Categories
okcupid PUA

Dating by the Numbers: Why “hacking” OkCupid is a waste of everyone’s time

This data point is an outlier.
This data point is an outlier.

Is there no problem out there that can’t be solved by SCIENCE? Apparently not. Indeed, it has recently come to my attention that one heroic nerdy dude actually used MATH to get a girlfriend. No really, an actual girlfriend. As in, a living human female that he’s seen naked. (We can only presume.)

Wired magazine found the story so astonishing that they devoted an entire 3000 word feature to it.

The piece tells the tale of Chris McKinlay, then a grad student in mathematics at UCLA, who went searching for love on OKCupid, a dating site that uses daters’ answers to various questions, ranging from silly to profound, in order to calculate a “match score” that supposedly measures your compatibility with a potential date. But McKinlay wasn’t getting as many dates as he wanted.

So he decided to “reverse-engineer” OkCupid. As McKinlay — ever the romantic — explains on his own blog, he used his mathematical skillz to analyze the “high-dimensional user metadata in [the] putatively bipartite social graph structure [of] OkCupid,” and adjust his own profile accordingly.

Basically, he crunched a lot of numbers to figure out how the kinds of women he was most interested in — in particular one data “cluster dominated by women in their mid-twenties who looked like indie types, musicians and artists” — tended to answer questions. And then he fiddled with his own answers — and his choice of which questions to answer — so he would score higher match percentages with them. Ta da! Suddenly he had more matches.

He claims not to have answered any questions dishonestly, but as Wired notes “he  let his computer figure out how much importance to assign each question, using a machine-learning algorithm called adaptive boosting to derive the best weightings.”

It doesn’t take a math degree to figure out that fudging your answers so they’re more like those of the women you’re targeting will make it look like you’re more like them. You can pull this same trick in real life by pretending to agree with everything a person says.

But you don’t have to be a psychologist to see that doing this kind of defeats the purpose of OKCupid’s match algorithms in the first place. You’re creating the illusion of chemistry where there may be none. Essentially, you’re cheating, but in a really self-defeating way.

And by focusing so intently on statistically crunchable data, he also ignored a lot of the more intangible “data” that the profiles provide if you actually sit down to read them. The numbers don’t reveal anything about a person’s verbal charm, or their sense of humor. They don’t tell you about the interesting little details of the person’s life.

As Katie Heaney notes in a Buzzfeed piece on McKinlay’s strange quest:

[M]uch of the language used in the story reflects a weird mathematician-pickup artist-hybrid view of women as mere data points … often quite literally: McKinlay refers to identity markers like ethnicity and religious beliefs as “all that crap”; his “survey data” is organized into a “single, solid gob”; unforeseen traits like tattoos and dog ownership are called “latent variables.” By viewing himself as a developer, and the women on OkCupid as subjects to be organized and “mined,” McKinlay places himself in a perceived greater place of power. Women are accessories he’s entitled to. Pickup artists do this too, calling women “targets” and places where they live and hang out “marketplaces.” It’s a spectrum, to be sure, but McKinlay’s worldview and the PUA worldview are two stops along it. Both seem to regard women as abstract prizes for clever wordplay or, as it may be, skilled coding. Neither seems particularly aware of, or concerned with, what happens after simply getting a woman to say yes.

And that’s where McKinlay’s system seems to have fallen down entirely. Though Wired is eager to present his “hacking” as a great success, it took McKinlay more than 90 dates  — 87 of them first dates with no followup — before he found his current girlfriend.

In other words, his wondrous system produced a metric shit-ton of “false matches” and wasted a lot of people’s time, including his own.

And in the end it wasn’t his data crunching that brought his girlfriend to his door; as Wired notes, she found him on OKCupid after doing a “search for 6-foot guys with blue eyes near UCLA.” Happily for him, McKinlay already matched her preferences in these areas. In addition to appreciating his height and eye color and location in physical space, she apparently was also charmed by his cynical approach to OkCupid dating, so maybe they are a match made in heaven, if not in his data crunching techniques.

While McKinlay was going on first date after first date after first date after first date after first date after first date after first date after first date after first date after first date after first date after first date after first date after first date after first date after first date after first date after first date after first date after first date after first date after first date after first date after first date after first date after first date after first date after first date after first date after first date after first date after first date after first date after first date after first date after first date after first date after first date after first date after first date after first date after first date after first date after first date after first date after first date after first date after first date after first date after first date after first date after first date after first date after first date after first date after first date after first date after first date after first date after first date after first date after first date after first date after first date after first date after first date after first date after first date after first date after first date after first date after first date after first date after first date after first date after first date after first date after first date after first date after first date first date after first date after first date after first date after first date after first date after first date, people I know have found wonderfully compatible matches — and long-term relationships — through OkCupid without having to date dozens of duds along the way.

How? Partly because OkCupid’s match algorithms led them to some interesting candidates. But mainly because they read profiles carefully and looked for compatibility in the words, not the numbers.

700 Comments
Inline Feedbacks
View all comments
kittehserf
10 years ago

I haven’t any grounds for comparison but that game/test/whatever was a hoot (not least the commentary). 😀

Kiwi girl
Kiwi girl
10 years ago

I think this guy deserves some slack.

No, he doesn’t. And all we’ve done is talk about how skeevy he is, no-one is marching to his house with pitchforks.

For me, the only way to find out if you have a connection with someone is to head out and actually talk…in person.

Comment not relevant to using a dating *website*.

He wasn’t getting enough actual dates out of his profile, so he mixed it up and got a lot of dates.

Use of OKCupid doesn’t guarantee any number of dates. OKCupid does not owe him dates, and therefore he can do what he likes with their data in order to get some. The women on OKCupid do not owe him dates just because he made a fucking profile on there.

He still had to go out and make a real impression, and he presumably accepted (or delivered) the verdict regarding the “one-date only” women with a modicum of grace, so in the end, what’s creepy about this?

*He* misrepresented himself so his profile would get higher up the women’s lists of matches. *He* therefore got dates with people he wasn’t actually that compatible with, on the basis of his misrepresentations. *He* then gave a “verdict” (what a nice word to use in this context) on a mis-matched date which *he* caused to happen by his misrepresentations. In this telling, *he* has all the agency/autonomy.

*He* also data mined OKCupid’s profiles against their terms and conditions. When he got blocked – and that is what happened to his first lot of bots – he then created “new improved” ones to *get around the blocks*. He used vast quantities of personal information about women in a way that it is not intended to be used.

What about what those women wanted? What about the fact that a number of them went on a date with someone, and then found out he wasn’t like he said he was in his profile? But I guess, because they’re only women their feelings and time doesn’t count.

How fucking hard is this to understand?

kittehserf
10 years ago

katz – seconding that. Plus, the idea of someone gaming the system to find some hypothetical Ideal Woman (or womanthing) says he’s not actually interested in women as people, but as trophies, or like he’s going shopping or something. For me, it’s one thing to say there are dealbreakers, reasons you wouldn’t date someone (they’re a smoker, or heavy drinker, or politics, for instance) but that’s not lying about yourself, which this guy was, and it’s not putting everyone into some sort of categories as if they’re data, as has been said many times already.

Low-grade PoS is all this guy is, in my view.

cassandrakitty
cassandrakitty
10 years ago

In order to think that what this guy did was no big deal there are some things you have to believe first.

1. Women don’t have a right to have preferences (so who cares if this guy actively tried to make sure they couldn’t select based on those preferences)
2. Women’s time is worthless (so who cares if the women who went on all those failed dates with him might not have done so if he’d presented himself more honestly)
3. Businesses have no right to try to serve their customers well and make policies based on that principle (so who cares if OK Cupid has policies against botting)
4. Businesses have no right to try to protect the personal data of their customers (so who cares if this dude or someone like him wants to harvest it, and uses technical means to get around the systems put in place to stop that from happening)
5. Woman aren’t people, they’re a public resource that companies like OKC are unfairly preventing the actual people (men) from accessing, so we should celebrate those who figure out a way to access the womenresources in the most efficient manner

There’s more, but really, is it necessary to spell all this out? It’s pretty obvious why this isn’t OK.

kittehserf
10 years ago

He still had to go out and make a real impression, and he presumably accepted (or delivered) the verdict regarding the “one-date only” women with a modicum of grace, so in the end, what’s creepy about this?

Lying to get a date (or get into someone’s pants?) = lying. As in, lying little shit who doesn’t give a damn about the other person, has no respect for them, and just sees them as a thing for him to choose or reject.

Why the fuck should he be cut slack? Even leaving aside the way he rorted the system, his treatment of the women involved is wrong.

kittehserf
10 years ago

Ninjaed every so much more eloquently by cassandra.

kittehserf
10 years ago

*ever

I blame the weather

cassandrakitty
cassandrakitty
10 years ago

Also, before another whiny misogynist comes in to complain that we don’t care about men – what about the male users whose profiles were downgraded in the compatibility rankings by this guy’s bots? Do we just not care about them because their willingness to accept the rules built into the system makes them too beta, or what?

Kiwi girl
Kiwi girl
10 years ago

The statistical thing, which makes me keep doing /headdesk, is that those 7 categories were constructed by the computer algorithm. There was *nothing* in the clustering algorithm that was instructions along the lines of “put all the women I’d prefer in this cluster, and BTW these are the things I prefer in women”. In statistical terms, we would call that weighting the data, so that things with larger weights would be given more priority in deciding the clusters.

So after he got 7 clusters, he still had to look at them to decide which clusters (2 initially) he preferred. And because the clustering algorithm wasn’t given his preferences, who knows what other women met his preferences but were in the other clusters (lucky break for them).

Professionally, I use clustering when I have a bunch of data and I don’t know if there are any sensible (i.e. theoretically explainable) groups in there. Surely he knew what types of women he preferred before he did this data scraping and dredging, so why did he use a clustering method? There are statistical methods to use when one knows the group membership, and these will always be more efficient than naive clustering.

His method wasn’t even intelligent or efficient from a maths/machine learning perspective. I don’t know why people are thinking he did something numerically clever, aside from all the ethics issues.

hellkell
hellkell
10 years ago

Dan, maybe you should have read the comments.

buttboy69
buttboy69
10 years ago

Reading this thread has been interesting. For one, it’s abundantly clear that the women here don’t understand how frustrating OKC can be for male users (and moreover, have no interest in understanding).

From my limited experience on that site, I can completely understand the urge to do something like this, if for no other reason to satisfy some perverse intellectual curiosity. It so often feels like a crapshoot for men, anyway.

Which isn’t to say I don’t understand the objections that have been raised here. Yes, he’s gaming the system. Yes, in a very indirect sense, he’s denying his dates the right to their preferences. But, you know, give me a break. In the real world, people of both genders engage in analogous behavior all the time. We lie about our pasts, our tastes, our skills. We put up a facade in an attempt to impress people we like, because of course we aren’t good enough.

It’s not a good thing to do, and it rarely leads to healthy relationship. And granted, there’s an extra layer of manipulativeness in this case, what with the botting and such. But I think calling him a “low-grade PoS” is, well, dumb.

kittehserf
10 years ago

You whine about “frustration” on OKC – ever been stalked? Harassed? Sent rape/death threats?

Oh, but of course you have with that pretend profile you really truly made yet somehow never show anyone.

Go fuck yourself with the world’s smallest violin, creeper.

buttboy69
buttboy69
10 years ago

^^ As I said- no interest in understanding.

hellkell
hellkell
10 years ago

Buttboy, shut up. You bring exactly nothing to the table.

hellkell
hellkell
10 years ago

We lie about our pasts, our tastes, our skills. We put up a facade in an attempt to impress people we like, because of course we aren’t good enough.

Stop universalizing what YOU do.

serrana
serrana
10 years ago

Go fuck yourself with the world’s smallest violin

You’re cracking me the fuck up over here.

buttboy69
buttboy69
10 years ago

But, you know, here’s a hint: it’s not about “whining” or me trying to portray men as “having it worse” or any of that. It’s just me trying to look at it from his perspective, and how he might not see what he’s done as particularly horrible or manipulative. I don’t think it’s that hard.

But forget it. I shouldn’t have posted.

hellkell
hellkell
10 years ago

For once, you are correct. You shouldn’t have posted. See if you can’t keep that up.

kittehserf
10 years ago

Fuck off, skeeze. You’re just a mealy-mouthed misogynist. We’ve seen your type plenty of times.

Why the hell should we be interested in seeing something from the perspective of a lying piece of shit who cheats women and cheats a company’s protections of its clients – breaches of privacy, what are they – for a quick date/fuck with someone who he probably isn’t even interested in, and whose right to make her own choices he’s subverted?

I’m not interested in the perspective of men who treat women as less than human. That’s the only other thing you’ve got right. Your repeated defence of them puts you right into the “thing to be wiped off the shoe” category as well.

Serrana – it gets better, too, because splinters . 😈

cassandrakitty
cassandrakitty
10 years ago

We lie about our pasts, our tastes, our skills. We put up a facade in an attempt to impress people we like, because of course we aren’t good enough.

Who is this “we” you speak of? Because it certainly doesn’t seem to include the subgroup “me”.

LBT
LBT
10 years ago

RE: Dan

the only way to find out if you have a connection with someone is to head out and actually talk…in person.

Then why on earth would you do online dating at all? Why would you do all this elaborate programming and spambotting against OKC policy? Why wouldn’t you just GO TO PEOPLE?

You’re talking about something completely different than what this dude did.

RE: buttboy69

For one, it’s abundantly clear that the women here don’t understand how frustrating OKC can be for male users

Dude, I’m male. I think this guy is an ass. Don’t pretend you’re spokesman for the male gender.

In the real world, people of both genders engage in analogous behavior all the time. We lie about our pasts, our tastes, our skills. We put up a facade in an attempt to impress people we like, because of course we aren’t good enough.

Do you create robots who go out and socialize for you? This guy was spambotting. Also, why would you want to date someone you have to put on a facade for?

^^ As I said- no interest in understanding.

Quoth the kettle.

It’s just me trying to look at it from his perspective, and how he might not see what he’s done as particularly horrible or manipulative. I don’t think it’s that hard.

I can look at it from his perspective. He sounds like a sad, lonely man so desperate for a relationship he was willing to ignore his dissertation and sleep on his table to kludge together this stupid program and spambots to get in with a girl. That’s sad, and it’s pathetic, and it doesn’t make excuses for his behavior, which is what you’re doing.

Again, who died and made YOU spokesman of the male gender?

Kiwi girl
Kiwi girl
10 years ago

@Argenti, if you’re still reading this, have a look at the Buzzfeed comments that skeevy guy thinks skewered the Buzzfeed journo (one of David’s links, but here so you don’t have to work out which one: http://www.buzzfeed.com/katieheaney/sorry-guy-math-cant-get-you-a-girlfriend)

The two main bits of fail for me:

“That personal likelihood doesn’t change whether she’s one of 10 or one of 50,000.”
No, but even assuming a 5% chance that any given woman says yes, the probably that at least one out of 50,000 women will say yes is essentially 100%. Rather, 1 – (.05)^50000.

But this remains true irrespective of whether a clustering algorithm was used or not. The point to the clustering method is to *increase efficiency*. When a mathematical argument boils down to “but if he talked to 50,000 women then he would have close to a 100% chance of a date”, *the number of women is the driver, not the fucking selection method*. There’s no evidence that his clustering method increased the average probability of a woman saying yes. And seriously, did they think he was going to date anything like 50,000 women? And “genius” mathematician thinks this is a skewering argument against the Buzzfeed piece /rolls eyes

Slight nitpick: the outcome of the flip of a quarter is totally independent of any other flip. That much is true. The example you use of what happens on the 10th flip is not correct, though. What happens the 9 previous throws does matter. Then it’s a simple matter of conditional probability to figure out how likely it will be to get either a heads or tails on the 10th flip.

No, this is wrong because it requires that a law of small numbers exists. There is no law of small numbers. If this argument were correct, lots of people would win big on games of chance like roulette in casinos (it came up black/odd 9 times in a row, red/even is due now). Runs are common, but it all reverts to the distribution defaults in the long run.

It’s ironic that the people trying to skewer the Buzzfeed author for her “poor understanding of statistics” came unarmed to a game of wits.

Howard Bannister
10 years ago

It’s not a good thing to do, and it rarely leads to healthy relationship. And granted, there’s an extra layer of manipulativeness in this case, what with the botting and such. But I think calling him a “low-grade PoS” is, well, dumb.

Really? Cuz I thought it was a quite restrained and normal reaction to, yanno, that ‘extra layer of manipulativeness’ you glossed right over.

LBT
LBT
10 years ago

Seriously, this is the first time I’ve seen folks DEFEND spambotting. Truly now, would any of you miss those damn spambot comments? I’d be delighted to see them all freakin’ vanish. Yes, even the entertaining word salad ones about viagra.

1 6 7 8 9 10 28