“My bounty is as boundless as the sea,
My love as deep; the more I give to thee
the more I have, for both are infinite.”
-William Shakespeare, Romeo & Juliet
When Juliet spoke of her love for Romeo as infinite, she was speaking metaphorically. After all, love is more art than science, and can’t be quantified on any mathematical scale. Or can it?
Certainly many have tried. One Australian PhD student performed a statistical analysis to calculate his odds of finding a girlfriend, concluding his odds were pretty much zero. Another man, in London, arrived at essentially the same conclusion, though happily, he later married.
A Jewish woman in Philadelphia had arrived at a similar conclusion as these, calculating that of all the men in her city, there were approximately 35 men that might be a suitable mate for her. She decided to use data analytics to her advantage though — she wrote out everything she was looking for in a match, turned them into data points and produced a scale to help rate potential dates and whether she should go out with them. She also used data to optimize her own online dating profile. This all worked remarkably well — she fell in love, had a daughter, did a TED talk on her love analytics, and wrote a book called Data: A Love Story.
This isn’t a new solution to our oldest problem. As early as 1611, a famous astronomer tried to use math to find himself a wife, interviewing and keeping data points on various candidates. More recently, an article in a 1960 issue of Scientific American described the Secretary Problem, a mathematical problem that attempts to calculate how many people an executive would have to interview to find the ideal secretary (using Optimal Stopping Theory). Also known as the marriage problem, it has also been used to find the ideal time to pick a spouse. Optimal Stopping Theory, by the way, says that you should stop 37 percent of the way into your search, and then pick the first person that comes along who is better than the previous 37 percent.Easy, right?
Thanks to online dating, much data is available about the successes and failures of dating, marriage, love and relationships in modern times. OK Cupid, one of the world’s largest online dating website, has been collecting extensive user data for over a decade, which are used to match potential mates. The site even posts analytical conclusions on its blog, a goldmine of information for the mathematically-inclined searching for love.
The site asks users multiple choice questions — after you answer one, you are able to view another person’s answer to the same question when looking at his/her profile. The site uses algorithms to utilize the users’ answers and profiles to calculate the likeliness of a match on a 100-point scale. Users can also leave a comment to explain an answer on questions, of which there are hundreds. One blog lists the questions that have the most commentsleft on it, which range from “Are you annoying?” to “Would you sleep with a serial killer?”
OK Cupid’s blog also offers interesting data that shows how one might statistically improve his or her chances at success in finding a mate. They have found, for example, that liberal women have more success and that the optimal length for a successful first message is 200 characters.
A lot of data analysis has also been done on the success or failure of marriage. One group of researchers found that they could predict with 90 percent accuracy whether couples would divorce after listening to conversations they had with each other. They measured certain vitals, such as heart rate, blood pressure and facial expressions. The single most accurate accurate predictor of a marriage’s success boiled down to a remarkably simple statistic — whether the couples were negative or positive during their interactions, also known as the negativity threshold.
In her TED Talk about the mathematics of love, Hannah Fry says, “For me, equations and symbols aren’t just a thing. They’re a voice that speaks out about the incredible richness of nature and the startling simplicity in the patterns that twist and turn and warp and evolve all around us, from the world works to how we behave.”
The nature of true love will always remain mysterious and for some, elusive. But you can take comfort in knowing that there is a certain predictability to human behavior and the chances that you’ll find happiness can at least be optimized. In the meantime, if you find someone you like and you want to fall in love, try this.
One thought on “The Analytics of Love”
Cool stuff. I actually developed a number of related analytics. One is the “Taxi Driver Analytic”, which doesn’t work in the US, but where I come from, I could measure the level of feminine energy I was transmitting by the percentage of taxi drivers who asked for my number. The more pertinent analytic is the measure of how much you love someone. I concluded that I could accurately measure how much someone meant to me by the amount of time I could sustain anger at the person without calling to apologize. The people closest to me in my life were somewhere around 10 minutes (daughter, partner, etc.). With a partner, it should be mutual, that is, the partner should also have a fairly low tolerance for being angry at you. Offspring are different. In those relationship, it’s almost your child’s job to be angry at you, so you as a parent don’t want to spend too much time in dissonance, but a teenager is allowed to be angry at you perpetually.