Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

The stable marriage problem and modern dating

arvarik.com

Post reply on HN
Print view View original

201–210 of 224 posts

Re: The stable marriage problem and modern dating

#201

Post by arvarik »

Shorel wrote: post #195
albntomat0 wrote: post #43

Earlier quoted context omitted.

My understanding is the "advantage" men have over women in this article is due to the average man finding the average women more attractive than the average woman finds the average man, at least in the data used. Without removing that, and assuming everyone pairs up, the "advantage" remains.

No. According to the article, the advantage they have is they choose first because they start the interactions, while women wait to be approached. As soon as women decide to also approach, this advantage disappears.

The first graph shows that choosing first does give an advantage, however like another poster has said, the advantage is amplified with how each gender sees each other.

Re: The stable marriage problem and modern dating

#202

Post by arvarik »

viklove wrote: post #87
polote wrote: post #52

Earlier quoted context omitted.

The advantage is amplified by the distribution of preferences, but in the classic simulation the advantage was already of 13%. It is due to the fact that they are the one choosing first

Are they the ones choosing first? I think it's completely the other way around. Consider the common Tinder/OkCupid situation where a man sends out hundreds of messages to only receive a few replies, while a woman has hundreds of messages waiting in her inbox, from which she gets to select her preferred match.

This is an interesting point. The way Tinder works is a little different than how this problem is normally modeled since both people need to swipe (propose) in order to get a match and therefore proposals are bidirectional. There is still some subtleties because people don't know who swiped right on them (unless they have gold).

A more apt comparison is how Hinge works. People don't need to actively go out and 'like' people, they can just sit and wait for likes to come in and they can choose if they want to match or not.

I would be very interested to see Hinge data for how often men give out likes versus women.

Re: The stable marriage problem and modern dating

#203

Post by coldnose »

ambicapter wrote: post #71
YetAnotherMatt wrote: post #37

One factor that I didn't see addressed in the post which does make a significant impact is that men in general rate women more similarly than women rate men. See for example: https://www.marketingcharts.com/demographics-and-audiences/m... As such, male advantage is probably a lot lower in the real world, if (hyperbolic) every woman has a different man on nr 1 of their preference list, and every man has the same woman…

> One factor that I didn't see addressed in the post which does make a significant impact is that men in general rate women more similarly than women rate men. That is addressed halfway through the post. Methinks you didn't actually read the article.

This is not the same point addressed in the post. The post references the 80/20 attractiveness bias. YetAnotherMatt is referencing a different phenomenon - that men rate women similarly to each other, whereas women rate men with much higher variance. See https://www.marketingcharts.com/demographics-and-audiences/m...

Re: The stable marriage problem and modern dating

#205

Post by wolfram74 »

arvarik wrote: post #200
wolfram74 wrote: post #153

Earlier quoted context omitted.

There's one simulation I'd have very much liked to have seen the results of, and that would be the female preference advantage under the paretto-normal asymmetry. When you have a normal-normal preference, you'd assume the advantage to be equivalent for whoever was reaching, but that assumption isn't as well grounded when the distributions aren't equivalent. TLDR: do the women have a similar advantage if they are the…

So there's two reasons why there is a male preference advantage here: 1) I ran the algorithm assuming the men make the first move and 2) Male/female normal/pareto correlation in attractiveness Do you mean what would happen if the distributions were switched but men still make the first move? I could change up the script and run it again if that's what you mean!

calling it male/female at this point is kind of arbitrary, when it's the asymmetry in the distributions that I thinks interesting, but yeah, what happens if the pareto-preference population was the first mover rather than the normal-preference population that acted first. It's well documented that the first mover in the stable mariage problem has an advantage, It's why they switched from residency programs to being first movers to med students being first movers about a decade ago for The Match. This is the first I've seen it done with different preference distributions in the population. I'm curious how strong of effect the advantage is for different distributions.

Re: The stable marriage problem and modern dating

#206

Post by CarbyAu »

ksdale wrote: post #131
mcv wrote: post #117

Earlier quoted context omitted.

I think the primary requirement for making a relationship work is that both parties are committed to making it work. The problem is: we're not, because we've been fed fairy tales about True Love and we keep looking for a Mr/Mrs Perfect that doesn't exist. Finding out what you don't want is definitely important too. Some people figure that out quickly, some go through a lot of drama before they finally figure it out.…

My wife and I talk about this occasionally. There's also a sort of prisoner's dilemma that disappears when you have both credibly committed to making it work. Suddenly each disagreement is... lower stakes(?) because you know that the other person is acting in good faith to solve it as well. Committing to make it work makes it easier to make it work.

The One Trick Relationship Experts Hate! :-P

My take on your comment.

Accepting the premise of: "you have both credibly committed"(a nice phrase) -> removes reason to be suspicious.

Removing suspicion -> removes fear. Removing fear -> removes anxiety/anger. Less suspicion, fear, anxiety and anger -> Less effort

Hence: "Committing to make it work makes it easier to make it work." (another nice phrase)

On the one hand you are committing to work. Who wants to sign up for more of that!? On the other hand, it really does reduce the overall workload. Long term relationships may be less scary than people realise.

When my wife does something that irritates me, my first question is "What error in communication led to our team play resulting in this?", because I have already accepted she is not doing it from spite.

Re: The stable marriage problem and modern dating

#207

Post by dartdartdart »

minikites wrote: post #59
CoolGuySteve wrote: post #54

Earlier quoted context omitted.

I'm always surprised that while men apparently rate women's attractiveness on a bell curve, women seem to rate men on a Pareto distribution. Considering how much I hear about beauty standards and how they harm women, I almost never hear anything about male beauty standards. If society is to be improved by having more realistic female beauty standards, then there's apparently a lot more headroom for society to be impr…

Unrealistic beauty standards are absolutely a problem for both men and women, the difference is that in society, men can be valued for other things (occupation, knowledge, athletics) and that is much less true for women.

Though if men can be valued for other ways, that doesn't explain why most men are seen are far below average for the average woman. I think an amendment for your previous statement is that though they can be valued from different areas, the starting value for men is starts way below that of a woman.

For example in Asian cultures there is the idea of 666 standard for males, 6 feet tall, 6 figure salary, 6 pack, which make up less than 0.01% of the eligible population. It may be due to the gender ratio being off, but the standard leaves a lot of women unhappy in their later years

Re: The stable marriage problem and modern dating

#208

Post by asguy »

danans wrote: post #164
gerbilly wrote: post #68

Schopenhauer believed that we can't control who we are attracted to, nature does, and further that the mates nature predisposes us to aren't the one most likely to make us happy, it's the ones most likely to produce genetically compatible offspring. "Love... casts itself on persons who, apart from the sexual relation, would be hateful, contemptible and even abhorrent to the lover. But the will of the species is so mu…

> Only from this is it possible to explain why we often see very rational, and even eminent, men tied to termagants and matrimonial fiends... That's Schopenhauer demonstrating his well attested misogyny.

So what? Plenty of women write from their perspective, which could be taken as misandry.

There's something to learn by putting yourself in someone's shoes, and seeing from their perspective (e.g. the current protests, and Black Lives Matter).

Re: The stable marriage problem and modern dating

#209

Post by watwut »

Retric wrote: post #183
btilly wrote: post #122

Earlier quoted context omitted.

You are not offering evidence against the hypothesis. Divorce spikes around various life transitions occur because predictable events create predictable stresses that predictably are hard on marriages. Examples include the death of a child, children leaving the home, financial crisis, retirement and long-term illness. Pairs who proved compatibility by settling down fast last longer because they are more likely to sur…

Divorce after 25 years is statistically below average for a first marriage in the US. Waiting would have significantly reduced your odds of divorce. Overall, most relationships don’t involve marriage and many people never get married, so it’s really a question of what you consider below average. I don’t think a relationship that ends is always worse than one that continues. Many stay together out of habit, fear, fina…

> Waiting would have significantly reduced your odds of divorce.

You can not use statistic thay way. General stat for divorce does translate into individual one like this.

Re: The stable marriage problem and modern dating

#210

Post by sizzle »

poorman wrote: post #40

Having worked on a dating app startup with a 2m seed (it's since pivoted to a brewery run by active gang members), I can say the problem of dating is looked at horribly wrong. People assume dating is about finding the person "most compatible" with you. Compatibility is an illusion. Most people tend to date until they have an idea of "what they don't want". Once they know and they find someone who checks the boxes, th…

"it's since pivoted to a brewery run by active gang members"

Say what?? Sounds like there's an interesting story here...

Post reply on HN