How to Use Game Theory to Figure Out if Your Crush Likes You Back

So when I was a dumb kid, and then when I was a dumb teenager, and then when I was a dumb young adult, I would spend a lot of time on DOES MY CRUSH LIKE ME BACK? quizzes. In the vast, swirling chaos of love and emotion, these quizzes seemed like a rare bastion of logic: if Quizilla user xnarutoxsasuke1997x’s exhaustive inventory of scientific questioning has revealed that I have a 53% shot with my crush, well, who am I to argue?

But I’m 24 now, which means that my adolescence is officially over. I am an adult now. I simply can’t rely on xnarutoxsasuke1997x’s methodology anymore. I need something more precise. More exacting. And last night, my best friend Lena, who is smarter than me and studied more econ in college than I did, interrupted me while I was whining about my love life.

“Peyton,” she said to me, “it’s time for some game theory.”

And so, ladies, gentlemen, and honoured non-binary guests, welcome to the wonderful world of romantic game theory. You’re gonna learn about math and nurse your emotional neuroses, all at the same time! Hooray!


So, first off, what’s game theory? Essentially, it’s the study of logical decision-making in any number of fields. There are game theorists in economics, politics, sports, biology, psychology; now, there are even game theorists on niche blogs that post nonsense about Robert Sean Leonard being the pope! The possibilities are endless!

What we’ll be doing today is making a game tree. This is a kind of graph (see above) which exists to determine and evaluate all the moves possible in a game, or in a decision. A complete game tree allows you to determine a sequence of moves you can follow to find the best possible outcome. Our first step is to create a game tree that contains every decision you could possibly make, and every foreseeable outcome.


To demonstrate how this applies to a crush, we’ll use Niche favourites Paris and Rory from Gilmore Girls. In one episode, the name of which I can’t be bothered to look up right now, Paris asks Rory to come to Florida with her for spring break, and they basically date each other all week and then Paris kisses Rory on the mouth and Rory reels back in horror and Paris apologizes profusely and Rory shouts, “You’re not my type!” and then their friends Madeleine and Louise tell them that they’d  make a good couple and Rory says, “No!” and Paris says, “Why not?” and Amy Sherman-Palladino owes me reparations.

ANYWAY. Let’s roll it back just a second, to the beginning of that episode, and let’s set up our game tree. We’ll start with the most basic decisions (in black), and then we’ll add some primary potential outcomes (in red).


EDIT: I only just now realized that I misspelled Elliott Smith’s name in literally all of these images. I am so sorry. Forgive me, Elliott.

So now we see that we can make two choices, and those two choices can yield three potential outcomes. But in order to evaluate these choices, and mathematically determine the best option, we’ll need to create a layer of potential outcomes that Rory’s responses could lead to. Let’s do that now.


Okay! So now we have a nice little map of some of the most likely potential outcomes here. What Paris now needs to do is decide how much she personally weighs all the potential outcomes. You can use any numerical scale you like for this, but just to keep things simple, we’ll have Paris rate the possibilities on a scale of -10 to +10. -10, naturally, is the least preferable option, and +10 is the most ideal outcome.


Now, Paris loves Yale, and she loves studying, so she decides that staying at Yale for the break really isn’t all that bad! She gives that possibility a +7. If she asks Rory and Rory says no, but just for logistical reasons, well, Paris is neutral on that. Paris gives that outcome a 0. A disastrous trip is just as scary as Rory clocking Paris’s feelings for her and rejecting her, so she gives both of those possibilities a -10. If the trip happens and it strengthens Paris’s friendship with Rory, that’s a positive – she’ll give that one an 8. But of course, the outcome Paris really wants is to form the power couple of all power couples with Rory Gilmore, and rise above the breathtakingly mediocre spate of men that Amy Sherman-Palladino has written for them to date.

Our final step here, the thing that will make the tree work, is to evaluate Rory’s likelihood of saying yes or no. This is the most difficult and subjective step — because game theory was designed for logic, not romance — but we’ll do what we can. We’ll make a guesstimate. We know that Rory’s schedule is clear for the week, which makes her more likely to say yes; we also know that the weather in New Haven is miserable, whereas Florida is lovely, which is another tilt in the direction of yes. However, we know that Rory is fundamentally Not Really A “Spring Break” Person. It’s actually fairly likely that she’d rather sit in her dorm room and read all week rather than drive down the coast and sit on a beach.

So let’s look at our complete tree again.


If Paris feels that Rory is even slightly more likely to say no than yes, she can opt not to ask Rory at all — that’s a clear gain of +7. If Paris knows Rory isn’t super-into the spring break idea, and Paris asks anyway, Paris is most likely going down a path where the best possible outcome is a 0. Would she rather have a +7 gain, or 0? Easy. Game theory says to go with this choice; this is the least risk for the greatest reward.

But, you know, nothing ventured, nothing gained. Fortune favours the brave and all that.

So let’s say that Rory is just as likely to say yes as she is to say no. Let’s say, based on her open schedule and her hatred of the weather in New Haven vs. her general non-springbreakness, we are looking at perfect 50/50 odds. In that case, in this simplified model, there is then a 40% chance that asking Rory will yield a better outcome than not asking her. Now, that’s a real roll of the dice, for sure, but a 40% chance is not nothing. 40% is substantial. If someone was like, “There’s a 40% chance an asteroid will hit your house in the next five minutes,” you’d probably leave your house, right? If you’ll recall, FiveThirtyEight’s model gave Donald Trump a 29% chance of winning the Electoral College, and… well.

Anyway, the important thing is that Paris now has a firm mathematical model she can use to objectively determine how likely it is that asking Rory out will lead to something good. How she uses the game tree, and what decision she ultimately makes, is up to her. But we have a layer of clarity, and a way of visualizing the decision, that we didn’t before! We can minimize risk! We can maximize reward! Synergy! Dynanism! Blockchain! True love!


2 thoughts on “How to Use Game Theory to Figure Out if Your Crush Likes You Back

  1. Sam says:

    This is both awesome and terrible. I am 100% going to use it. God i love science. I’ll keep you updated on how that worked for me and if my heart breaks in a million pieces i will blame the niche.

    Liked by 1 person

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s