A PhD student from Boston used algorithms and robot profiles on OkCupid to find his wife in 90 days.
Chris McKinlay, a UCLA maths student, was working on his dissertation when he devised a system to crack OkCupid’s matching systems.
He realised that the 350 questions a user answers would only make you compatible with women who answered the same questions as you.
The questions he answered only generated 100 women above 90% compatibility, in a city with 2m women.
So the 35-year-old set about creating a system that used 12 robot profiles he wrote scripts for, to gather data on 20,000 women.
McKinlay, speaking to Wired, said: “I started thinking about it when I was in dissertation mode, so I was applying grad student mentality to everything back then.”
This data was then split into 7 categories, to which he applied other algorithms in order to sort via preferences like proximity to LA and San Francisco, as well as age and other personal preferences.
His processes were recognised by OkCupid and the fake profiles were blocked, but McKinlay already had the data he needed.
He went on 87 dates with women all closely matched with him, and on the 88th, he met his future wife, artist Christine Tien Wang.
Mckinlay explained the mathematical system he had used to generate these dates to Wang, who “thought it was dark and cynical. I liked it.”
Wang said: “People are much more complicated than their profiles.
“So the way we met was kind of superficial, but everything that happened after is not superficial at all. It’s been cultivated through a lot of work.”
Speaking to Wired, McKinlay said: “It’s not like we matched and therefore we have a great relationship.
“It was just a mechanism to put us in the same room. I was able to use OkCupid to find someone.”
To read further into the details of McKinlay’s mathematical processes, read the full piece in Wired here.
McKinlay has also written a book, called Optimal Cupid, Mastering The Hidden Logic of OkCupid, which you can buy here.