Peak dating season is quickly approaching and soon dating sites like Match.com and OkCupid will see a surge in people searching through countless profiles to find a prospective match.
Thanksgiving, New Year’s Eve and Valentine’s Day can be wonderful when you have someone to share them with, but for the chronically single among us, these holidays can be an especially difficult time.
Online dating has become an increasingly popular option in today’s digital age, and there are so many apps and websites with millions of users worldwide, you would think that finding a compatible match would be easy as swipe, match, date.
However, the reality is that people using apps like Tinder and Bumble often find themselves swiping into oblivion with few prospects, lots of messages that lead nowhere, and an ever-growing collection of “matches” that just sit gathering dust as the swipes keep coming.
As someone who has tried their fair share of dating sites and apps, I’ve always been curious about the science and math behind the scenes, especially the ones that claim to match you with the most compatible people based on your shared interests and what you’re looking for.
Can science find your perfect match? How do the algorithms that determine your rank within the system work, and who or what decides who sees who?
You may not realize it, but behind the mega-successful dating websites like OkCupid and the apps like Bumble, there’s math, more specifically a series of algorithms that keep the cogs turning so you can swipe to your heart’s content.
“Algorithms learn from users’ preferences,” wrote Ally Marotti, a reporter for the Chicago Tribune, in a piece about perfecting your dating profile. “They gather data on users and how they interact, and calculate which profiles will appear in feeds or as matches. If a user tends not to engage with people with tattoos, the app may stop showing that person people with tattoos, for example.”
These algorithms work to take the guesswork out of the equation and can help you meet a lot of people that you probably wouldn’t if you decided to try your hand at dating the old fashioned way and leave it up to chance.
A Financial Times profile, Mandy Ginsberg, CEO of Match group, discussed the importance of algorithms in making matches. Match group is the parent company of Match.com, OkCupid, Tinder, and Plenty of Fish among other dating brands.
Ginsberg explained how Match.com’s algorithm, “Synapse,” uses many different factors in a person’s profile to find a match and also learns from the different actions of users on the site.
This means that for Match.com users, not only will their matches be based on their indicated preferences in their profile, but also on the profiles that users view, who they send messages to, and what they search for.
Tinder’s algorithm claims to weed out inactive users and keep them in the back of the line. Other algorithms may make sure to keep the most popular profiles in a certain area at the head of the queue.
In Tinder, every user has a ranking called an Elo score, and the fact of the matter is, who sees you and how often you’re profile is seen boils down to that ranking.
“It’s not just how many people swipe right on you,” Sean Rad, Tinder’s CEO, told Fast Company. “It’s very complicated. It took us two and a half months just to build the algorithm because a lot of factors go into it.”
For Tinder, they say the Elo score measures desirability in the hopes of matching equally desirable couples.
It’s important to note that although some dating websites claim otherwise, dating algorithms have yet to crack the code of human attraction and love, and no amount of math will find you your perfect match.
Researchers from Northwestern University and the University of California, Davis found that out the hard way after conducting a series of speed dating experiments where romantically unattached individuals were asked to fill out questionnaires rating their own qualities and attractiveness and what they were looking for in a partner.
Then the participants took part in a speed dating exercise, where people were matched up and given four minutes to make a connection before moving on to the next person.
After the speed dating was done, the participants told the researchers which people they liked and which they didn’t.
The researchers then created a machine learning algorithm based on all the data from the questionnaires and the participant’s responses to the dates to see if machine learning could predict compatibility.
After all the analysis and number crunching was complete, the researchers found that machines and algorithms were useless at predicting human compatibility.
“We couldn’t predict compatibility at all,” said Paul Eastwick, a leader of the study, on the podcast Science Verses, while discussing the results. “It’s very hard to predict how much people are going to like each other once they first meet.”
If you find yourself looking for love and think online dating is the way to go, or if you’ve been a member of every online dating website and still have yet to find a good match, unfortunately, science can’t magically find you a partner.
However, what these algorithms do well is expose you to tons of people while learning from your online dating activity.
Don’t get too caught up in your rank, or your number of matches because as we’ve learned, these algorithms know very little about actual matters of the heart.
However, online dating can be a great way to get things started, put yourself out there without looking for an expensive local matchmaking service, and meet different people.