Do men on OKCupid follow the Standard Creepiness Rule?

Written by: Randy Olson

Primary Source: Randal S. Olson

It seems that there’s an XKCD comic for every life situation that we run in to. Is there an XKCD comic for that yet?

One of my favorites, by far, is the comic titled “Dating pools.”

This comic highlighted the Standard Creepiness Rule, a.k.a. the “half-your-age-plus-seven” rule, which states that no person should date someone under (age / 2 + 7), otherwise they will look like a creeper. This seems arbitrary, but if you crunch your age into that equation, I’m willing to bet that you wouldn’t even consider dating someone under that age. (I would never consider dating someone under 21!)

If we plot the Standard Creepiness Rule out for men:


(Note that you can easily just change the axis labels in the above chart and it works just as well for women.)

It just so happens that Christian Rudder released his book Dataclysm last week, which features a chart showing us the age range that men search on OKCupid for when looking for women to date. One of my first thoughts when I saw this chart was: Do men on OKCupid follow the Standard Creepiness Rule?

(And now we see why the last panel of the XKCD comic above applies so well to me…)


Sure enough, if we overlay Rudder’s OKCupid data over the first chart, we see that men follow the rule almost exactly. There are a few spots in the mid-30’s where men seem willing to deep ever so slightly past the safe zone of non-creepiness, but that trend quickly ends by their 40’s.

Another interesting trend is how men aren’t even close to reaching the upper bound of the zone of non-creepiness. According to the Standard Creepiness Rule, it’d be perfectly fine for a 30-year-old man to date a 45-year-old woman, but apparently 30-year-old men are already struggling with the idea of dating a 37-year-old!

The following two tabs change content below.
Randy Olson is a Computer Science graduate research assistant at Michigan State University in Dr. Chris Adami’s lab specializing in artificial intelligence, artificial life, and evolutionary computation. He runs a research blog where he writes about Python, scientific computing, evolution, and AI. Randy is an ardent advocate of open science and regularly travels the U.S. to teach researchers scientific computing skills at Software Carpentry workshops.