There’s a mathematical theory that claims to know exactly when you’ve dated enough people to make a smart decision about settling down. It’s called the 37% rule, and it comes from optimal stopping theory—a branch of mathematics that, of course, someone eventually applied to romance.

Psychology Today notes the theory has been used to solve everything from hiring decisions to real estate, and the dating version works the same way. Before you’ve dated anyone, you have no real baseline for what’s out there. So the theory suggests spending the first 37% of your dating life purely in observation mode, treating those relationships as data points rather than candidates. Once you’ve cleared that threshold, you commit to the next person who outperforms everyone in your sample. According to the math, searching beyond that point produces diminishing returns and a low probability of finding anyone better.

In practice, the calculation is harder to execute than it sounds. If you’re going on roughly one date a week for a year, that’s about 50 dates total—meaning your first 19 are just calibration. Alternatively, if you’ve given yourself a decade to find a partner, the first three years and eight months are essentially a research phase. The 13th day of the ninth month is apparently when you start actually choosing.

Maybe Don’t Take the 37% Rule as Gospel, Though

The problems with this approach are about as numerous as the dates themselves. For one, life doesn’t sort itself into neat little categories. Situationships, work crushes, and blurry almost-relationships don’t come with timestamps. Do those count? The theory also torches the concept of high school sweethearts and anyone who met their person on the early end of the timeline. And it assumes you’re the same person throughout—that your taste, self-awareness, and circumstances stay fixed. Anyone who dated in their early 20s and then looked back in horror knows that’s not how it works.

The bigger issue is that the person you are at 22 isn’t the person you are at 32—and the dates you went on during those years reflect that. Someone who met their partner before fully knowing themselves might have walked right past them. The sample only works if the sampler is self-aware enough to read it correctly.

The better approach to the 37% rule is to hold it loosely rather than follow it literally. Get enough experience to know what you actually want. But once someone fits, stop running the numbers. Some decisions just don’t have a formula, and this is one of them.