Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

OkCupid on Trial - 4 Month Experiment

jonmillward.com

Post reply on HN
Print view View original

51–60 of 101 posts

Re: OkCupid on Trial - 4 Month Experiment

#51

Post by erehweb »

Whenever there's an experiment, we have to weigh the costs against the benefits.

Benefits here seem slight - we relearn that women, and pretty women in particular, get more messages than men.

Costs are non-trivial - about 1000 people deceived. Admittedly a minor deception, but experimental subjects normally get some protection, and a minute of time wasted by each translates to 16 hours overall. I'm presuming he got consent to use all the photos in his experiment and on the site, but he's silent on that. At least we have a thousand small disappointments.

Experiments involving deceiving people can be useful (e.g. ones investigating whether employers / landlords discriminate), but this one doesn't seem to rise to that level.

Re: OkCupid on Trial - 4 Month Experiment

#52

Post by alttab »

goodside wrote: post #44
tghw wrote: post #40

Earlier quoted context omitted.

Is there any chance of OkTrends coming back any time soon, if ever?

There are no plans to permanently stop doing posts as far as I know. OkCupid's user base has been growing rapidly over the past year, to put it mildly, but we're still only ~30 people so we haven't had much time for the more peripheral aspects of the site. I know there are some things in the works, but I'm not directly involved so I can't really give a time estimate.

I glad we have someone from OKCupid on here. Just want to let you know personally that I tried everything when I moved to a new city. Match, PlentyOfFish, and a billion others I can't remember. I got good at writing profiles.

OKCupid did an auto-match for someone I didn't find during my own searches. I thought I looked at everything. We apparently thought each other were attractive enough, and thats what spawned the initial prod from OKCupid.

The similarities didn't stop there. This was 4 years ago and now I have an amazing wife, the perfect wife. She makes me a better person.

It is fairly easy for me to say I would have never met this woman without OKCupid (the site, and the science that drives it). Your service has literally changed my life permanently, for the better. And you did it without charging me a thing. So thank you, and hopefully you guys keep your spirit and innovation as part of a larger match.com portfolio!

Re: OkCupid on Trial - 4 Month Experiment

#53

Post by personlurking »

untog wrote: post #2

For anyone not aware, OkCupid has a fantastic blog where they do all sorts of statistical analysis on their members: http://blog.okcupid.com/ It hasn't been updated in over a year. Downer. Getting bought by match.com has clearly had an effect. I have the extremely dubious honour of having had my profile featured on there once as being of "average attractiveness". People still visit my profile to this day, despite me…

There's a very interesting behavior that happens on the streets of Brazilian cities which your comment on anecdotal evidence made me remember. Due to a seemingly high percentage of men here acting like they've never seen a woman before in their life (ie, doing a quadruple-take when one walks by) and doing lots of cat-calling (generally in groups), women have learned to not make eye contact. Being that I'm not like these kinds of men, I still get treated like them thus women don't make eye contact with me, in passing.

Re: OkCupid on Trial - 4 Month Experiment

#54

Post by personlurking »

"If they are hot, the girls can pick and choose which men they interact with."

There was a video on Youtube on dating and game theory where 5 women and 5 men were given a number which corresponded to their attractiveness. Basically, those who were 5's, when accepted into the 8's group were made more attractive in the eyes of the opposite sex, even if they were still 5's physically. Conversely, those who were 10's got the pick of the litter and had to do the least work to find a mate.

Re: OkCupid on Trial - 4 Month Experiment

#55

Post by wtvanhest »

goodside wrote: post #37

Hi. I work at OkCupid. I coud have saved you a lot of trouble. Your results confirm what all dating web sites have always known, and which we've made no secret about: Nearly all initial messages are sent by men. This does not mean that the site is "better" for women than for men, relative to any other form of heterosexual courtship. It's a deeply ingrained cultural, and perhaps biological, norm that men make proposal…

I just started using it after I moved to a new city. I have had some great dates, and even met some friends who introduced me to their friends so I’m certainly not complaining about the site, but I have two observations I would love to have the understanding behind.

1st, what is the purpose of the green/yellow/red dot? To me, if I see a women with a yellow or green dot I rarely if ever contact them. Also, the percentage of women with red dots is ridiculous compared with the other colors. It basically seems like a worthless indicator of anything and may even be harmful since it causes people to select the red over the yellow etc.

As a guy, if I get messages from women and they are someone I may or may not have contacted, I never respond to get my color back up to yellow. I can’t keep it there long, and have to check using a fake profile, but feel like the gamesmanship is worth it.

2nd, why say exactly when someone was online last? This encourages me to not login random times since I don’t want people seeing that I am always checking in. Even if I am bored somewhere, have nothing else to look at, I purposely don’t go to the site to let it sit. You would generate more ad revenue, at least from me if the site said, visited in last 48 hours or whatever.

FYI, for those trying the site, 2 things really surprised me. 1, the photo selection tool, I can’t remember the name, was really useful as it let me pick better photos of myself. 2, focusing on having a profile that was more about stuff I was doing was better than a profile about who I am. (that may be me specific).

Re: OkCupid on Trial - 4 Month Experiment

#56

Post by jaems33 »

mootothemax wrote: post #19
jsnk wrote: post #3

>The fact that the first stage of online dating is so heavily stacked in women’s favour doesn’t necessarily mean that it’s any easier for them, compared to men, to reach the end goal of pure love or perfect sex. Of course, men have it harder than women in the mating game. This conclusion seems like a white lie the author created for women. No one is looking for perfect mate. Both men and women choose their partners f…

Of course, men have it harder than women in the mating game. I think this is a myth. I wish years ago someone had told me all I needed to do was walk up to someone I found attractive and say "Hi, my name's Tom, how're you?" The day I started doing so, my dating life was revolutionised. Actually, this applies to most situations: parties; conferences; seeing a new face at work. With exception of special cases (maybe al…

Eh, sort of.

I'd be interested to know your race and your height.

At the end of the day, women still have deal breakers regardless of your personality, boldness, income, skills, etc.

The reason why I think women have it easier is that they can both wait for someone to approach AND take the initiative and go up to a guy if they so choose. Whereas guys not named Ryan Gosling almost always must take the initiative.

Re: OkCupid on Trial - 4 Month Experiment

#57

Post by wpietri »

ashconnor wrote: post #33
jacques_chester wrote: post #25

Earlier quoted context omitted.

So if I'm prepared to directly pay a woman, I can get company? I kinda sorta already knew that.

But you pay in air tickets and hotel rooms. It's not an escort site...honest.

What is it, then? Anybody have enough experience to give us the HN-style breakdown (as opposed to the marketing pitch)?

Re: OkCupid on Trial - 4 Month Experiment

#58

Post by teuobk »

dude_anon123 wrote: post #30

In other news, gay people don't even exist. Seriously, if you are going to attempt this type of "study" gay people are a perfect group, as you can mostly control for gender bias.

It's an interesting question: what determines how many messages a gay guy receives? What type of guys are most likely to send the initial message?

In the first four months I was on OKCupid, I got unsolicited messages from about 115 guys. In the same period, I sent unsolicited messages to about 10 guys, of which 7 elicited responses. My sense is that my numbers are higher than if I were a straight guy, but I have no idea how they compare to "typical" for gay guys.

Re: OkCupid on Trial - 4 Month Experiment

#59

Post by sophacles »

It may be an artifact of the different city problem in data collection, but assuming it wasn't, I find it interesting that the woman who was 'second most attractive' got significantly more messages than the 'most attractive'. To me this suggests interesting game theoretical stuff going on here. Does anyone know any good research/math on strategies that would optimize for this phenomenon?

e.g. I can see a thought process happening along the lines of "she is too good looking for me, I won't try, rejection sucks. I'll try this slightly less attractive woman who probably isn't hit up as often because my chances will be better, because everyone is hitting up the really good looking woman". Of course if most men follow this, you get the discrepancy I noted.

Re: OkCupid on Trial - 4 Month Experiment

#60

Post by roel_v »

yummyfajitas wrote: post #47
roel_v wrote: post #45

Earlier quoted context omitted.

No, but rounding gender distribution off to 50/50 and assuming that polygamists are a rounding error, there are an equal amount of male and female singles. Or did I miss your point?

You did. In a real life discrete time instant (e.g., an evening at the bar), most women will choose to go home alone rather than accept a second best partner. In the stable marriage algorithm, not a single woman will. I.e., the stable marriage algorithm is a poor model for reality.

But the case at hand wasn't talking about a time span that short, either. When you look at it in the long run, I'd say that both men and women have an equal likelihood of selecting a mate who is not the optimum for all qualities they look for. (i.e., to use less Greenspan-eque language, that I think that both men and woman will in the end "settle" for someone if they can't find someone who they'd consider "perfect" rather than remain alone).

However, come to think of it, I guess a prerequisite would be that there are an equal amount of men and women in the pool; which I think is so, I seem to remember from one of the okcupid analysis posts that the majority of profiles are male. Then again, on the whole, there are roughly the same amount of both, so I'm not so sure about your statement that "the stable marriage algorithm is a poor model for reality"; maybe for dating sites, but on the whole?

Post reply on HN