How this instrument works
In 2010, economics researcher Peter Backus, then at the University of Warwick, wrote a paper titled "Why I Don't Have a Girlfriend," borrowing the structure of astronomer Frank Drake's famous equation for estimating the number of contactable alien civilizations in the galaxy and applying it to dating instead. Both problems share the same shape: start with an enormous population, then multiply by a chain of independent qualifying probabilities to shrink it down to a small, realistic count.
Each fraction in the chain represents one qualifying criterion narrowing the pool further — the fraction of the right gender, living in your area, in your preferred age range, with your preferred education level, whom you'd find attractive, who are single, and who'd likely be compatible. Multiplying probabilities together like this assumes each criterion is statistically independent of the others, which is a simplification real dating preferences don't perfectly satisfy — people's traits correlate with each other in ways a straight multiplication ignores — so the true number is often somewhat higher than what the naive chain produces.
Treat the output as a fun, thought-provoking estimate of just how selective a stated set of criteria really is, not a literal headcount of people you'd date. The most interesting part of running the exercise is usually adjusting a single fraction — widening a preferred age range, say — and watching how much the final number swings; because every fraction multiplies rather than adds, small changes near the start of the chain compound into large changes at the end.
- Enter your population pool — the city, region, or country you're drawing from.
- Enter each qualifying fraction from 0 to 1: gender, location, age range, education, attractiveness, single status, and compatibility.
- Read the estimated number of potential matches.
- Adjust any single fraction to see how much it swings the final estimate — usually the most revealing part of the exercise.
Worked example — a city of one million people
Starting from a population of 1,000,000: applying gender (0.5) gives 500,000; location (0.1) narrows that to 50,000; age range (0.2) to 10,000; education (0.3) to 3,000; attractiveness (0.1) to 300; single status (0.5) to 150; and compatibility (0.2) to a final estimate of exactly 30 people.
That steep drop — from a million down to 30 — illustrates why the exercise resonates: each individual fraction sounds reasonable on its own (half the population is the right gender, a fifth is the right age), but multiplying seven such fractions together compounds quickly, which is precisely the point both Drake's original equation and Backus's dating adaptation are built to demonstrate.
Questions
Where does this idea originally come from?
It adapts the Drake equation, formulated by astronomer Frank Drake in 1961 as a way to structure discussion about the number of communicating extraterrestrial civilizations in the galaxy — multiplying factors like the rate of star formation and the fraction of planets capable of supporting life. Economics researcher Peter Backus repurposed the same multiplicative-fraction structure for dating in his 2010 paper "Why I Don't Have a Girlfriend," swapping astronomical factors for demographic and romantic-compatibility ones.
Does this account for whether the match would also be interested in me?
Not directly — the calculation estimates how many people in the population meet your stated criteria, which is a one-directional filter. It doesn't factor in the reverse condition (whether you'd meet their criteria too), so the true number of realistic, mutual matches is generally smaller than this one-sided estimate suggests, not larger.
Why does the final number swing so wildly when I change just one fraction?
Because every fraction multiplies the running total rather than adding to it, so a change near the front of the chain (like population size or gender fraction) has an outsized effect on everything downstream. Doubling any single fraction anywhere in the chain doubles the final result, which is why loosening even one criterion slightly — widening an age range from a 5-year window to a 10-year one, for instance — can move the estimate by a large multiple rather than a small increment.
Is this meant to be taken as a literal, accurate count?
No — it's a simplified model built for perspective and reflection, not a rigorous demographic calculation. The independence assumption behind multiplying separate fractions together doesn't perfectly hold in reality (traits like education and location often correlate with each other rather than being statistically independent), and the inputs themselves are necessarily rough personal estimates rather than measured data, so the output should be read as illustrative rather than precise.
What happened to Peter Backus, the paper's original author?
Despite calculating in his own paper that he had roughly a 1-in-285,000 chance of finding a suitable partner in the UK, Backus met a woman named Rose a couple of years after publishing the paper and got engaged — a widely reported epilogue to the story that the University of Warwick itself later wrote about, and one that underscores the paper's own point: it was written as a playful, thought-provoking exercise in applying a scientific framework to an everyday question, not a genuine prediction of romantic odds.