3 ms·
Is this more than just speculation? According to the linked occupid data (https://web.archive.org/web/20170330131859/https://theblog.okcupid.com/your-looks-and-
by ketzu 7y ago
Is this more than just speculation? According to the linked occupid data (https://web.archive.org/web/20170330131859/https://theblog.okcupid.com/your-looks-and-your-inbox-8715c0f1561e https://web.archive.org/web/20170330131859/https://theblog.o...) while women underrate attractiveness of men, the distribution of messages fllows the attractiveness curve: Men receive about the same amount of messages independent of attractiveness (https://web.archive.org/web/20170331084200im_/https://cdn-images-1.medium.com/max/800/0*aWz0dYzuUR7PO3dP.png https://web.archive.org/web/20170331084200im_/https://cdn-im...). Especially compared to the behaviour of men, which tend to value attractiveness much more.
One hypothesis is that it depends on the service you use. E.g., Tinder vs Okupid, where Tinder reinforces decision based on looks while Okupid encourages much more detailed profiles.
- animal531 7y agoThe study from OkCupid that I saw a long time ago showed that the top 78% of women were only interested in the top 22% of men (where top denotes a score that rates how often others are interested in you, the amount of messages you receive etc). The reverse then meant that the bottom 78% of men had to compete for the attention of the bottom 22% of women. They retracted that study after a while, but other dating sites roughly get the same percentages eventually.
- watwut 7y agoIf it is 80% male, that would mean that the women generally simply matched available males and then were done with it all. There is no way to achieve monogamous one-to-one pairing for everyone in place that is 80% male. Of course only minority of men succeeds. And while here it is not exactly 1-to-1, it is not actually all that much far off.
- JohnJamesRambo 7y agoIt’s interesting to see the Pareto Principle working here as well.