The official blog is the cleaned-up version, they removed the most interesting articles when they sold out to match.com

Famously, the article "Why You Should Never Pay For Online Dating" got deleted during the acquisition.

[1] Mirror: https://www.gwern.net/docs/psychology/okcupid/whyyoushouldne...

It's a good article, and one of the key takeaways:

If a dating site makes you pay to send messages, then they have an incentive to make you send messages to inactive accounts rather than active accounts, since people with inactive accounts have to pay in order to reply.

There were also articles that ran counter to popular gender theory/politics.

This is off memory but I believe their stats showed that men rated women's photos on what resembled a classic bell curve, shifted to the right slightly. Ie, dudes were generally reasonable if not a wee bit overly kind.

Women were exceptionally brutal in ranking men's looks. Women's ranking of men was a triple-diamond ski hill with damn near most of the userbase falling in (I believe, again, this is from memory) the bottom third. "Women are held to unrealistic beauty standards" seems to be more than a bit of projection.

Funny story: I got banned from OKCupid once for calling out other volunteer flagmods (people suckered into wasting their time policing OKCupid user photos for free) for body shaming and transphobia (the latter almost exclusively toward transfemmes, but both coming almost exclusively from white, straight women.) Hilarious.

Some "controversial" blog posts they deleted.

How men and women perceive attractiveness https://archive.is/489UV#selection-282.0-282.1

We finally answer the age-old question: should men keep their shirts on? https://archive.is/9fJQh#selection-282.0-282.1

We Experiment On Human Beings! https://archive.is/QNCbf#selection-278.0-278.1

How Your Race Affects The Messages You Get https://archive.is/kMP32#selection-278.0-278.1

Don’t Be Ugly By Accident! https://www.gwern.net/docs/psychology/okcupid/dontbeuglybyac...

More can be found here:

https://www.gwern.net/docs/psychology/okcupid/index

i'd love to pay for the substack of whoever wrote that blogpost, they have to be sitting on a mountain of unpublished insights.

What people say they sort on: personality, values, morals, political views, friendships, etc.

What people sort on when they don't think they're being observed: genes

If I'm remembering correctly, it was way more specific than that. The only genetic thing is there were some extreme racial biases. You really don't want to be an Asian man or a Black woman on a dating site.

But plenty of non-genetic things. Back when they let you list an income range, men with higher incomes got much better response rates. Men heavily favor women who are at least ten years younger than them. There were weirdly specific things about your photos that mattered, too, like you'd get a much better response rate if other people weren't in the photo with you, you'd get a better response rate if you weren't looking at the camera. Women were more attractive if they were smiling but men did better if they were not smiling.

Christian Rudder used to publish gold mines for anyone who wanted to just game hot-or-not. Plenty of this was stuff you could control, not genetic. Though I guess you can't exactly control your age even if it isn't genetic. It also let you sift through the lies, like women would always say they were turned off by shirtless pictures, but based purely on response rates, that definitely wasn't true for men who actually had lean bodies.

I haven't been on OkCupid in a long time. I think they first started publishing these data mining studies in maybe 2006? A lot of the old blog posts were purged after the Match purchase. I'm sure someone saved them off or they might still be on the wayback archive, but I don't even remember what the url for the blog was at this point and I doubt it's still even public. Your best bet at this point is probably just to read Dataclysm, the book Rudder ended up writing about all of his findings.

Interestng, have a link for the study? Can't seem to find it

Not OP, but https://www.gwern.net/docs/psychology/okcupid/raceandattract...

This talks about a follow-up study 5 years after the first one, searching "okcupid race and attraction" doesn't find me the link to the 2009 article.

I'm pretty sure 99% of people would openly agree that physical attraction is a core element of partner selection.

Not anymore, but before online dating people were hiding it much more

People were trying to hide that they actually want to be sexually attracted to their partner?...

To me (a not attractive man) yes. But I'm from Eastern Europe, the culture is different there.

Out of curiousity in terms of attractiveness and dating or choosing a partner, what do you feel is culturally different in Eastern Europe?

Generally non-US countries are lagging behind by a few years in dating culture. In Colombia for example I was already used for a foodie date (and some other girls have tried quite aggressively to go to dinner on the first date), but here in Eastern Europe it's not trendy yet. Younger girls are looking for equality based relationships, 30+ girls are looking for more traditional marriage.

OKCupid doesn't allow you to sort on anything anymore. It's all part of their business model of preventing people from creating permanent relationships. Yes, some slip through, don't @ me.

The way I see it, people sort on both genes (aka looks) and personality/values/morals/etc.

The thing is, by just scrolling through the feed/list of people to swipe on, you don't get to see much personality, mostly looks. To get to personality, you gotta talk to the person.

So when you swipe, you filter mostly by looks. And once you match and start talking, that's when you filter by personality.

Yes, one can say that you can get personality from their bio/profile, but that's such a non-consistent metric with tons of noise and misleading data (cliched/copypasted bio, nothing standing out, outdated bio, etc.). You need to have a conversation with a person to get a gauge of their real personality (of course, exceptions apply; if you see a profile/bio claiming that vaccines give kids autism and that the only valid covid treatment is essential oils, you kinda already have an idea who you are dealing with).

And out of all those people you spend a lot of time intensely reading thru profiles of before swiping, most of them won't even match with you. So imo, it makes sense to initially swipe based purely on looks and a 5-10 second glance at the profile, and then try to gauge their personality only after you match.

One of the founders published an excellent book that is an extension of the blog: https://www.goodreads.com/book/show/21480734-dataclysm

Wow, the in app questions are pretty politicised to say the least.

Is that something that really matters in dating these days?

Sadly a lot of their most interesting posts went away after they were acquired