number of times a female seemed to check all the boxes and when I actually followed her scent and met her it was like....meh.
Not done reading but I'd love to see a poll on into/neutral/eww for your fake profiles. Hana sounds obviously awesome and I have a hard time imagining anyone (anyone here?) ranking someone above her, but also would love to be wrong about that.
Is it just me or do Hana and Jane seem the least bad options? Hana sounds a bit mad, but maybe in a harmless interesting way. The other three sound just awful.
As someone who tends to be "that relentlessly match-making friend"...to me, dating docs aren't so much about preference filtering but a *distribution mechanism*--if I don't know that you are actively looking for a partner, I am not likely to look out for someone for you!
As unromantic as it is, the probability of finding a compatible match has a large marketing component to it...and classic marketing issues like distribution and awareness applies.
I have a lot of nostalgia for old OkCupid, and one thing that really ought to exist but doesn't is a nonprofit dating app with the sole goal of being _really, incredibly good_ at matching people. It seems like a relatively inexpensive way to create a quality of life increase in US cities, because dating is a really big but tractable allocation problem right now. It would be really "easy" to ingest a list of hard preferences and return matches in the area guaranteed to meet those. It doesn't solve all the problems of modern relationships, but it would be exponentially better than the current state of the art!
Not sure if this is a good epistemic rule, but, in a situation where Scott and Gwern disagree, I'm leaning towards Gwern with low confidence. Given that this discusses dating, if Jacob Falkovich weighs in, I would go with his tiebreaker.
Where do you post them? I saw Chris Olah's dating doc because it was posted on Marginal Revolution, but this will not work for people who have zero fame. Is the idea to become at least a little bit famous?
Do people who use dating docs also use swiping apps? How does their experience compare? Since Scott seemed to meet people on OKCupid, it seems like that at least used to be about as good as a dating doc, although now OKCupid is hot garbage ime
What do you write on dating docs that you couldn't easily convey in a swiping app? Many of these apps allow you to post your political affiliation, race, desire for children, and include a bio that shows any other traits you wish to see. Many dating docs that I've seen are quite similar to a profile on one of these apps
Do dating docs mostly help you date friends? This seems like the main benefit, in that if you create a dating doc, then your friends will probably be the first to know about it and it's a way to signal that you'd potentially be open to dating them.
Are there any people who exclusively do dating docs and do not use apps? The type of person who would make a dating doc is at least potentially interesting given the effort put in, but I'm wondering if I would see that type of person eventually on an app. This is probably unanswerable in general, but only via anecdotes
The problem with the above argument is that is slides from a few (and it's important that they are just a few) points that almost everyone everyone agrees upon ("I want a human <gender> who speaks my language and shares my culture") to the points that are actually in dispute, namely that people know what they want for almost everything else.
gender, language, culture. No-one is disputing these. But what people also think is important includes age, children (wanting to have in the future, not wanting partner to have any, etc), and a variety of "vanity culture" signifiers (being vegan, must like the outdoors, must be a fan of Phish, etc etc).
Of course there is imprecision here, such is life. What I consider a "vanity culture" signifier might, in your case, you are the one special snowflake unlike all others in the world, be an actual deep "cultural setting", so deep that it cannot be overridden. But the way to bet is that it's not.
Ultimately those of us who are older, or happily paired off, don't much care. You want to waste 20 years of your life and suffer a grossly diminished dating pool when you come to your sense, go ahead.
But there might be some wisdom in assuming that there is something to our accumulated years of experience rather than dismissing it out of hand.
The Western fad for assuming that the olds are idiots who know nothing of "true" love or "real" sex or how to give meaning to life has a long dreary history beginning in at least the mid-19th C. Yet not one of those rebel generations, from silents to beats to hippies to millenials, seems to achieved anything significantly different or worthwhile compared to their predecessors. Perhaps the accumulated experience of the ages is, in this case, mostly correct?
The below article cites some research favourable to Scott's case. A quote -
"Research in laboratory settings consistently shows that what people say they want in a partner has virtually no bearing on who they actually choose to date [citations] And yet, once people are in established relationships, they are happier when their partners match their ideals [citations]."
I was under the impression that love strikes like lightning from the blue, unbidden and unpredictable, rendering the sufferer powerless - and unlike Lichtenberg figures, the scars last forever.
It's always fascinating to learn about the different models others have of seemingly commonplace phenomena.
During the sexual-pairing phase of my life (I'm beyond that phase now), I was never able to figure out which women attracted me. Some just did — like a thunderbolt or cupid's arrow hitting me — but the vast majority didn't. I was never able to understand which particular physical or personality traits attracted me. Those hidden variables defeated all my attempts at self-reflective deduction. Worse yet, even when I made the effort to get to know the women who I thought should attract me (liberal-smart-educated-creative-regular facial features), I discovered that getting to know them didn't increase my attraction. So, given that one in a hundred women with some hidden variable which I couldn't pin down attracted me, I never gave dating apps a try.
I think what criteria you filter on would depend on what your goals are. A lot of people, at any one point in time, are looking *for* a one night stand and are trying to avoid commitment. But the real problem are interaction effects--I know many people use sex as a gateway to more committed relationships (that is, sex is considered the first or early step on that path). Therefore, sex appeal would be a necessary but not sufficient criterion for a relationship, followed by whatever personality characteristics you're into.
How much research is there on the olfactory side of things? The people I find attractive in photos does not fully overlap with real life and for my wife and myself, smell has a huge influence on attractiveness, and possibly for other relationship variables.
I'll chime in here with a rough description of the dating process in my sub-section of the Orthodox Jewish population.
- Every single has a resume/profile. These can be extremely spartan with just basic bio info like name/parents name, DOB/place of birth and residence, school history, current employment details, or can be more detailed with information about what the person is like and what they are looking for. A picture of the person may or may not be included. But every single profile will include references.
- A "shadchan" (matchmaker) will have the idea that Jacob might be a good match for Rachel. She (the vast majority of shadchanim are women), will suggest the match.
- Jacob and Rachel's parents or other close relative or friend will call the references on the profiles, and will additionally try to find people who know them and are not listed as references (the expectation being that people on the resume are vetted friends who are pre-disposed to say nice things about the person).
- They will be asking about most of the following: What is the single's personality? What is their level of religious observance? What kind of home did they grow up in? Are there any medical or mental health issues? Where do they want to live? What kind of lifestyle are they looking for? They will not ask whether or not they want to have children; that is considered to be a given for any single choosing to participate in this dating system.
- If both camps like what they hear, the shadchan will coordinate a 1st meeting.
- The 1st date is formal. The man will take the woman to a nice bar or restaurant, and they will spend around 3-4 hours talking. There will be no physical contact, not even a handshake or hug. This is almost always purely a vibe check. Not much of substance will be said or shared, and the expectation is that there will be a 2nd date unless one party really doesn't like the other person. The shadchan will coordinate the next date as well.
- The 2nd - 5th date will be spent verifying compatibility on the smaller things you can't ask about in reference calls. The minutiae of religious passion and observance, interests, lifestyle preferences, how many children you want to have, are you actually doing well in your career, etc. At any point past the 2nd date it is considered appropriate to end things (via the shadchan) if you just don't enjoy spending time with that person. Also at any point past the 2nd date the level of formality of the date is entirely the choice of the couple themselves.
- The 6th-10th dates are for DMC's and chemistry building. At some point in this stretch the couple will stop using the shadchan as a go-between and communicate directly. If either party wants to end it, they will have to end it directly too.
- Beyond the 10th date the couple is assumed to eventually be getting engaged. This usually happens in the 12 to 18 date range. Couples taking longer than this is usually due to commitment issues on the part of somebody, family difficulties, or some characteristic on the part of one single that the other one is trying to get over.
- After engagement, the couple will get married within 3-5 months.
Results: (non-scientific numbers)
- 80% marriage rate by age 30, 95% by age 40.
- 5-10% divorce rate.
- 70% happy/functional marriage rate.
Context:
- We are raised from birth to see getting married and having children (as many as possible) as a pre-requisite to living a religiously correct and happy life. Our social structures encourage this at every level, and there is very little comfortable space for older unmarried people.
- We are raised and educated gender-segregated. We do not socialize with the opposite gender who are not our close relatives.
- We are raised from birth with firm gender roles. It's the man responsibility to provide, and the woman's to raise children.
- We are taught what to look for and value in a marriage partner: religious compatibility, someone who is kind to us and others, someone who will be a good parent someone who can fulfill their assigned gender role, someone who we enjoy spending time with. Men are told to look for a woman who can make them feel masculine, woman are encouraged to look for a man who makes them feel feminine.
- We are told to avoid getting hung up on physical attraction at the early stages. Attraction will grow naturally as you spend more time together and your chemistry increases. Declining to meet someone based on just their physical appearance is heavily frowned upon.
- We are also told to avoid getting hung up on any number of minor things: How they dress, if they talk too fast, if they work an uncool job, if you don't like someone in their family, etc. We are taught that all of those things will fade or matter less as you get to know each other and as your chemistry grows.
- There is zero physical contact throughout the dating process until marriage.
- The man is expected to pay for everything throughout the dating process (if he is young or under-resourced his parents will usually pitch in).
- The parents of the couple will pay for the entire wedding (with community support for those needing it) and will also help setup their first home with furniture/appliances/dishes and the like.
Why did I write this megillah? Because Scott's argument is entirely correct, and should actually be taken much farther than he did. People need to be preference-maxing. It's easy to get married and stay happily married when you know what you want and have the tools to find potential partners who want the same thing.
Are we convinced that the OKCupid algorithm was actually more accurate than some other simple sorting algorithm like matching people by IQ or something?
There is a considerable amount of research that indicates that people regard a quick spontaneous response to be more honest and authentic than one the other person took their time and thought carefully about the options. For most people I have met, intellectual forethought is considered suspicious. Initial impressions are trusted more than calculated ones. There seems to be an idea in popular culture that one's intuition is a better gauge of oneself and one's own needs and desires than objective consideration. Given that, it isn't surprising that a great many people believe that you can't objectively articulate what you want or need in a partner.
This could be a case of the self-fulfilling prophesy. That is, two intuitive people, both of whom believe in first impressions and spontaneity as a sign of authenticity, may actually be more compatible with each other than they would be with someone more rational.
But rational people might be well advised to filter on the exact opposite set of criteria.
> And all of this is separate from the types of preferences mentioned above - ie it’s not just the easy things like race, religion, income, number of children desired, politics, sexual compatibility, etc. Everything here is after the 1/500 even-getting-started number listed above! So learning about people from profiles must allow an even stronger filter than that!
I really don't think all of this *is* separate from the types of preferences mentioned above. What are the odds that Cindy is poly or willing to tolerate a mostly asexual partner?
OKCupid was a huge nerd snipe of a site, as it fed the introvert's tendency to think the answer was just around the corner from answering another hundred compatibility questions, rather than in just messaging and meeting up with people. In reality, most people's dealbreakers are all correlated with each other, and you can easily get an idea if someone is in your "market" from five photos and two paragraphs. Scott has much more idiosyncratic criteria than the vast majority of people, but even still there's only a couple questions that will essentially narrow down the pool with a high degree of confidence.
A worrisome trend is that corporations have adapted the concept of dating docs to employment. Using so called "job ads", they publish stringent criteria they have for employment and have the job applicants bear the load of checking for compatibility. This is despite Science(TM) showing that these formal criteria (like having an MD, or truck driver licence, or a PhD) have no correlation with the outcome of three minute long HR interviews.
Seriously, I think that dating docs are mostly useful in situations where there is an asymmetry between the supply of both prospective partners. For example, the rationality crowd seems to be overwhelmingly male (84% cismale on the 2022 ACX survey). Or I would expect that high status individuals within that group (e.g. Scott, EY) have probably way more people interested in dating them than they can possibly date. In such cases, dating docs serve multiple purposes:
* They advertise the fact that the writer is looking for dates (which can override the common courtesy of not cold-messaging people regarding romantic intentions especially in settings where universalization of such a behavior would see them flooded with such)
* They pre-filter candidates so that flooding does not occur
So dating docs sound like an obvious positive sum action for all the parties involved (except for more brazen people who would have cold-messaged their target audience before and now (1) have to compete with more considerate people and (2) may be summarily told "RTFM, you do not qualify".
Even for the ones filtered out by the criteria stated in the dating docs, the process is actually helpful, because it gives them feedback on a process which is normally completely opaque. If nine out of ten prospective partners state that they prefer a provider enabling a lavish lifestyle or physical beauty or BMIs below 20 or sizes above 1.9m, this is actually helpful information. Perhaps not directly actionable (unless one is willing to embark on a career in the drug trade or have ones leg bones enlarged or whatever), but still helpful (perhaps the correct choice is to search elsewhere for partners or romance a LLM or whatever).
Of course, one problem is that openness about certain criteria is generally frowned upon. Writing "your net income should be at least 100,000$ a year" is not the optimal way to filter for high earners, because many people (including people who qualify) will consider this criterion indicative of a shallow character. I think much of the traditional dating strategies revolve about maintaining plausible deniability about such things (which is why paying for dinner in a fancy restaurant will lead to better outcomes than just sending a copy of ones the bank balance).
One solution here might might be to provide aggregated information about the previous relationships. "my previous partners made between 100k and 130k a year" sounds at least less shallow than "I won't date anyone who makes less than 90k". One might still appear less shallow if one lied about that, though.
In general, I have two extreme models of the dating market, with the truth probably in between.
The one-dimensional model would be that most people of one gender, orientation and phase of life desire similar qualities in their partners. Few people are attracted by poverty or ugliness, after all. So for heterosexuals, you can order both by their hotness score, and the stable equilibrium will be the Nth hottest woman dating the Nth hottest man.
The high-dimensional model would assert that romantic potential is determined by lots of different preferences. People want partners with shared interests, similar (or complementary) kinks, members of their own subcultures and so on.
Obviously the real dating markets are mixtures of both, but I am totally puzzled as to regard what mixture. Any suggestions?
I built a dating app that does boolean matching based on users' selected dealmakers & dealbreakers to filter matches (the app is now mothballed). Suffice to say the obvious that the format of the app is insufficient unless you have a billionaire friend that is willing to give you a lot of money for marketing. If no one is on the app, then it is hard to get traction, no matter the algorithm.
So if someone is thinking about building such an app, either be really young and build it over the long run, or have a clear business & marketing strategy that leads to user growth despite initial users having few matches.
Such a document could have prevented the Jonah Hill situation - the text messages his ex shared were all about incompatible expectations for partner behavior.
I think the mistake here is confusing two questions:
* Are people most attracted to people who match their stated preferences?
* Are people most likely to date people who match their stated preferences?
Sadly, people don't get to date the people they are most attracted to. Imagine if the undergrads in any of the speed-dating studies were sent on a desert island and left to incubate until they're somehow all paired up. When asked about their preference, maybe every girl said "I want a tall boyfriend" and every boy said "I want a girlfriend with a 0.2 waist-to-hip ratio". But, at the end, the most attractive girls will pair with the most attractive boys, and you will have no correlation between stated preferences and who people actually date. However, you will find a high correlation between the two partners' objective attractiveness.
In that regard, the utility of date-me docs is not to find people who would match your preferences, but to find people in your league.
I look forward to your upcoming column “Yet Another Bay Area House Party Redux,” starring Hana, Jane, Larissa, and Sky (Cindy makes a cameo appearance, decides the party is boring/weird, and leaves after 5 minutes). Please do it! The ACX commentariat will be eternally grateful. 😊
Dating docs are an attempt to solve a problem (the lack of partner). Unfortunately, trying to solve a problem = signaling that you have one; and that is low-status. The people making fun of dating docs are signaling that they don't have this problem; and whether that is true or not, the signal is high-status.
Note that reading or writing dating docs implies a preference for reading or writing. Some people hate reading and writing, so this solution wouldn't work for them.
I wonder how honest can people actually be in their dating docs. People are hypocritical -- maybe you want to date a person of the same race, and your potential partner also wants to date a person of the same race... but your potential partner may still perceive this information in your dating doc as evidence of racism and reject you. (Maybe they actually don't mind you saying this, but are still afraid that their friends might judge them for dating a racist. Problem is, your dating doc is a public document.) Talking about your partner's income may be a similar problem: you may want to date someone in the same social class, but don't want to be seen as a person for whom money is so important. Among woke people, expressing a preference for your partner being cis could be a similar problem. So a lot of important information may be unmentionable.
In a comment in this thread, Jack mentions preference for "kindness, honesty, and emotional stability". Can these things be talked about explicitly? Do honest people write "I am honest" and dishonest people write "I am dishonest" in their dating docs? Does anyone admit being hateful, or a psychopath? It could even be the other way round, a normal person admitting "sometimes I am not as nice as I would like to be" and a psychopath saying "actually, I am perfectly nice all the time". Or maybe admitting that "I am a little bit X" actually means that person is X a lot. In other words, you probably can't trust most information like this.
Dating doc is a potential weakness, e.g. if you are an attractive girl looking for a long-term relation, but there are many guys wanting to have a one-night stand with you. You just told them exactly what they should pretend to be!
Also, knowing yourself (sufficiently that you can describe yourself in a document) is quite difficult. I might be able to do it now, but I doubt that I would do a good job in my 20s. But maybe it's a part of the process of learning: figuring out what went wrong, and updating your doc accordingly. For example, in a parallel universe where LessWrong and the rationalist community never existed, I would have difficulty describing that I am looking for a *rationalist*. I would probably write something like "looking for an educated and open-minded person", but that's not the same. It is not even a narrow filter because many people are educated and almost everyone considers themselves open-minded.
This probably sounds too negative, but I actually like the idea. I just notice the imperfections. Freestyle description sounds nicer than checkboxes: I get more of the feeling that I know the person. Also, when other people design the checkboxes, sometimes you just don't fit any category; text allows you to describe it using your own words. We should probably use something similar also for finding friends.
Would you necessarily want to restrict the potential candidate pool *before* meeting them in person? I think preferences are a complex interaction--I may weight certain personality traits higher than physical attractiveness, but the weights aren't "1.0" and "0". So it's entirely possible that a sufficiently attractive candidate could make up for a lack of certain personality traits, but only in comparison to whoever else showed up. You can't really predict these things in advance, so perhaps best to use filters conservatively.
Then again, spending an evening with a member of your preferred gender isn't always a complete waste of time, even if neither sex nor romance is likely to result. What else are you going to do for two hours? So maybe it makes sense to be liberal with who you agree to meet.
(Aware that many women are forced to proactively protect themselves from the potential for sexual harassment, so filtering for "creepy" as early in the process as possible is non-optional).
This post is f-ing great! The day before, I was listening to "Tea for the Tillerman". And I need a hard headed women... And I updated my profile on e-harmony, and added pictures. For me the problem with a dating doc is exposure. I live far away from any big city, so number of people is much bigger on e-harmony. (though the interface totally sucks, and I can't put my likes into whatever pigeon holes they've put up.)
> Entry number 1... Hello. I'm Marianella. I'm a Virgo and I'm 15. My hobby is **PoisonGas**! Nice to meet you.
> Entry number 2... Petra the Pisces here! I'm 16 and I love tall, dark boys! I love to sing in the **BlackRain**! Nice to meet you.
> Entry number 3... I'm Carolina, and I'm 10. When I grow up I want to be Miss Universe. I do a mean **Tornado**! Nice to meet you.
> Entry number 4... I'm Richard. I'm 29. And I'm here to protest beauty pageant sexism! My proudest accomplishment is the **BlackHole**! Nice to meet you.
Dating docs are cringey but also kind of inevitable, right?
1. Scarcity exists. There aren't hundreds of men with Richard Feynman brains and Henry Cavill shoulders lobbing pebbles at my bedroom window.
2. The dating market wants to clear. Everyone wants love or at least a good time.
3. Information asymmetries exist. Why not clear the fog a bit? As credit ratings and standardized test scores do.
I'm glad I haven't had to make one! My husband and I got sloshed around in a tiny office dating pool of numerate 22-year-olds, which probably happens less in a WFH world.
Separately, how do polyamorous people with kids have the time and energy to keep going on dates?
I think a large part of why people are against dating docs is two-fold:
- it's not mainstream (yet?) and as such seems, like all online dating, desperate/weird/... . Basically it goes against the fairytale "you will bump into your perfect partner in the library/on the way to work/... and one of you will drop all their papers/books and the other will help pick them up and it will be cute and fate and the universe have this perfectly organized for you and if you try to make it more likely to meet a suitable partner you hate romance and mess up the universe's plans"
- This checklist mentality adopted at large scale would lead to even more segregation than we have now. If I only meet people who are in my social class, I'm in even more of a bubble/echo chamber etc. (I do personally agree that this could have pretty bad societal implications if everyone could filter for a lot of describable preferences/hard facts before seeing a list of people that want to date.)
> Many people want a partner of the same religion and level of religiosity.
This can't be right. Traditional literature/religion is full of indications that men need to be forced to marry within their religion, because otherwise they'll just convert to whatever their wife wants.
And in general - and this is easily observed today - women are much, much more religious than men. Insisting that you want a partner of the same level of religiosity as yourself is much like insisting that you want a partner of the same height as yourself: at a system level, almost nobody can possibly accomplish this, and it's not really worthwhile to try.
I think that most negative reaction on dating docs is sorta emotional and arguments is rationalization. Why it often provokes negatives? First thing - it strikes all "marriages are made in heaven" and "instant strike of romantic love" narratives, and this is still a big thing in our culture - dating docs are much more about "rational partner choosing" than "arrows of cupid". Second - we still mentally more villagers than citizens, and in village being too picky is a bad plan - but much more OK in big city
The way this topic was taught in my sociology class referenced Intimate Relationships by Rowland Miller, Chapter 3 of which I think covers the current research well.
For casual relationships/one-night stands: pretty much everyone's mainly looking for someone physically attractive. If you can get along with them at least a little bit, then you might stick around for a while and call it FWB or a situationship, but looks are still the main criterion for both genders.
For longer-term relationships, here's how it breaks down:
* Everyone's got similar dealbreakers (which are generally pretty obvious: kids, career, smoking/drug use, poly/mono, whether or not someone's very politically-minded or religiously observant and if so which religion/party are they a member of, etc.).
* Everyone's looking for a certain baseline level of assortative matching; the biggest difference between what straight men vs. women are looking for is that men tend to more heavily favor good health/looks and women tend to more heavily focus on wealth/earnings potential, but everything else is pretty similar and there's absolutely a lot of trading-off between how much of each criterion partners satisfy.
* How the assortative matching shakes out for long-term relationships is that they're all *benchmark* requirements: people look for a certain bare minimum (depending on how desirable/attractive they themselves are), but beyond that they're not super-important.
* Once the benchmark's met on both sides, what everyone's really looking for is how well you two get along: agreeableness, compassion, loyalty, and so on.
The latest edition of that book came out in 2022, and Chapter 3's got an *extensive* bibliography. I'm probably not pulling out the right citations, but here's a sampling:
* For similarities in requirements: Thomas, A. G., Jonason, P. K., Blackburn, J. D., & Li, N. P. (2020). Mate preference priorities in the East and West: A cross-cultural test of the mate preference priority model. Journal of Personality. doi:10.1111/jopy12514
* For a summarized list of what people want: Lam, B. C. P., Cross, S. E., Wu, T., Yeh, K., Wang, Y., & Su, J. C. (2016). What do you want in a marriage? Examining marriage ideals in Taiwan and the United States. Personality and Social Psychology Bulletin, 42, 703–732.
* More on what people look for in a casual relationship (not sure which one they were citing!):
* Eastwick, P. W., Luchies, L. B., Finkel, E. J., & Hunt, L. L. (2014). The predictive validity of ideal partner preferences: A review and meta-analysis. Psychological Bulletin, 140, 623–665.
* Eastwick, P. W., & Hunt, L. L. (2014). Relational mate value: Consensus and uniqueness in romantic evaluations. Journal of Personality and Social Psychology, 106, 728–751.
* For people adjusting their standards to meet who they're already dating: Kučerová, R., Csajbók, Z., & Havlíček, J. (2018). Coupled individuals adjust their ideal mate preferences according their actual partner. Personality and Individual Differences, 135, 248–257.
Suffice it to say that I think the research isn't in a completely terrible place right now (knock-on-wood), and that at least some of the criteria are well-understood.
The real criticism of date-me docs, then, wouldn't be that the preferences themselves are undescribable: it'd be that what people put into the date-me docs doesn't effectively depict what matters.
And I think that's a bad criticism, too; I think writing about yourself is a great way to show off your personality, especially if you write a lot in general. (Even if you're just showing off the fact that your first instinct to show yourself off is to do it in writing!)
So really, I think that the main criticism might just boil down to an example of The Worst Argument In The World.
A not-so-hot take: I hope AI-assisted dating will take off someday. I won't need it, I hope, since I'm already married, but other men and women like me are suffering. Tinder-like apps are shit - "matching" people on attractiveness seems like anathema to me. And long-form dating sites are mostly dead and not coming back, unless someone finds a way to monetize them (well, Substack found a way to monetize long reads, which were going the way of dodo with advent of social networks, so there is hope there). But if I had to date again, it would be so, so nice to let an AI assistant comb the internet for a likely woman who matches my preferences...
From reading the article and the comments here, it seems like we have dating apps that are actually good: OKCupid and eHarmony. Why, then, do we hear so much about how dating is broken these days and loneliness is at all-time highs?
I would guess it's a combination of a few things
1. These apps are not as good as people say they are
2. People don't actually want to settle down that badly
3. People's standards are too high
Anyone have insight into why dating isn't basically solved? We can put aside the question of what's wrong with the existing social science and consider the efficacy of the existing social engineering.
I came for the headline and stayed for the glimpses into Scott’s love life.
I think the argument here is correct, but wonder if there is something more to be gotten from the skeptics on stated preferences. I think lots of people (applies to me at various points in my life) overstate the importance of, for example, matters of taste (liking the same music, pop culture ephemera, sports teams, whatever) and how important alignment on that is for a happy, long-term relationship.
Basically I think there’s a bucket of important things you should be aligned on - monogamy, religion, politics, whatever, and a bucket of things that feel like they are important but probably aren’t - Taylor Swift fandom. Of course person to person these categories can change and there is no one size fits all solution to disentangling what is important and what isn’t. That’s probably something best left to your friends to actually help you sort through. But a study that predicts that e.g. shared musical taste doesn’t mean much as far as lowering divorce risk is a decent enough hint that maybe for most people that’s a criterion that should fall in the “not that important” bucket.
People don't think dating docs are weird and repellent because there are no legible criteria to screen potential partners. But these criteria should be a blunt, first-pass tool. If I want an atheist guy in his 30s in DC, and you're a 20-year-old Christian woman in Ohio, we're not compatible. Great. Tell people your basic criteria. But once you have someone who meets those first few basic, legible, criteria...go on a freaking date! You'll be able to tell within 30 minutes whether you're attracted to the person, like the person, and feel comfortable with the person. And you cannot determine this from any amount of reading and writing google docs.
Dating docs are weird because they're incredibly long and detailed, and because the effort you spend writing ten pages about yourself and reading other people's manifestos is effort that could more usefully be spent going on dates, and actually figuring out whether you like the person.
If you like them, feel comfortable with them, want to have sex with them...you'll find that all the details on your 15th page about your ethical philosophy and tastes in video games absolutely do not matter at all.
I think you're overestimating how significant those numbers say political compatibility is. The study said that 4% of marriages are between a Republican and a Democrat, which *sounds* low, but given that something like 30% of people are Republicans and 30% Democrats and 40% Independents, you would only expect 9% from pure mixing. There are 17% between Independents and non-Independents, but from random mixing you would only expect 24%.
Looking at these numbers actually made me realize that this level of political compatibility appears to be far *less* important to most people than I would have expected. It looks like 16% of Democrats are married across party lines, even though 70% of their options are across party lines (about 1/4) of the expected number), and 84% of Democrats are married to other Democrats, even though only 30% of their options are Democrats (about 3x the expected number) so there's something like a factor of 12 between their probability of marrying a particular non-Democrat compared to their probability of marrying a particular Democrat. But a factor of 12 is much smaller than I would have expected in this era of high polarization - and low enough that you wouldn't actually want a filter that eliminates 100% of the people from the wrong side of that line.
Mark my words, soon dating apps will happen in VR. Assuming we get to a point where it can read and show your facial expressions. And possibly upper body language as well.
In fact VR would be perfect for meeting strangers. You can sort of be yourself in an avatar with a different voice (or the same voice), it would be like a real life meet up, except you can end it anytime and move on to the next one.
I noticed this post mentions race a few times, seemingly with the implication that most Americans would only want to date someone of the same race as themselves. Now I'm curious, is that really true, and to what extent? My intuition says there are plenty of people who don't have an explicit preference regarding race, even if they might filter potential partners on other factors like political views. I'm wondering if my social circles and I are outliers on this.
For context: I'm a straight white (ethnically part-Jewish) guy in my late 20s, politically liberal, who has lived most of my life in the NYC area and the Boston area. I don't think it's wrong if someone has a preference towards dating someone of their same race, but I've never had that kind of preference. When I was more actively trying to date and using the apps, I went on dates with both white and non-white women without it ever coming up as an issue. And many of my friends seem to feel similarly.
My grandpa was a medical doctor and he had zero difficulties finding a date or a spouse. Aren't you a psychiatrist, Scott? So maybe your ease in finding a catch isn't that informative. I mean, how many women who are moderately indisposed to poly life would flip flop to snag a doc? Or men.
This is an absurd question, Scott. I don't know if it's the asexuality or what, but the answer to your puzzle, or whatever you want to call it, of the five dating ads is:
1) Which of them is the best-looking?
2) Which of them is the best at giving head?
I don't care about *anything* any of them wrote. None of that has absolutely any relevance for a relationship. Have your weird hobbies, girl!
Speculatively, I think dating docs (as opposed to broad categorical pre-screening) will, if they grow in popularity, have the effect of essentially making women increasingly convinced that few men if any are good enough for them.
I would push back on the claim that a public "date me" profile reduces pipeline and exposure. I suspect that a well formed and publicized one actually both gets you a much better top-of-the-pipeline, and acts as a more efficient filter. You and your extended social network are probably much better about targeting single people with your profile than any dating app would be. So way more eyeballs on first pass, much better chance of multiple exposures to the same eyeballs (it's still a weird enough thing that folk might read and talk about your profile multiple times; an early adopter advantage), and network effects (they'd be a perfect match for Jane! Let's introduce them)
I spent years on OkCupid in two countries and never met Hana. Surprisingly low amount of resulting relationships. But I did find that free text profiles gave some sense of the person.
In my experience, the chance that Hana also loves Taylor Swift is surprisingly high, relative to what stereotypes would suggest. Swift is actually hip with rock critics.
Back when I used to use dating apps, similar music preference + photos were uncannily good at predicting compatibility for me. I think music taste can reveal most other preferences fairly quickly like for instance where you prefer to live, religion, interests, personality.
There's definitely an uncanny valley of people who say they're into anime yet pluralize 'animes'.
> Except for the twin study, all of these focus on initial attraction, not on which relationships work and survive. They don’t rule out a situation where the “initial spark” of romantic attraction is random, but people with similar interests and personalities are more likely to stay together. Maybe everyone in these studies is very stupid (cf. they’re mostly undergrads), and they all feel attraction to random unsuitable people at speed dating events, but in real life, selecting people who are long-term compatible with you is the way to go.
Something along these lines, minus the 'very stupid' and most of the 'random', seems really obvious to me. *Of course* short-term attraction is based more on hotness + charisma + other 'shallow' instinctive preferences than on a checklist of deeper qualities or life-plan compatibility! And of course that doesn't imply those deeper qualities and preferences are unimportant for long-term relationship success!
I think the main issue there is that all of those women sound like insufferable cliches. I mean, I know you exaggerated them into insufferable cliches to make a point, but 1) I don't think you're far off, and 2) if I saw any of those profiles I wouldn't read it as "I am this kind of person." I would read it as "I am choosing to represent myself as this kind of person, because I've found that this kind of person gets dates with people who represent themselves as the kind of person I want to date." Which is different. And a dystopian hellscape you can't opt out of if you want to participate meaningfully in the dating market. I'm mostly just glad I'm married.
Lots of obvious counterarguments ("What? People in bars don't lie?") but I think this is a different case. The issue is the same as all social media - the incentives are to go broad and shallow. My friends who are in the dating scene feel big pressure to present themselves as a type. There's no room to be a driven lawyer who is also a Taylor Swift fan.
The thing that rubs people wrong about these dating docs isn’t so much that people express preferences or even that people try to make themselves sound good (this is bog standard for a dating profile) but that there is an implied “look at how high status I am” embedded in the assumption that lots of people are going to open your particular document and read through your personal essay to determine whether you would think they are a good match for you. On a normal dating site (or in real life) there is a certain symmetry: you see their (short) profile and say “yes”, they look at your (similarly short) profile and say “yes” back, then there’s a conversation that eats up time equally for both of you. It’s arguably true that requiring potential suitors to sign off on your 26 points of agreement before even starting to talk is more efficient since there are fewer false positives, but putting the burden of figuring that out on the other person just feels wrong to fairness-obsessed humans.
I think the docs are a great idea. Swiping apps are horrible at connecting people. I am sure some company will streamline it soon, too. As your samples show, people are trying very hard to sound exactly the same as one another. The docs enable people to highlight their differences.
Shouldn’t dating “success” be based on marriage length, lower divorce rates, happiness level after long periods of time, etc. and not just whether your individual preferences are being temporarily met? This seems like evaluating a diet based on whether you enjoy the immediate taste of the food and not on whether you are actually losing weight.
Am I the only person who would go on a first date with all of the fake dating profiles mentioned, aside from the first? I’ve been partnered for a while but #2-#5 remind me of people I actually dated at some point, and enjoyed the time with. Although with #5 it could be really hard to keep a conversation going unless you happened to both love the same obscure anime.
(I wound up marrying #4, fyi)
98% of whites marry other whites? Where did that figure come from? The source states that 90.6% of white newlyweds in 2010 married other whites.
I am very likely an anomaly, but my dating life has been WILDLY variable. My dates/partners have had about a 40 year age span, different classes, education levels, drastically different Big 5 spreads, different races, nationalities. I don't feel very surprised when I hear about these things not mattering in practice.
I obviously do have preferences, but they are subtle and flexible enough that they can be satisfied in ways that look superficially dissimilar.
Maybe I'm just a mutant, but I'm a mutant that aligns with the science here, so?
Also, I think you may be missing that people sort of merge when they are together. Like I would love god to tell me how many of those 96% matching political people shifted their politics specifically in contact with their partner. It doesn't surprise me that 96% end up converging, but that doesn't tell me how they started, and I bet that matters for politics and a lot of other dimensions that can change over time. A different example is that my first wife was a working class high school dropout, but also blazingly smart and hard working, and in contact with my upper middle class intelligencia sensibilities (and a fair bit of cultural metis), she got an advanced degree in math and now is a math professor at a local university. That sort of thing is not captured in the data.
Is eHarmony still in business? Thier model is explicitly the OKCupid thing but with more "designed by a phd" and "aiming for marriage" veneer.
On this, suum cuique.
I would not have preferred to know my wife’s preferences before our first date, and I’m pretty confident she feels likewise. That was the late 90s, but I don’t think that dynamic would be different for us now.
Moderately liberal democrat and irreligious NY+Miami, conservative evangelical old Yankee family. The only real thing we both had in common when we met was science as an avocation. We grew as adults, and became way more compatible, then got married. The change from who we each were ~25 years ago to now is stark. The same dynamic was even more true of my parents, this month is their 50th anniversary.
The only information that would have been a dealbreaker before or during a first date? Smoking. Fortunately for both of us, she quit cold turkey about a month before I asked her out. Re: a describable dating preference, for me it would be fine to learn this during a first or second date, rather than published in writing.
Perhaps these are too anecdotal to have any value for reasoned debate on the principle of dating preferences. But within our spheres of friends, most of whom have been married once and still are, the degree of political, religious, and cultural change each person has traversed between 20-40 years old is between “a large degree” and “a difference in type.” I hold to my mom’s advice to me, with my daughter: date different types of people, to explore and learn what you are looking for and esp what you can’t abide.
I suspect a lot of the objection is just coming from the tell-don't-show style that many of these are written in, and foregrounding the boring-but-important stuff instead of putting it off in a corner somewhere. I recall Aella complaining about how may of these are tell-don't-show and should be more show-don't-tell, and claimed hers was, but on reading hers, I still found it to be tell-don't-show! So, yeah.
I feel like someone needs to start a poll about these five women.
To me it seems like Hana (4) is awesome, Jane (5) could be a good friend, and the rest are terrible.
It's hard for me to imagine anyone else feeling differently.
I am reminded of a novel, The Rosie Project, by Graeme Simsion, about a man who's tired of deal-breakers turning up among his dates (e.g. his dates habitually don't tell him that they're vegetarian until after arriving at his house for dinner), so he compiles a questionnaire that all prospects must answer. His friends futilely try to explain that this is not how you go about it.
I guess sometimes you might not know that you would like someone with a given trait until you meet a person with that trait. Also, people change their preferences, sometimes influenced by people they love. The problem with prefiltering on a set of criteria is that you might miss an opportunity with a person which fails on some of your preferences, but scores very high on preferences you did not know are also important to you.
The problem here isn't that profile description is a bad filter - it's just that mutual physical attraction is a much more reliable one.
Text profiles are hard to verify, reflect self-image more than personality, and it's still a cointoss whether you two actually vibe IRL. Meanwhile the photos are pretty foolproof - if someone shows up on a date not looking like the photos, the jig is up straight away. And it takes a split-second to judge a photo, when reading a dating doc takes a minute or more - making it faster to find your 1/500 person.
With all the dating apps switching to a swiping-based system, it seems like using apps as a first-pass attractiveness filter is a much more workable model than expecting them to sort people by compatibility - which was way easier 20 years ago where online dating was a niche for a specific kind of person.
In fact, from what I see, the vast majority of dating profiles you see on apps like Tinder don't resemble any of five examples in this post. They're simply empty or have a pithy one-liner. It's very easy to discard these profiles as "not taking dating seriously enough" or "what do I even talk with her about".
I strongly recommend in favour of attempting to match with these profiles anyway! Empty bio does not equal empty brain. Some of my best dates came out of matching with these wildcard profiles - purely on physical attraction alone. If there's no long-term compatibility, that's okay, spending an evening in the company of a stranger who finds you attractive is still better than staying at home. And anything more than that comes as an unexpected bonus.
The long-form dating profiles found on sites like OKCupid and PlentyOfFish were great back in the 00s, but they lost out to swiping apps and their ease-of-use because at the end of the day the only truly key issues are looks, money, and status (and also partly industry consolidation). Unless you are looking for something really specific, like a religion or someone into polyamory, nobody much cares about reading your carefully written profile--especially all the cliched ones.
There's two questions here:
1. Should you have expressible dating preferences, and
2. Should you actually express them?
I would say that of course yes, it's reasonable to have dating preferences, but you should be very careful about actually expressing them lest you alienate or repel even the people who satisfy them.
I think I'm in love with Hana. Or possibly the internal simulations of Aaron Smith Teller and the Comet King have hijacked by brain's attraction centers to make me think I'm in love with Hana. God dammit.
OK Cupid used to have a blog called OKTrends where they did statistics with data from their userbase. One post I remember in particular looked at the match questions that predicted resulting relationships (which they knew from a survey when people disabled their profiles and answered, I met someone via OKCupid). They suggested that the questions people thought were very important (religion, smoking, etc) were less predictive of relationship-forming that some others ("Do you like scary movies?"). Here's the link: http://web.archive.org/web/20110315063632/blog.okcupid.com/index.php/the-best-questions-for-first-dates/
Jane has remarkably narrow tastes in anime for someone who likes anime. Someone should introduce her to Monster and Nodame Cantabile. But she sounds like a nice person.
You should see the matrimonial section of Indian newspapers. However, I hope you won't mock them.
They're very earnest and absolutely open about things Americans would find non PC.
Often, the parents are placing ads. I only saw about 10 of these and one was an ad for a gay son - the mom was looking for a man for him.
Westernized folks there wouldn't deign to use ads to find spouses for their offspring and while I think they'd have very similar preferences, they'd pretend to be very open minded. Therefore I find these ads charming.
I live in America and every time I visit India I browse a few ads in the papers because they tell me so much about things that have changed (or not) - since I grew up there.
Hi, I'm an ultra-orthodox Jew, and I would like to point out a few things about "shidduch resumes."
1) The single biggest thing that's looked at are the schools the person attended. Ultra-orthodox high schools are generally selective, and most schools have a defined stereotype and perceived rank. This becomes even clearer after high school, when the boys go to yeshivas that are very often smaller than most high schools (my options for yeshiva were ~25 guys, ~30 guys, & ~160 guys), leading to even more sorting. Girls go to 'seminary' after high school and similar dynamics are at play for them. In addition, prospective in-laws are now able to ask the staff of the school about the person. This works pretty well only because ultra-orthodox jews generally get married less than one year out of the religious school system.
2) Another useful feature of these resumes is to provide references who are willing to talk about the person, which allows prospective in-laws the chance to see who their friends are, and how their friends describe them without meeting the prospective partner.
3) The basic outline of the reference will also tell you if the person has deviated in any significant way from the standard ultra-orthodox educational path, which is then interpreted in different ways according to personal preference. (If someone is currently in Yeshivas Brisk in Jerusalem, the path they took to get there may show that they are not a typical 'Brisk Bochur,' for better or for worse.)
4) Ultra-orthodox parents often support their sons-in-law in Torah study for several years after marriage, the resume may indicate how long such support is being sought or offered.
5) It also bears mentioning that in the orthodox jewish world, parents are heavily involved in their childrens' dating lives, and the possibility of comparing potential partners is antithetical to all of the social & religious norms surrounding the dating process. People in the shidduch process are not supposed to see one profile that says "I'm in pre-med" alongside one that says "I'm a secretary in an elementary school," instead, resumes are given to parents. Parents reject any that are extremely obviously incompatible, kind of as in Scott's example, and then if they receive one that seems to make sense, they investigate. This includes general fact-finding (personal history, looks, family, sub-subsectarian religious affiliation, etc.), reputation checking (are they a 'catch'?), and trying to get a sense of whether they would be likely to hit it off. The potential partners are then supposed to consider the combination of resumes, 'backround check', and any other information and decide if they are interested in dating each other. If they agree to meet, the partners are understood to be interested in the possibility of marriage; at this point the potential partners are now running things, mostly. This extremely high degree of parental involvement may make the high level of filtering easier than it would be for the person themselves, who may have a bias toward saying yes.
6) It is also worth noting that the general sentiment of the community is that resumes are a necessary evil, and we commit a gross injustice when we attempt to pin someone down to what can be stated on a resume.
7) Finally, it is worth noting that resumes are a relatively recent innovation, and while I am not certain when they became common, they definitely were not a feature of jewish life in prewar European communities.
TL, DR:
The orthodox Jewish process is a lot more complicated than resumes, and resumes are not a longstanding tradition.
This is absolute insanity. I have never found a girlfriend on a dating app who was in the same country as me, let alone one who was in a similar age range or had remotely similar views as me. Or maybe it's just that guys with Muslim names either have to be super-successful, or date Muslim girls by pretending to be super-religious.
I’m more of a Hana-Jane type of guy. Were those biographies written with AI?
My experience having a dating doc: with one or two exceptions, everyone whose emailed me about it has been unusually interesting and compatible with me (I think less based on specific listed preferences than on being a good match for my writing tone). The exceptions were also alright in that at least we had a more interesting initial conversation than I usually have with a random app match, because they were working off knowing my interests.
I've also had one or two people I matched with on a dating app that clicked through to it, saw I want kids, and cancelled the date on me (which wasn't fun, but probably a net positive).
For what it's worth, my marriage is a statistical anomaly. I'm an atheist liberal Democrat with an undergraduate engineering degree, and my wife talks to ghosts, voted for Trump, hasn't finished community college, and, most astoundingly of all, does read books but does not like Terry Pratchett novels.
I started dating her around nine years ago. If you had asked me back then what I wanted in a girlfriend, she would have checked almost none of the boxes and probably should have set off a lot of alarm bells, but when you're 32, not employed, living with your parents, and have never had a girlfriend before, when someone shows interest in you you're inclined to give them a chance; it only took one date to decide she was a keeper, and ever since, in spite of problems that would probably make a sane man or woman run for the hills, I've never been happier or felt like my life has been more meaningful.
(Btw, she did give up on Trump after Jan 6.)
Different take on why people react negatively to dating docs:
For some people, writing a dating doc is socially similar to saying "I'm smart".
People don't think you're smart because you say you're smart, they think you're smart because either
1. You convincingly "perform smartness"
2. You demonstrate social proof of smartness (e.g. you have a PhD in Mathematics from
Failure to do either of the above make people think you're either too stupid to understand the rules, or trying to deceive people too stupid to understand the rules.
I see two main (overlapping) groups of people for whom a dating doc will trigger a similar reaction:
1. If the reader lives in a low trust environment (maybe due to their filter bubble), the idea of telling people who you are will trigger the trust issue. They believe you know that only a gullible person would believe this doc, and thus you are manipulative.
2. Many people are not comfortable writing structured documents - to them this sort of guide is likely to come off as trying *way* too hard, which signals either desperation or deceptiveness. The idea that this is something someone could casually type out never cross their mind. (perhaps adjacently ,the idea of solving problems through structured documents may be alien to them)
Now here's the twist I find interesting - even if you're able to understand why these documents may be a good idea in principle, if you have the emotional reaction that sharing this doc was a rather large social faux pas that is generally going to short-circuit any "pro dating doc" argument because people filter *hard* against people who do things they consider large social faux pas.
I get all those arguments, but my personal experience with any kind of online dating was not so good, mostly because the subconscious aspects of attraction were missing. Most of the time, I knew after 5 minutes into a date that this is not it.
After my divorce, going to events where like minded women could be expected worked much better. For me it was tantra, and the criteria were interest in self development, a certain openness in relationship, willingness to grow together and an interest in new experiences. I found my second wife and later most of my lovers there, people that I probably would have neglected in online dating
Anyone know if Hana is single?
Since we're doing this about the dating profile stereotypes (I'm married in real life, but whatever):
1) I'd totally give Cindy a chance. I don't go to bars, or at least I don't drink at them, but I did enjoy dancing back in high school and she sounds like a lot of fun. It could go wrong, and I imagine that the most likely outcome is that she rejects me, but I'd still like to see what happens. And I do happen to like Taylor Swift's music.
2) This girl is out of my league. I don't have much in the way of career ambitions and tend to be lazy and procrastinate a lot. I could offer to be a househusband or assistant-type the way a wife might have been in The Bad Old Days to a high-achieving husband, but I couldn't be her equal. I'd probably leave her to other people who are Going Places and just not respond to the profile.
3) Astrology? Eww, red flag. She sounds like a totally sweet and caring person, but I'd ruin it with my intolerance toward woo. Pass. (I don't like my actual wife's belief in the supernatural either, but that's another story.)
4) Yeah, I'm totally interested in this kind of person; I took one economics class in college and the textbook totally changed my way of thinking. Definitely contact.
5) On paper, this is my dream girl. I would totally be down to hang out and let her show me her favorite animes and play video games together. I'd be afraid of coming on too strong, though, or that I'd end up with yet another one-sided attraction to someone who only wants to be friends. Once upon a time in college, I ended up in a conversation with a girl I while we were waiting for a campus bus; when I asked her what her favorite Final Fantasy was and she said Six, my response was a only half-joking "Will you marry me?"
"98% of whites marry other whites; 90-95% of blacks marry other blacks."
Those figures are too high, at least for American newlyweds. According to the Wikipedia article you link to, in 2010:
"Among all newlyweds, 9.4% of whites, 17.1% of blacks, 25.7% of Hispanics and 27.7% of Asians married someone whose race or ethnicity was different from their own."
> [after speed dating] realistically probably people just chose whoever was hottest and maybe most personable
If true, this would be a pretty damning indictment of the usefulness of speed-dating. Instead of going to a speed dating event to talk to people for 3-4 minutes each and then asking to be put in touch with the hottest of them, you can open tinder and swipe right on the hottest profile pictures in a smaller number of seconds each without leaving your couch.
Reeeeeally want to date Hana.
Knowing 'Hana' is just a Scooby-Doo-style mask and if you ripped it off you'd find either Scott Alexander or GPT-4.. doesn't exactly seem to diminish this longing.
Suspect this sensation is not unique amongst ACX readers.
I am currently building a dating app that is similar to Tinder but WITHOUT any chatting. It will launch in about three months.
I agree that certain "basic" info like what the person is looking for (gender, long or short term relationship) is obviously significant. But the rest is largely bullshit. You only really get to know a person in real life.
I'm still reading, but wanted to add to first part, where Scott describes critical incompatibilities. In my reality (country close to Russia), there is most important point, probably. I totally won't be able to date someone who supports Russia in the war with Ukraine. It is just not imaginable.
"For a while, the person in the entire US with the highest match percent with me was my actual girlfriend (who I had met separately, not using the site)."
I'm a bit confused about how this could happen, isn't it extremely improbable that the person you're most compatible with in the entire country just happens to be someone you've also met irl? Seems a bit like when people claim to have met their one soul mate out of the 8 billion people in the world and it just happens to be someone that lives in the same building or works at the same place, rather than say... a village in Mozambique.
On this topic, maybe a bit of an odd question, I've been wondering for a while reading the blog, how Scott manages to have so much success dating people selected on extremely niche psychological traits. The readership for ACX skews about 90% male, right? You'd think there'd be 10 guys similar to Scott for every one woman similar to his partner. I'd guess the competition for guys following this strategy is extremely fierce. Is this something most guys could feasibly pull off, or would you need to be some kind of rationalist chad?
>For a while, the person in the entire US with the highest match percent with me was my actual girlfriend (who I had met separately, not using the site). She told me I was her second-highest match percent; her ex was #1.
Surely this demonstrates that these questions only tease out short-term surface-level compatibility rather than long-term compatibility? After all, your then-girlfriend and her ex broke up, as did you and your then-girlfriend.
IRL you have three seconds to get someone's attention and it takes ten seconds max for them to decide if they even want to talk to you, which is why your first words to them must always be a question. The swipe-right apps are little different. You have only seconds to make an impression. On first contact a salesperson doesn't try to sell the sale. The salesperson tries to sell the appointment. Perhaps there are intellectuals who set aside time and sit down purposely to read through dating resumés, dull though they may be with little attempt to capture the reader from the first sentence. I think that a dating doc should only be communicated after initial contact and a primary expression of interest. At the very least it's more graceful than going straight to intimate pix.
Whatever happened to frustration, rejection, not lying about yourself and dignity? All character building experiences. I guess I'm showing my age, but I'm so glad I don't have to try to date anyone (married for 47 years). I rather think online dating is like social media: you lie about yourself and your life for 'likes' of a more direct kind.
My wife and I were married in 1985. We were 23 and 22 when we got married. My parents were Republicans, hers were Democrats. It wasn't something we ever discussed or felt worthy of discussion.
So the 4% number of cross-party marriages really stood out to me. That's a huge decline in the available pool of potential spouses. Which is a real shame.
Who wants to marry someone like themselves? Many great relationships have a yin and yang that a dating doc would actually filter against. Larisa and her ultra ambitious husband would end up at each others' throats. She'd probably be better with a laid back, good looking, funny guy, who could also look after the kids' emotions while Larisa grew her multinational shoe empire. Maybe I'm old fashioned, but I never felt like I needed a partner to share too many of my interests. She does have to be smart (good breeding genetics), good looking, and from the same social class or higher. But if I am into anime and reptiles then i actively don't want my partner to be into that stuff. It is just too much.
The 5 women described here are hobbies, not people. I can't get any judgement out of them other than "their hobby seems fun/boring". #1 is the only one I strongly dislike, I don't have much preference between 2-5.
I myself find dating docs useful. Anybody who writes one is a hostile tentacular alien to which I am anathema to, as is anybody who thinks they are a good idea. This is because they signal wildly divergent cultural norms to which I view as monstrous, being that they are almost entirely represented in the Rationalist/EA sphere. One could say they identify low match percentage. Online dating profiles more broadly signal only mildly divergent cultural norms from my own, to the extent that it's theoretically possible for somebody I find normal to be found by sheer accident.
I could then assume that the monstrous aliens would love them as a tool for finding each other, if only the population wasn't skewed heavily towards male.
I recently did a quick study where I asked people for their personality-adjective partner preferences, and then asked them what they meant by those. I think the answers were pretty revealing: https://twitter.com/tailcalled/status/1680269225418522628
For instance, one guy who preferred an honest partner brought up the importnace of her not cheating on him. And one guy who preferred a nice partner brought up that he wanted her to take on a housewife-like role.
I think part of the issue is in using abstract terms like "nice", when really everyone wants something more specific. Like the conservative guy wants a nice housewife, you probably want a nice girlfriend who doesn't mock nerds and doesn't aggressively push woke stuff, and a woke guy probably wants a nice woman who cares about the oppressed. You all agree that you want someone nice, but you disagree about what "nice" means.
”How do you translate these to single-dimensional scores on a psychological exam?"
A lot of the things you bring up in your profiles could totally be measured psychometrically by the way, they'd just be in the interests domain rather than in the temperament domain. There are correlations between interests and personality, but they are relatively modest, so practically they are pretty different.
Usually the way you measure interests is by asking people whether they are interested in the relevant things. So for instance, "Are you interested in business?", "Would you be fine with combining vacations and work?", "Would you be interested in going skydiving?", "Would you be interested in hiking in the Alps?", etc..
Probably this isn't gonna be entirely unidimensional (I imagine this profile is combining nearly-orthogonal factors of thrill-seeking/outdoorsy interests and business interests), but that's not really a problem psychometrically (you're probably familiar with the Big five, which is a multidimensional measure of personality). As long as it's relatively low-rank, it's still gonna work fine, and I think you'd be surprised about how much variation you can stuff into surprisingly low-rank models.
For all of you Hana fans: think about supply and demand here.
She's either married to a guy who owns his own tech company, or is in a polyamorous relationship with two of them.
(Or potentially one of the other Hanas.)
If pairings ultimately end up being sorted without the docs then who needs them.? Without them you'll date people from the broader spectrum of humanity and maybe learn. These docs strike me as an effort to sanitize human contact. A contraction of human possibility.
How much of the problem is the fact that the set of single men and the set of single women are incompatible with each other? Single men skew conservative, and single women skew liberal. Small group differences overall compound into large differences at the margin. I think most counterintuitive dating market conclusions can be explained by adverse selection. The most compatible people are already matched up.
I'm surprised at the negativity. For me, someone having a dating doc signals some degree of confidence, intelligence, good communication, awareness of their wants, commitment to taking a potential relationship seriously, and an interest in organising and optimising their life. That's before I've read the contents.
Those are all attractive qualities to me and I would happily filter based on them were I still in the dating pool.
I’ve been married for 38 years (married at 19). This is all so baffling to me - life and love can just happen if you let it - if you enmesh yourself in other humans. The heart wants what the heart wants...
Serious question for all the lonely nerd guys here: which personality type did work for you? (One of Scott's or another.) We all know there aren't enough Hanas to go around, so who did you find happiness with?
> OKCupid, back when it was good.
Oh, man, I miss those days so, so much.
I have a close straight male friend where part of our friendship is acting as one another's dating coach and commiseration buddy. We've both noticed how OKCupid and all its sibling websites and apps across Match have been degraded to the point of uselessness. Someone whose profile I haven't seen yet sends me a message? Maybe I receive it, maybe I don't, maybe it hits my inbox THREE YEARS after they sent it (true story). Someone who I've matched with sends a message? Again, maybe it shows in the inbox right away, and maybe it shows up a month after it was sent, and probably it shows up never.
Dealbreaker preferences, like smoking or parenting status? Frequently ignored.
Premium membership does not fix these issues, either. While I've never paid for membership, my friend has, and he likewise often receives messages weeks/months/years after they were originally sent and also has his dealbreakers ignored. Plus, as a straight man he is inundated by messages from laughably transparent scammer profiles that OKC is absolutely capable of shutting down but does not.
It's so clear that OKC is constantly being tweaked to thwart most daters' ultimate goal: To find a partner and no longer need the dating platform. I don't necessarily mind a business giving users the choice between a shitty free product and a terrific premium product, but OKC can't even do *that,* likely because (duh), once they've identified that someone is willing to pay $40 a month for their profile, then they REALLY don't want that person to be successful enough in dating to leave.
It just sucks. How do you solve the fundamental problem of dating companies and their customers having perfectly opposed goals?
(Use Fetlife, I suppose, but that's a very tough putt for those with weak stomachs.)
Okay, I was dubious about "dating docs" but the comments section has converted me on them. The amount of people going "Oooh Hana my dream girl" while I'm going "she has to make up some silly jokey name for the rooms in her house in order to remind her to buy groceries???? how does she cross the street unaccompanied?" 😦 Being interested in your pet subject or hobby is one thing, being economics-obsessed to that degree is.... worrying. I think you all know I like the works of Tolkien but dang it, I'm not naming the rooms in my house out of his writings in order to remember "I need to change the bed sheets and put the washing on".
Definitely need dating docs to keep the likes of that away from me! (And me away from them, because I'm sure I'm every bit as much not their cup of tea).
(Also I am so glad I have never had the faintest interest in dating. So, so glad. I now see why the sexual instinct has to be so strong, otherwise the rest of you would never manage to reproduce the species if left to your own devices).
On the other hand people are quite capable of rapidly filtering through basic profile information themselves. The big problem with matchmaking isn't the matchmaking part, it's getting enough people in and helping everyone filter through everyone else as efficiently as possible. So helping people put together profiles containing relevant information including, most importantly, their picture, is a good thing, but telling them who they should pair with is not.
This is all making me wonder - what's the male version of Hana? Stereotypes suggest it's "I like Taylor Swift concerts, long chats about feelings, and bringing you flowers", but does anyone know a real organic example that looks like it?
Dear Hana, can we please be friends?
Reminds me of this blog- https://putanumonit.com/2022/09/02/date-me-doc/
Partner trait preference is all well and good, but the instant you turn dating into a game of app algorithm minmaxing or find yourself writing a goddamned resume you've type-A'd yourself and the whole scene you're in into a region where I consider you more an animate object than a human being, for better or worse.
Interestingly, I’m an atheist (Dawkins, Dennett, the whole thing) and my wife of thirteen years is an ordained minister and our marriage is solid.
Apparently some subconscious attraction between individuals in physical proximity is determined by distinct smells caused by variations in their combinations of MHC (Major Histocompatibility Complex) genes:
https://en.wikipedia.org/wiki/Body_odour_and_sexual_attraction
So if these could ever be classified and expressed in a fairly compact set of letters, analogous to blood groups, then perhaps there is a marketing opportunity for a new dating app, or a supplement to existing apps:
Clients registering would mail a cheek swab, from which their relevant DNA could be analyzed. Then the app would pop up possible matches whose MHC differs widely (which it seems is the positive factor, as opposed to similarities). A bonus is that presumably the offspring of a successful match on that basis, in combining differing MHC from both parents, would have an unusually healthy immune system!
But it is questionable whether these instinctive chemical indicators could trump incompatibilities of the kind being discussed here, such as different interests and goals or conflicting opinions on politics for example.
I had no idea how statistically rare my relationship with my spouse is. I'm a Christian Republican with a JD and they are a Agnostic Democrat that dropped out of college. I never would have guessed those were strong sorters. However, both of us have the same energy as Jane in that last profile. I think it might be really hard to use survey's to best pin down what makes people have a good marriage, but it would be great to have more research.
An unexplored angle: That’s also some “is/ought” confusion (and a bit of naturalistic fallacy) in the objection that Scott dive into.
Things like dating docs are an (attempt at) innovation. So, by definition, objecting with “but that’s not how people do it” is more of a non sequitur than it first appears. If Henry Ford showed someone his Model T plans and someone said “but, statistically, most people travel by horse or foot, not by machine,” I wouldn’t begrudge Henry saying “no sh*t, what’s your point?”
Along a similar vein, when divorce rates are 45% in the US, an objection that’s basically “we can’t innovate long-term partner selection in X way because people tend to do Y instead” is pretty weak. Clearly Y has plenty of room to improve, and innovation necessarily assumes a future that looks different than the past.
Here is my take on this: you can infer a ranking of people from their decisions and preferences (People’s choices determine a partial ordering over people’s desirability: https://nunosempere.com/blog/2023/06/17/ordering-romance/). Then, it's not implausible to think that someone using a dating doc reveals that they are lower down that ranking--because they couldn't find romantic success without the dating doc. Therefore, from an evidential decision theory perspective, it's a bad move to have a dating doc**.
** caveats may apply. other decision theories may provide other recommendations.
As an aside: I’d never heard of the ‘dating doc’ before this article, but they remind me of the Personal User Manual idea that I learned about ~5 years ago, which I love and always try to proselytize.
Highly recommend you set one up: https://friday.app/p/personal-user-manual-for-work
As someone currently in the muck of dating apps after the end of a long relationship, I haven't encountered the 'dating doc'. But I think the issue with listing preferences generally is that lists of traits and profiles are bad a capturing personality - and personality determines a lot of the 'spark' of compatibility.
That, and there's a tendency to treat the date like a dry job interview, and it can be tough to break out of that format. People ask formulaic questions, and are guarded about potential flaws and their pasts in ways that don't allow them to act like the interesting and idiosyncratic people they are.
Anyway, I've met a lot of people in my preferred demographics who I don't otherwise seem to have much to converse about, whereas most of those demographics besides gender (age, cultural background, education, politics interests...) aren't rigidly defined and can be flexed.
So clearly these can be done well or done poorly.
Done well, a dating doc is an asynchronous speed dating event. Hundreds or thousands of potential matches can spend 4 minutes with "you," and you do the same, without having to coordinate a specific time.
Looked at that way, a video where you say all the stuff you would put in a doc should also be an option. Less crunchy data transmitted per minute, but 1000x more "vibes"
Hmm. I feel like the negative views of date-me docs don't really have much to do with doubts about the efficacy of describable dating preferences, but rather a concern that writing a date-me doc is a yellow flag for potential narcissism and entitlement. That doesn't mean that this is necessarily a reasonable view, but it at least explains some of the revulsion.
I wanna write a longer reply but I wanna preliminarily say I have no idea why it’d be possible to pick one significantly more interesting person out of those hypothetical girls based on just that, and I really really don’t understand why I so obviously should be. The idea is really strange to me!
I feel similarly about Myers-Briggs as you do about dating compatability docs. It's so obviously useful once you understand it. Existing studies that don't show value tend to be poorly designed.
This quote from Reddit is a sentiment I’ve seen around a lot:
“Jesus. I'm not trying to be insulting at all but even just the existence of a Date Me Doc, or an online dating profile is just so antithetical to how I grew up. (I'm 38) I see people addicted to these apps and now "burning out" from them, it all seems kinda crazy and gamified. The over analysis and "sciencification" of attraction really take the fun out of life, and it happens in every game where people min max to suck the joy out of it. Just talk to people...it is still easy for people from my generation.”
But what it misses — one thing it misses, at least — is consent. If you join a dating website, you can reasonably assume that the people you find there want to find dates and will not be offended if you ask them. Compare that with asking someone out that you’ve met at a local board game meetup or folk dance meetup or whatever: someone very well could be offended by that, since they’re just there to do $HOBBY and now they feel like they’re being treated like a piece of meat. I often see this expressed through a feminist lens — “women go to $HOBBY meetups to do $HOBBY, not to get hit on!” — but it seems more broadly applicable.
"only 4% of marriages are between Democrats and Republicans". I think this means if you get one partner who's willing to state "Democrat" on a survey and the other willing to state "Republican". As opposed to say all those "independents" who still often vote the straight ticket as being easier. Before you object to that characterization, other surveys have found more than 30% of people answering "independent".
"The most popular dating apps, like Tinder, almost push you into that mode." I think it's more that the companies that can monetize their apps successfully have found that the users who go for attractiveness are the ones who pay (or are susceptible to ads or however else they actually get money).
I marveled at the profiles, but then thought better. It's a signaling device - you care about x, and if your potential partner is repulsed by that it's better not to waste both your time. On the other hand, I've had a couple of relatives who ended up divorcing because (probably) not discussing these signals enough - but not before producing kids. Are we better off for not having these now? Unsure.
Whenever the topic of relationship choice comes up, I’m stuck by the question, ‘how do people KNOW this person will be a good match? Dating docs try to solve the problem, but people (or at least some people) seem surprisingly good at it without help. (Finding the person is a separate problem.)
Example, my husbands best friend in grade school now lives in another country though they’ve remained in touch over the years. Their initial attraction was largely based on a mutual love of fart jokes and their accomplishments were of the video game variety. They’ve grown into very different people, both from who they were then and from each other. So it was surprising to eventually meet him and discover how strikingly similar he is to me. We took different paths, but ended up with the same career. We have read many of the same books, blogs, etc. - notably ones of no interest to my husband. I think it is true that this friend is the sort of person my husband is very compatible with, but how on earth did he know that based on information available when they were in kindergarten?
Having thought more about this I have a more coherent objection than my "I don't like Scott's made-up profiles and want to make that a complaint about actual profiles" comment earlier.
Determining dating preference is a multi-armed bandit problem.
For those unfamiliar, this is a problem in computer science/machine learning. Imagine you have two slot machines - one of which has a 40% chance of giving you a dollar and one of which has a 60% chance of giving you a dollar. You want to make as much money as possible but you don't know which is which. Assuming you have limited resources to play these machines you want to spend as much time as possible at the 60% machine, yes. But you also want to spend some amount of time confirming that you have guessed the right machine. Since you can never be completely sure which machine is the 60% machine, you employ a strategy that balances exploration to increase your confidence that you have the right machine with exploitation of the one you suspect is the 60% machine.
In dating you should also optimize for both exploration of your preferences and optimization of those preferences. Some folks like Scott may have a strong understanding of their preferences, plus be extremely high status and able to whittle down a large dating pool by several orders of magnitude and still find a match. Such a person should spend a lot of time exploiting those preferences because they're optimized and possible to exploit. But for most folks, preferences are a guess at what would make them happy in a partner, not a 100% certain formula for partner perfection. You may have dimensions along which you're very certain ("I want to date a woman"), dimensions along which you're not so certain ("I think sharing my political affiliation is important in a partner but I'd be open to meeting someone who thinks differently") and worst, dimensions along which you think you're certain but you're not ("Oh turns out I didn't want to date a woman, something it took me 10 years to learn because I only went on dates with women."). Plus for most folks, some amount of preference compromise is necessary, so knowing what is and is not a dealbreaker is another exploration/exploitation puzzle.
It's not necessarily true that dating docs/expansive profiles/checklists/whatever have to silo you, preventing exploration and putting you all in on your guessed preferences. Someone who thinks critically about their own assumptions about their preferences might use these tools in a different way that actually helps optimize. But in practice, I suspect they do way more often than not.
I simply do not believe that people will give truthful answers to "are you attracted to your wife's sister?" in a poll.
When I wrote a dating profile, I updated it after every unsuccessful date in such a manner as to be less attractive to (the sort of person) who I went on a date with, targeting in particular the ways in which we were incompatible.
It worked; my spouse messaged me because my profile came across as a likeable asshole. (Fair cop.)
Where dating profiles fail to be useful (for the specific purpose of finding a compatible life partner), I think they tend to fail because the users are trying to attract as many people as possible (complicated social reasons for this, but this isn't what you actually want to do, even if you are struggling to find matches now*).
And dating docs are basically just dating profiles that have been divorced from the specific user interface of a specific dating app. All the same failure modes apply.
Maybe somebody should create an industry standard for dating profile documents, so they can be imported into any dating app that supports the standard, using whatever data the dating app supports. I don't think it's a secret that people using data apps are likely to have accounts on several (although probably their actual usage is mostly limited to one, as regional variations in the popularity of specific applications means that a given dating app may be locally dominant).
*I know some of the responses I'll get to this. "I struggle to find any partners / get any responses now, why would I try to be deliberately unattractive to people?"
Well, first, you aren't trying to be deliberately unattractive, exactly, so much as being honestly unattractive; posting a picture of yourself sneezing ... okay, that might work, as I think about it, but maybe not the way you might hope. Maybe.
Trying to attract as many people as possible is what everybody else is doing, second. Your goal is to stand out, yes? If you can get somebody to go "Wait what?" and keep reading, you've won your first battle already, getting somebody to spend more than a second or two to say "Nope" and move on.
Third, trying to attract everybody is, unless your social skills are pretty on-point, is obvious, and comes across as desperate. Don't come across as desperate!
You can come across as desperate if you go the honest unattractive route as well, note. Don't snivel. Don't apologize. The attitude you should take: These are the things you may not like about me, and if you don't like them about me, move on to somebody else, so we can both save our time, because I'm looking for somebody for whom these things are not a problem / are a positive. The attitude you should not take: These are the things nobody likes about me, and I'm sorry I'm such a terrible person, but maybe you can take pity on me and date me anyways. With the caveat that maybe the "attitude you should not take" is the attitude you should take if that's who you are; there are people out there who will like that. But there aren't a whole lot of them, and you'll have more success if that isn't your entire dating pool, so only go that route if that is exactly who you are, and that is exactly who you need to date.
If you're going into dating for the purposes of finding a life partner - then take that purpose seriously. You are trying to find a life partner. Quit messing around with trying to date people who aren't that, and quit trying to date people who are your ideal life partner, but for whom you are not their ideal life partner; even if that succeeds, do you -want- to spend the rest of your life with somebody who is unhappy in their relationship with you?
Mind, there are other reasons to date which are perfectly valid.
Even assuming that humans actually know what they really want in a partner and can/will formulate their desires into words, I get the idea behind describable dating preferences, but I cannot count the number of times a female seemed to check all the boxes and when I actually followed her scent and met her it was like....meh.
Not done reading but I'd love to see a poll on into/neutral/eww for your fake profiles. Hana sounds obviously awesome and I have a hard time imagining anyone (anyone here?) ranking someone above her, but also would love to be wrong about that.
Is it just me or do Hana and Jane seem the least bad options? Hana sounds a bit mad, but maybe in a harmless interesting way. The other three sound just awful.
As someone who tends to be "that relentlessly match-making friend"...to me, dating docs aren't so much about preference filtering but a *distribution mechanism*--if I don't know that you are actively looking for a partner, I am not likely to look out for someone for you!
As unromantic as it is, the probability of finding a compatible match has a large marketing component to it...and classic marketing issues like distribution and awareness applies.
I have a lot of nostalgia for old OkCupid, and one thing that really ought to exist but doesn't is a nonprofit dating app with the sole goal of being _really, incredibly good_ at matching people. It seems like a relatively inexpensive way to create a quality of life increase in US cities, because dating is a really big but tractable allocation problem right now. It would be really "easy" to ingest a list of hard preferences and return matches in the area guaranteed to meet those. It doesn't solve all the problems of modern relationships, but it would be exponentially better than the current state of the art!
Not sure if this is a good epistemic rule, but, in a situation where Scott and Gwern disagree, I'm leaning towards Gwern with low confidence. Given that this discusses dating, if Jacob Falkovich weighs in, I would go with his tiebreaker.
I have questions about dating docs
Where do you post them? I saw Chris Olah's dating doc because it was posted on Marginal Revolution, but this will not work for people who have zero fame. Is the idea to become at least a little bit famous?
Do people who use dating docs also use swiping apps? How does their experience compare? Since Scott seemed to meet people on OKCupid, it seems like that at least used to be about as good as a dating doc, although now OKCupid is hot garbage ime
What do you write on dating docs that you couldn't easily convey in a swiping app? Many of these apps allow you to post your political affiliation, race, desire for children, and include a bio that shows any other traits you wish to see. Many dating docs that I've seen are quite similar to a profile on one of these apps
Do dating docs mostly help you date friends? This seems like the main benefit, in that if you create a dating doc, then your friends will probably be the first to know about it and it's a way to signal that you'd potentially be open to dating them.
Are there any people who exclusively do dating docs and do not use apps? The type of person who would make a dating doc is at least potentially interesting given the effort put in, but I'm wondering if I would see that type of person eventually on an app. This is probably unanswerable in general, but only via anecdotes
The problem with the above argument is that is slides from a few (and it's important that they are just a few) points that almost everyone everyone agrees upon ("I want a human <gender> who speaks my language and shares my culture") to the points that are actually in dispute, namely that people know what they want for almost everything else.
gender, language, culture. No-one is disputing these. But what people also think is important includes age, children (wanting to have in the future, not wanting partner to have any, etc), and a variety of "vanity culture" signifiers (being vegan, must like the outdoors, must be a fan of Phish, etc etc).
Of course there is imprecision here, such is life. What I consider a "vanity culture" signifier might, in your case, you are the one special snowflake unlike all others in the world, be an actual deep "cultural setting", so deep that it cannot be overridden. But the way to bet is that it's not.
Ultimately those of us who are older, or happily paired off, don't much care. You want to waste 20 years of your life and suffer a grossly diminished dating pool when you come to your sense, go ahead.
But there might be some wisdom in assuming that there is something to our accumulated years of experience rather than dismissing it out of hand.
The Western fad for assuming that the olds are idiots who know nothing of "true" love or "real" sex or how to give meaning to life has a long dreary history beginning in at least the mid-19th C. Yet not one of those rebel generations, from silents to beats to hippies to millenials, seems to achieved anything significantly different or worthwhile compared to their predecessors. Perhaps the accumulated experience of the ages is, in this case, mostly correct?
The below article cites some research favourable to Scott's case. A quote -
"Research in laboratory settings consistently shows that what people say they want in a partner has virtually no bearing on who they actually choose to date [citations] And yet, once people are in established relationships, they are happier when their partners match their ideals [citations]."
https://www.psychologytoday.com/gb/blog/dating-decisions/201412/the-real-reason-we-date-people-we-shouldnt
I was under the impression that love strikes like lightning from the blue, unbidden and unpredictable, rendering the sufferer powerless - and unlike Lichtenberg figures, the scars last forever.
It's always fascinating to learn about the different models others have of seemingly commonplace phenomena.
During the sexual-pairing phase of my life (I'm beyond that phase now), I was never able to figure out which women attracted me. Some just did — like a thunderbolt or cupid's arrow hitting me — but the vast majority didn't. I was never able to understand which particular physical or personality traits attracted me. Those hidden variables defeated all my attempts at self-reflective deduction. Worse yet, even when I made the effort to get to know the women who I thought should attract me (liberal-smart-educated-creative-regular facial features), I discovered that getting to know them didn't increase my attraction. So, given that one in a hundred women with some hidden variable which I couldn't pin down attracted me, I never gave dating apps a try.
I think what criteria you filter on would depend on what your goals are. A lot of people, at any one point in time, are looking *for* a one night stand and are trying to avoid commitment. But the real problem are interaction effects--I know many people use sex as a gateway to more committed relationships (that is, sex is considered the first or early step on that path). Therefore, sex appeal would be a necessary but not sufficient criterion for a relationship, followed by whatever personality characteristics you're into.
How much research is there on the olfactory side of things? The people I find attractive in photos does not fully overlap with real life and for my wife and myself, smell has a huge influence on attractiveness, and possibly for other relationship variables.
I'll chime in here with a rough description of the dating process in my sub-section of the Orthodox Jewish population.
- Every single has a resume/profile. These can be extremely spartan with just basic bio info like name/parents name, DOB/place of birth and residence, school history, current employment details, or can be more detailed with information about what the person is like and what they are looking for. A picture of the person may or may not be included. But every single profile will include references.
- A "shadchan" (matchmaker) will have the idea that Jacob might be a good match for Rachel. She (the vast majority of shadchanim are women), will suggest the match.
- Jacob and Rachel's parents or other close relative or friend will call the references on the profiles, and will additionally try to find people who know them and are not listed as references (the expectation being that people on the resume are vetted friends who are pre-disposed to say nice things about the person).
- They will be asking about most of the following: What is the single's personality? What is their level of religious observance? What kind of home did they grow up in? Are there any medical or mental health issues? Where do they want to live? What kind of lifestyle are they looking for? They will not ask whether or not they want to have children; that is considered to be a given for any single choosing to participate in this dating system.
- If both camps like what they hear, the shadchan will coordinate a 1st meeting.
- The 1st date is formal. The man will take the woman to a nice bar or restaurant, and they will spend around 3-4 hours talking. There will be no physical contact, not even a handshake or hug. This is almost always purely a vibe check. Not much of substance will be said or shared, and the expectation is that there will be a 2nd date unless one party really doesn't like the other person. The shadchan will coordinate the next date as well.
- The 2nd - 5th date will be spent verifying compatibility on the smaller things you can't ask about in reference calls. The minutiae of religious passion and observance, interests, lifestyle preferences, how many children you want to have, are you actually doing well in your career, etc. At any point past the 2nd date it is considered appropriate to end things (via the shadchan) if you just don't enjoy spending time with that person. Also at any point past the 2nd date the level of formality of the date is entirely the choice of the couple themselves.
- The 6th-10th dates are for DMC's and chemistry building. At some point in this stretch the couple will stop using the shadchan as a go-between and communicate directly. If either party wants to end it, they will have to end it directly too.
- Beyond the 10th date the couple is assumed to eventually be getting engaged. This usually happens in the 12 to 18 date range. Couples taking longer than this is usually due to commitment issues on the part of somebody, family difficulties, or some characteristic on the part of one single that the other one is trying to get over.
- After engagement, the couple will get married within 3-5 months.
Results: (non-scientific numbers)
- 80% marriage rate by age 30, 95% by age 40.
- 5-10% divorce rate.
- 70% happy/functional marriage rate.
Context:
- We are raised from birth to see getting married and having children (as many as possible) as a pre-requisite to living a religiously correct and happy life. Our social structures encourage this at every level, and there is very little comfortable space for older unmarried people.
- We are raised and educated gender-segregated. We do not socialize with the opposite gender who are not our close relatives.
- We are raised from birth with firm gender roles. It's the man responsibility to provide, and the woman's to raise children.
- We are taught what to look for and value in a marriage partner: religious compatibility, someone who is kind to us and others, someone who will be a good parent someone who can fulfill their assigned gender role, someone who we enjoy spending time with. Men are told to look for a woman who can make them feel masculine, woman are encouraged to look for a man who makes them feel feminine.
- We are told to avoid getting hung up on physical attraction at the early stages. Attraction will grow naturally as you spend more time together and your chemistry increases. Declining to meet someone based on just their physical appearance is heavily frowned upon.
- We are also told to avoid getting hung up on any number of minor things: How they dress, if they talk too fast, if they work an uncool job, if you don't like someone in their family, etc. We are taught that all of those things will fade or matter less as you get to know each other and as your chemistry grows.
- There is zero physical contact throughout the dating process until marriage.
- The man is expected to pay for everything throughout the dating process (if he is young or under-resourced his parents will usually pitch in).
- The parents of the couple will pay for the entire wedding (with community support for those needing it) and will also help setup their first home with furniture/appliances/dishes and the like.
Why did I write this megillah? Because Scott's argument is entirely correct, and should actually be taken much farther than he did. People need to be preference-maxing. It's easy to get married and stay happily married when you know what you want and have the tools to find potential partners who want the same thing.
Are we convinced that the OKCupid algorithm was actually more accurate than some other simple sorting algorithm like matching people by IQ or something?
There is a considerable amount of research that indicates that people regard a quick spontaneous response to be more honest and authentic than one the other person took their time and thought carefully about the options. For most people I have met, intellectual forethought is considered suspicious. Initial impressions are trusted more than calculated ones. There seems to be an idea in popular culture that one's intuition is a better gauge of oneself and one's own needs and desires than objective consideration. Given that, it isn't surprising that a great many people believe that you can't objectively articulate what you want or need in a partner.
This could be a case of the self-fulfilling prophesy. That is, two intuitive people, both of whom believe in first impressions and spontaneity as a sign of authenticity, may actually be more compatible with each other than they would be with someone more rational.
But rational people might be well advised to filter on the exact opposite set of criteria.
From just after the profiles:
> And all of this is separate from the types of preferences mentioned above - ie it’s not just the easy things like race, religion, income, number of children desired, politics, sexual compatibility, etc. Everything here is after the 1/500 even-getting-started number listed above! So learning about people from profiles must allow an even stronger filter than that!
I really don't think all of this *is* separate from the types of preferences mentioned above. What are the odds that Cindy is poly or willing to tolerate a mostly asexual partner?
OKCupid was a huge nerd snipe of a site, as it fed the introvert's tendency to think the answer was just around the corner from answering another hundred compatibility questions, rather than in just messaging and meeting up with people. In reality, most people's dealbreakers are all correlated with each other, and you can easily get an idea if someone is in your "market" from five photos and two paragraphs. Scott has much more idiosyncratic criteria than the vast majority of people, but even still there's only a couple questions that will essentially narrow down the pool with a high degree of confidence.
A worrisome trend is that corporations have adapted the concept of dating docs to employment. Using so called "job ads", they publish stringent criteria they have for employment and have the job applicants bear the load of checking for compatibility. This is despite Science(TM) showing that these formal criteria (like having an MD, or truck driver licence, or a PhD) have no correlation with the outcome of three minute long HR interviews.
Seriously, I think that dating docs are mostly useful in situations where there is an asymmetry between the supply of both prospective partners. For example, the rationality crowd seems to be overwhelmingly male (84% cismale on the 2022 ACX survey). Or I would expect that high status individuals within that group (e.g. Scott, EY) have probably way more people interested in dating them than they can possibly date. In such cases, dating docs serve multiple purposes:
* They advertise the fact that the writer is looking for dates (which can override the common courtesy of not cold-messaging people regarding romantic intentions especially in settings where universalization of such a behavior would see them flooded with such)
* They pre-filter candidates so that flooding does not occur
So dating docs sound like an obvious positive sum action for all the parties involved (except for more brazen people who would have cold-messaged their target audience before and now (1) have to compete with more considerate people and (2) may be summarily told "RTFM, you do not qualify".
Even for the ones filtered out by the criteria stated in the dating docs, the process is actually helpful, because it gives them feedback on a process which is normally completely opaque. If nine out of ten prospective partners state that they prefer a provider enabling a lavish lifestyle or physical beauty or BMIs below 20 or sizes above 1.9m, this is actually helpful information. Perhaps not directly actionable (unless one is willing to embark on a career in the drug trade or have ones leg bones enlarged or whatever), but still helpful (perhaps the correct choice is to search elsewhere for partners or romance a LLM or whatever).
Of course, one problem is that openness about certain criteria is generally frowned upon. Writing "your net income should be at least 100,000$ a year" is not the optimal way to filter for high earners, because many people (including people who qualify) will consider this criterion indicative of a shallow character. I think much of the traditional dating strategies revolve about maintaining plausible deniability about such things (which is why paying for dinner in a fancy restaurant will lead to better outcomes than just sending a copy of ones the bank balance).
One solution here might might be to provide aggregated information about the previous relationships. "my previous partners made between 100k and 130k a year" sounds at least less shallow than "I won't date anyone who makes less than 90k". One might still appear less shallow if one lied about that, though.
In general, I have two extreme models of the dating market, with the truth probably in between.
The one-dimensional model would be that most people of one gender, orientation and phase of life desire similar qualities in their partners. Few people are attracted by poverty or ugliness, after all. So for heterosexuals, you can order both by their hotness score, and the stable equilibrium will be the Nth hottest woman dating the Nth hottest man.
The high-dimensional model would assert that romantic potential is determined by lots of different preferences. People want partners with shared interests, similar (or complementary) kinks, members of their own subcultures and so on.
Obviously the real dating markets are mixtures of both, but I am totally puzzled as to regard what mixture. Any suggestions?
I built a dating app that does boolean matching based on users' selected dealmakers & dealbreakers to filter matches (the app is now mothballed). Suffice to say the obvious that the format of the app is insufficient unless you have a billionaire friend that is willing to give you a lot of money for marketing. If no one is on the app, then it is hard to get traction, no matter the algorithm.
So if someone is thinking about building such an app, either be really young and build it over the long run, or have a clear business & marketing strategy that leads to user growth despite initial users having few matches.
Such a document could have prevented the Jonah Hill situation - the text messages his ex shared were all about incompatible expectations for partner behavior.
I think the mistake here is confusing two questions:
* Are people most attracted to people who match their stated preferences?
* Are people most likely to date people who match their stated preferences?
Sadly, people don't get to date the people they are most attracted to. Imagine if the undergrads in any of the speed-dating studies were sent on a desert island and left to incubate until they're somehow all paired up. When asked about their preference, maybe every girl said "I want a tall boyfriend" and every boy said "I want a girlfriend with a 0.2 waist-to-hip ratio". But, at the end, the most attractive girls will pair with the most attractive boys, and you will have no correlation between stated preferences and who people actually date. However, you will find a high correlation between the two partners' objective attractiveness.
In that regard, the utility of date-me docs is not to find people who would match your preferences, but to find people in your league.
Hi Scott,
I look forward to your upcoming column “Yet Another Bay Area House Party Redux,” starring Hana, Jane, Larissa, and Sky (Cindy makes a cameo appearance, decides the party is boring/weird, and leaves after 5 minutes). Please do it! The ACX commentariat will be eternally grateful. 😊
Dating docs are an attempt to solve a problem (the lack of partner). Unfortunately, trying to solve a problem = signaling that you have one; and that is low-status. The people making fun of dating docs are signaling that they don't have this problem; and whether that is true or not, the signal is high-status.
Note that reading or writing dating docs implies a preference for reading or writing. Some people hate reading and writing, so this solution wouldn't work for them.
I wonder how honest can people actually be in their dating docs. People are hypocritical -- maybe you want to date a person of the same race, and your potential partner also wants to date a person of the same race... but your potential partner may still perceive this information in your dating doc as evidence of racism and reject you. (Maybe they actually don't mind you saying this, but are still afraid that their friends might judge them for dating a racist. Problem is, your dating doc is a public document.) Talking about your partner's income may be a similar problem: you may want to date someone in the same social class, but don't want to be seen as a person for whom money is so important. Among woke people, expressing a preference for your partner being cis could be a similar problem. So a lot of important information may be unmentionable.
In a comment in this thread, Jack mentions preference for "kindness, honesty, and emotional stability". Can these things be talked about explicitly? Do honest people write "I am honest" and dishonest people write "I am dishonest" in their dating docs? Does anyone admit being hateful, or a psychopath? It could even be the other way round, a normal person admitting "sometimes I am not as nice as I would like to be" and a psychopath saying "actually, I am perfectly nice all the time". Or maybe admitting that "I am a little bit X" actually means that person is X a lot. In other words, you probably can't trust most information like this.
Dating doc is a potential weakness, e.g. if you are an attractive girl looking for a long-term relation, but there are many guys wanting to have a one-night stand with you. You just told them exactly what they should pretend to be!
Also, knowing yourself (sufficiently that you can describe yourself in a document) is quite difficult. I might be able to do it now, but I doubt that I would do a good job in my 20s. But maybe it's a part of the process of learning: figuring out what went wrong, and updating your doc accordingly. For example, in a parallel universe where LessWrong and the rationalist community never existed, I would have difficulty describing that I am looking for a *rationalist*. I would probably write something like "looking for an educated and open-minded person", but that's not the same. It is not even a narrow filter because many people are educated and almost everyone considers themselves open-minded.
This probably sounds too negative, but I actually like the idea. I just notice the imperfections. Freestyle description sounds nicer than checkboxes: I get more of the feeling that I know the person. Also, when other people design the checkboxes, sometimes you just don't fit any category; text allows you to describe it using your own words. We should probably use something similar also for finding friends.
Would you necessarily want to restrict the potential candidate pool *before* meeting them in person? I think preferences are a complex interaction--I may weight certain personality traits higher than physical attractiveness, but the weights aren't "1.0" and "0". So it's entirely possible that a sufficiently attractive candidate could make up for a lack of certain personality traits, but only in comparison to whoever else showed up. You can't really predict these things in advance, so perhaps best to use filters conservatively.
Then again, spending an evening with a member of your preferred gender isn't always a complete waste of time, even if neither sex nor romance is likely to result. What else are you going to do for two hours? So maybe it makes sense to be liberal with who you agree to meet.
(Aware that many women are forced to proactively protect themselves from the potential for sexual harassment, so filtering for "creepy" as early in the process as possible is non-optional).
This post is f-ing great! The day before, I was listening to "Tea for the Tillerman". And I need a hard headed women... And I updated my profile on e-harmony, and added pictures. For me the problem with a dating doc is exposure. I live far away from any big city, so number of people is much bigger on e-harmony. (though the interface totally sucks, and I can't put my likes into whatever pigeon holes they've put up.)
This reminds me of "Chrono Cross":
> Entry number 1... Hello. I'm Marianella. I'm a Virgo and I'm 15. My hobby is **PoisonGas**! Nice to meet you.
> Entry number 2... Petra the Pisces here! I'm 16 and I love tall, dark boys! I love to sing in the **BlackRain**! Nice to meet you.
> Entry number 3... I'm Carolina, and I'm 10. When I grow up I want to be Miss Universe. I do a mean **Tornado**! Nice to meet you.
> Entry number 4... I'm Richard. I'm 29. And I'm here to protest beauty pageant sexism! My proudest accomplishment is the **BlackHole**! Nice to meet you.
Dating docs are cringey but also kind of inevitable, right?
1. Scarcity exists. There aren't hundreds of men with Richard Feynman brains and Henry Cavill shoulders lobbing pebbles at my bedroom window.
2. The dating market wants to clear. Everyone wants love or at least a good time.
3. Information asymmetries exist. Why not clear the fog a bit? As credit ratings and standardized test scores do.
I'm glad I haven't had to make one! My husband and I got sloshed around in a tiny office dating pool of numerate 22-year-olds, which probably happens less in a WFH world.
Separately, how do polyamorous people with kids have the time and energy to keep going on dates?
Someone (putanumonit Jacob?): "Date-me docs signal desperation."
Me: "But doctor, I..."
I dated someone who was not really that compatible. It was a mistake, don't do it. Prescreening compatibility is very important.
I think a large part of why people are against dating docs is two-fold:
- it's not mainstream (yet?) and as such seems, like all online dating, desperate/weird/... . Basically it goes against the fairytale "you will bump into your perfect partner in the library/on the way to work/... and one of you will drop all their papers/books and the other will help pick them up and it will be cute and fate and the universe have this perfectly organized for you and if you try to make it more likely to meet a suitable partner you hate romance and mess up the universe's plans"
- This checklist mentality adopted at large scale would lead to even more segregation than we have now. If I only meet people who are in my social class, I'm in even more of a bubble/echo chamber etc. (I do personally agree that this could have pretty bad societal implications if everyone could filter for a lot of describable preferences/hard facts before seeing a list of people that want to date.)
> Many people want a partner of the same religion and level of religiosity.
This can't be right. Traditional literature/religion is full of indications that men need to be forced to marry within their religion, because otherwise they'll just convert to whatever their wife wants.
And in general - and this is easily observed today - women are much, much more religious than men. Insisting that you want a partner of the same level of religiosity as yourself is much like insisting that you want a partner of the same height as yourself: at a system level, almost nobody can possibly accomplish this, and it's not really worthwhile to try.
I think that most negative reaction on dating docs is sorta emotional and arguments is rationalization. Why it often provokes negatives? First thing - it strikes all "marriages are made in heaven" and "instant strike of romantic love" narratives, and this is still a big thing in our culture - dating docs are much more about "rational partner choosing" than "arrows of cupid". Second - we still mentally more villagers than citizens, and in village being too picky is a bad plan - but much more OK in big city
The way this topic was taught in my sociology class referenced Intimate Relationships by Rowland Miller, Chapter 3 of which I think covers the current research well.
For casual relationships/one-night stands: pretty much everyone's mainly looking for someone physically attractive. If you can get along with them at least a little bit, then you might stick around for a while and call it FWB or a situationship, but looks are still the main criterion for both genders.
For longer-term relationships, here's how it breaks down:
* Everyone's got similar dealbreakers (which are generally pretty obvious: kids, career, smoking/drug use, poly/mono, whether or not someone's very politically-minded or religiously observant and if so which religion/party are they a member of, etc.).
* Everyone's looking for a certain baseline level of assortative matching; the biggest difference between what straight men vs. women are looking for is that men tend to more heavily favor good health/looks and women tend to more heavily focus on wealth/earnings potential, but everything else is pretty similar and there's absolutely a lot of trading-off between how much of each criterion partners satisfy.
* How the assortative matching shakes out for long-term relationships is that they're all *benchmark* requirements: people look for a certain bare minimum (depending on how desirable/attractive they themselves are), but beyond that they're not super-important.
* Once the benchmark's met on both sides, what everyone's really looking for is how well you two get along: agreeableness, compassion, loyalty, and so on.
The latest edition of that book came out in 2022, and Chapter 3's got an *extensive* bibliography. I'm probably not pulling out the right citations, but here's a sampling:
* For similarities in requirements: Thomas, A. G., Jonason, P. K., Blackburn, J. D., & Li, N. P. (2020). Mate preference priorities in the East and West: A cross-cultural test of the mate preference priority model. Journal of Personality. doi:10.1111/jopy12514
* For a summarized list of what people want: Lam, B. C. P., Cross, S. E., Wu, T., Yeh, K., Wang, Y., & Su, J. C. (2016). What do you want in a marriage? Examining marriage ideals in Taiwan and the United States. Personality and Social Psychology Bulletin, 42, 703–732.
* More on what people look for in a casual relationship (not sure which one they were citing!):
* Eastwick, P. W., Luchies, L. B., Finkel, E. J., & Hunt, L. L. (2014). The predictive validity of ideal partner preferences: A review and meta-analysis. Psychological Bulletin, 140, 623–665.
* Eastwick, P. W., & Hunt, L. L. (2014). Relational mate value: Consensus and uniqueness in romantic evaluations. Journal of Personality and Social Psychology, 106, 728–751.
* For people adjusting their standards to meet who they're already dating: Kučerová, R., Csajbók, Z., & Havlíček, J. (2018). Coupled individuals adjust their ideal mate preferences according their actual partner. Personality and Individual Differences, 135, 248–257.
Suffice it to say that I think the research isn't in a completely terrible place right now (knock-on-wood), and that at least some of the criteria are well-understood.
The real criticism of date-me docs, then, wouldn't be that the preferences themselves are undescribable: it'd be that what people put into the date-me docs doesn't effectively depict what matters.
And I think that's a bad criticism, too; I think writing about yourself is a great way to show off your personality, especially if you write a lot in general. (Even if you're just showing off the fact that your first instinct to show yourself off is to do it in writing!)
So really, I think that the main criticism might just boil down to an example of The Worst Argument In The World.
A not-so-hot take: I hope AI-assisted dating will take off someday. I won't need it, I hope, since I'm already married, but other men and women like me are suffering. Tinder-like apps are shit - "matching" people on attractiveness seems like anathema to me. And long-form dating sites are mostly dead and not coming back, unless someone finds a way to monetize them (well, Substack found a way to monetize long reads, which were going the way of dodo with advent of social networks, so there is hope there). But if I had to date again, it would be so, so nice to let an AI assistant comb the internet for a likely woman who matches my preferences...
From reading the article and the comments here, it seems like we have dating apps that are actually good: OKCupid and eHarmony. Why, then, do we hear so much about how dating is broken these days and loneliness is at all-time highs?
I would guess it's a combination of a few things
1. These apps are not as good as people say they are
2. People don't actually want to settle down that badly
3. People's standards are too high
Anyone have insight into why dating isn't basically solved? We can put aside the question of what's wrong with the existing social science and consider the efficacy of the existing social engineering.
I'm 99% sure I made the first date me doc
https://twitter.com/dschorno/status/1687141890016329728?t=1-UfpVSfzMNSYcVZrgHS-g&s=19
I came for the headline and stayed for the glimpses into Scott’s love life.
I think the argument here is correct, but wonder if there is something more to be gotten from the skeptics on stated preferences. I think lots of people (applies to me at various points in my life) overstate the importance of, for example, matters of taste (liking the same music, pop culture ephemera, sports teams, whatever) and how important alignment on that is for a happy, long-term relationship.
Basically I think there’s a bucket of important things you should be aligned on - monogamy, religion, politics, whatever, and a bucket of things that feel like they are important but probably aren’t - Taylor Swift fandom. Of course person to person these categories can change and there is no one size fits all solution to disentangling what is important and what isn’t. That’s probably something best left to your friends to actually help you sort through. But a study that predicts that e.g. shared musical taste doesn’t mean much as far as lowering divorce risk is a decent enough hint that maybe for most people that’s a criterion that should fall in the “not that important” bucket.
People don't think dating docs are weird and repellent because there are no legible criteria to screen potential partners. But these criteria should be a blunt, first-pass tool. If I want an atheist guy in his 30s in DC, and you're a 20-year-old Christian woman in Ohio, we're not compatible. Great. Tell people your basic criteria. But once you have someone who meets those first few basic, legible, criteria...go on a freaking date! You'll be able to tell within 30 minutes whether you're attracted to the person, like the person, and feel comfortable with the person. And you cannot determine this from any amount of reading and writing google docs.
Dating docs are weird because they're incredibly long and detailed, and because the effort you spend writing ten pages about yourself and reading other people's manifestos is effort that could more usefully be spent going on dates, and actually figuring out whether you like the person.
If you like them, feel comfortable with them, want to have sex with them...you'll find that all the details on your 15th page about your ethical philosophy and tastes in video games absolutely do not matter at all.
I think you're overestimating how significant those numbers say political compatibility is. The study said that 4% of marriages are between a Republican and a Democrat, which *sounds* low, but given that something like 30% of people are Republicans and 30% Democrats and 40% Independents, you would only expect 9% from pure mixing. There are 17% between Independents and non-Independents, but from random mixing you would only expect 24%.
Looking at these numbers actually made me realize that this level of political compatibility appears to be far *less* important to most people than I would have expected. It looks like 16% of Democrats are married across party lines, even though 70% of their options are across party lines (about 1/4) of the expected number), and 84% of Democrats are married to other Democrats, even though only 30% of their options are Democrats (about 3x the expected number) so there's something like a factor of 12 between their probability of marrying a particular non-Democrat compared to their probability of marrying a particular Democrat. But a factor of 12 is much smaller than I would have expected in this era of high polarization - and low enough that you wouldn't actually want a filter that eliminates 100% of the people from the wrong side of that line.
Mark my words, soon dating apps will happen in VR. Assuming we get to a point where it can read and show your facial expressions. And possibly upper body language as well.
In fact VR would be perfect for meeting strangers. You can sort of be yourself in an avatar with a different voice (or the same voice), it would be like a real life meet up, except you can end it anytime and move on to the next one.
I noticed this post mentions race a few times, seemingly with the implication that most Americans would only want to date someone of the same race as themselves. Now I'm curious, is that really true, and to what extent? My intuition says there are plenty of people who don't have an explicit preference regarding race, even if they might filter potential partners on other factors like political views. I'm wondering if my social circles and I are outliers on this.
For context: I'm a straight white (ethnically part-Jewish) guy in my late 20s, politically liberal, who has lived most of my life in the NYC area and the Boston area. I don't think it's wrong if someone has a preference towards dating someone of their same race, but I've never had that kind of preference. When I was more actively trying to date and using the apps, I went on dates with both white and non-white women without it ever coming up as an issue. And many of my friends seem to feel similarly.
Pretty sure patient zero was Richard Stallman and that should be enough to convince most people against the practice.
My grandpa was a medical doctor and he had zero difficulties finding a date or a spouse. Aren't you a psychiatrist, Scott? So maybe your ease in finding a catch isn't that informative. I mean, how many women who are moderately indisposed to poly life would flip flop to snag a doc? Or men.
Why would someone who is asexual be interested in polyamory?
This is an absurd question, Scott. I don't know if it's the asexuality or what, but the answer to your puzzle, or whatever you want to call it, of the five dating ads is:
1) Which of them is the best-looking?
2) Which of them is the best at giving head?
I don't care about *anything* any of them wrote. None of that has absolutely any relevance for a relationship. Have your weird hobbies, girl!
Speculatively, I think dating docs (as opposed to broad categorical pre-screening) will, if they grow in popularity, have the effect of essentially making women increasingly convinced that few men if any are good enough for them.
I would push back on the claim that a public "date me" profile reduces pipeline and exposure. I suspect that a well formed and publicized one actually both gets you a much better top-of-the-pipeline, and acts as a more efficient filter. You and your extended social network are probably much better about targeting single people with your profile than any dating app would be. So way more eyeballs on first pass, much better chance of multiple exposures to the same eyeballs (it's still a weird enough thing that folk might read and talk about your profile multiple times; an early adopter advantage), and network effects (they'd be a perfect match for Jane! Let's introduce them)
I spent years on OkCupid in two countries and never met Hana. Surprisingly low amount of resulting relationships. But I did find that free text profiles gave some sense of the person.
In my experience, the chance that Hana also loves Taylor Swift is surprisingly high, relative to what stereotypes would suggest. Swift is actually hip with rock critics.
This sounds like a great idea. An obvious one, even. But then again, I always found "show, don't tell" an incredibly stupid advice. Might be related.
Back when I used to use dating apps, similar music preference + photos were uncannily good at predicting compatibility for me. I think music taste can reveal most other preferences fairly quickly like for instance where you prefer to live, religion, interests, personality.
What was your process for finding compatible people, when you were looking, Scott?