4 ms·

"I don't really see how friendships and relationships in general don't align with that sort of model. Can you reduce them to purely mathematical transaction val

• luckylion 7y ago
  You cannot reduce any decisions on any markets, at some point you're going "huh, I guess free will or the universe or randomness? idk", but that doesn't make the models useless. They can be tested. Take Tinder, make predictions how people behave differently if e.g. you change the gender balance, test it. My point about "you cannot reduce it" isn't about the model, but about relationships themselves. You can't just go "look, here's the fact sheet" because it's way too complex and we don't know all the variables etc pp, but overall, in larger numbers, you can very much use models to predict general behavior. Not on an individual level, but very much so on a group level. Much like in any market, where you will always have outliers that act totally different than what your models predict, but you'll also have the bulk that behaves the way your model predicts. If they don't, your model isn't good. That's not the case for models looking at relationships as markets though.