How Tinder “Feedback Loop” Forces Men and Women into Extreme Strategies
The first study of swiping strategies on Tinder shows just how different male and female mating behavior can be.
Back in 2012, a new craze swept the Internet centered on a dating app called Tinder. The app shows users pictures of potential dating partners in their local area. Users swipe right if they like the picture or swipe left if they don’t. When two users like each other, the app puts them in contact with its built-in messaging service.
Tinder changed the ground rules for dating apps. Until then, most dating services had found matches using a range of factors such as shared interests, age, future plans, and so on. On Tinder, all that matters is first impressions.
That’s interesting for anthropologists who have spent decades studying how people select mates. This research is hard because there are so many factors to take into account. Tinder, on the other hand, is a much cleaner environment, since it is based only on first impressions, and so has fascinating research potential. And yet nobody has studied mating strategies on Tinder.
Today that changes thanks to the work of Gareth Tyson at Queen Mary University of London in the U.K. and a few pals who have studied mating strategies on Tinder for the first time. Their work reveals some remarkable differences between different groups using Tinder, some counterintuitive phenomenon, and they have even come up with some tips to help men in particular to maximize their chances of success.
The team does not have access to raw data from Tinder and so developed another way to gather information. They set up 14 different Tinder accounts designed to mimic ordinary users. They created three accounts using stock photos of white men, two accounts for white male volunteers with several pictures, and as controls a male account with no picture and a male account with a picture saying the account had been disabled. The team set up a similar set of accounts for white females.
They chose only white men and women, rather than a variety of ethnicities, to reduce the number of variables in the experiment. They also located the fake accounts in London to reduce location-based variability. However, they set up the volunteer accounts in New York to prevent the volunteers from being recognized in their home cities (although why they didn’t recruit volunteers in New York and set up their accounts in London isn’t clear).
All the accounts used pictures of ordinary looking people. “We emphasize that our study is not intended to measure attributes like beauty or attraction,” say Tyson and co.
Next, the team created an algorithm that searched through each profile’s matches, logged the details of each one—age, sex, bio, and so on—and then liked them all. In total, they crawled 230,000 male profiles in this way and 250,000 female profiles. By counting the likes each profile got in return, the team could determine the percentage of other users who responded favorably.
The data analysis reveals some interesting differences between the sexes. For a start, men and women use entirely different strategies to engage a potential mate on Tinder. Men tend to like a large proportion of the women they view but receive only a tiny fraction of matches in return—just 0.6 percent.
Women use the opposite strategy. They are far more selective about who they like but have a much higher matching rate of about 10 percent.
But curiously, the vast proportion of matches came from men, whether for the team’s male or female profiles. “Even though the male:female ratio in our dataset is roughly even, on average, 86% of all the matches our male profiles receive come from other men,” say Tyson and co.
This suggests that homosexual men play an important role on Tinder. “Homosexual men are far more active in liking than heterosexual women,” say the team.
Just as puzzling is that one of their male profiles—the one showing the account to be disabled—received all its matches from women. Just why this happened isn’t clear.
The way men and women gain matches is different, too. Men tend to pick up matches slowly over time, while women gain matches quickly, achieving more than 200 matches in the first hour. In total, the team received 8,248 male matches but only 532 female matches.
Another difference is the way men and women behave once they have received a match. Women tend to be far more engaged and more likely to send a message to their match. “Overall, we find that 21 percent of female matches send a message, whereas only 7 percent of male matches send a message,” say Tyson and co.
And women also take more time over their messages. Almost two-thirds of messages sent by men occur within five minutes of the match taking place, but only 18 percent of those sent by women. And men’s messages are shorter, too, averaging just 12 characters, presumable to say hi, hello, or something similar. By contrast, women’s messages are 122 characters long on average.
The number of pictures on a profile makes a difference, too, particularly for male profiles. “With a single [male] profile picture, after four hours, only 44 matches were made, whereas this increased to 238 with three pictures,” say the team.
Bios also make a difference. “Without bios, our male stock profiles received an average of 16 matches from women; this increases fourfold to 69 with a bio,” say the team.
That suggests two simple things men can do to significantly improve the number of matches they get on Tinder—include a bio and more photos.
Finally, the team sent out questionnaires to frequent Tinder users to ask about their motivation for using Tinder and the strategies they employ. Interestingly, men say that the low matching rate is one of the factors that causes them to like a higher proportion of the women they see on the service.
That implies the existence of a vicious circle of behavior that forces men and women into more extreme strategies. “Our findings suggest a ‘feedback loop,’ whereby men are driven to be less selective in the hope of attaining a match, whilst women are increasingly driven to be more selective, safe in the knowledge that any profiles they like will probably result in a match,” say Tyson and co.
One end point from such a feedback loop is that men will end up liking all the women they see, while women will be guaranteed a match every time they like somebody. In that case, Tinder will effectively be broken.
A more likely outcome is that some evolutionary stable strategy will emerge; and perhaps has already. Only Tinder, using its own data, will know, but the company is not saying.
Either way, that’s interesting work.
Ref: arxiv.org/abs/1607.01952: A First Look at User Activity on Tinder
The AI revolution is here. Will you lead or follow?
Join us at EmTech Digital 2019.