​

Last week, I posted an analysis concluding that Blue Pill and Red Pill were essentially explanations of the dating market for two separate groups (with BP explaining top 70% of men / 97% of women and RP explaining the remainder). Based on an OKCupid study exploring message distributions based on perceived attractiveness, I concluded that men focused more on looks when sending messages to women. Women, on the other hand, largely followed a two-tier strategy: the top 70% of men based on attractiveness got a relatively balanced distribution of messages with looks being a helpful, but not anywhere near essential, asset to have.

On the other hand, the bottom 30% of men got very scarce attention in a brutal looks-focused economy with a message distribution worse than the conditions of many war-torn third world dictatorships. Given that women only found 20% of men physically attractive, I posited that women had very divergent perspectives on what made men attractive to them, effectively creating a low-response but relatively equal message distribution for the top 70% of men. These results were found mainly by computing Gini coefficients (these are basically numbers between 0 and 1 that measure how unequal wealth is concentrated in an economy, with 0 denoting perfect equality and 1 signaling a world where one person controls all the wealth) for both men and women as well as different subpopulations.

However, I neglected to examine the first stage of the pipeline itself: likes/swipes. Given that many dating apps require mutual likes/swipes prior to messages even being possible to send (and that only a very small percentage of matches end up having messages - men message 7% of their matches while women message 21%), the like/swipe stage itself will have a large impact on creating these distributions. While I did see studies (Quartz, Medium, etc.) exploring this stage and concluding the presence of horribly unequal distributions, they all seemed to make inexplicably rookie mistakes that led me to ignore their conclusions (e.g. the Quartz study randomly doubled every country's Gini coefficient to compare to their undoubled male and female Ginis; Medium effectively concluded that men's larger willingness to swipe right equaled that 80% of women only chased 20% of men, a comically strawman argument). Upon taking a second look, however, the first half of the Quartz article (where the author cited a Hinge engineer) actually had a useful like distribution to explore, which (if you are paywalled like some of my devices) shows the following like shares:

top 1%/top 5%/top 10%/bottom 50%

Men to Women: 11.2/30.6/45.7/7.9

Women to Men: 16.4/41.1/58.0/4.3

These distributions are indeed quite unequal -- men give women a likes economy with a minimum Gini coefficient of 0.5676 (at best the fifth most unequal economy in the world; though the true Gini is likely much worse. For reference, recall that 20% of men getting 80% of the likes is 0.5999) and women giving men a likes economy with a minimum Gini coefficient of 0.6675 (more than any country today)! Ultimately, Hinge at its initial stage is largely a contest where everyone just likes the most desirable individuals and hopes for the best. Note, however, that Hinge's data does not state what about these profiles leads them to be so well liked (just that they are). However, if looks/photos are the main contributor to what produces likes (as intuition might predict, though intuition is a very noisy signal), then the app is basically just a "hot or not" contest - in other words, a stripped down Instagram clone. It should be noted, however, that this cartoonish distribution does not carry over to messages. Women get their 0.5676 Gini dropped to 0.3960 (a little better than the US), while men get theirs dropped from 0.6675 to a far more reasonable 0.3284 (around where Germany is).

Should a likes to looks relationship be explicitly proven, this finding adds two conclusions to those from last week: (1) women do not have massively varying preferences for what they find attractive (I previously concluded they didn't, so this is a refutation of my previous claim), and (2) "Chads" do exist insofar they get showered with tons of likes, but this ceases to be the case in the message stage.

So what results in the massive shift between "top 10% gets all the likes" to "messages are pretty balanced"? Far and away the main factor is that most swipes are not matches. Per this swipestats.io infographic reposted on Reddit, men swipe right on 53% of women while women swipe right on 5%. Women's match to right swipe ratio was found to be 36% (other studies have it as low as 10%), which is still low enough where it is possible that a majority of the likes women send out are all concentrated on the top 10% of profiles (i.e. Hinge data) but are mostly not reciprocated and the likes dispersed to the remainder of the population (which women end up having quite a bit of disagreement on who those men are, at least among the 30th to 90th percentile of men) are far more likely to be matches.

TLDR (both previous and current):

  • In the likes stage, men and women mostly shoot for the very top profiles (I assumed this to be based on looks for the analysis, but given the lack of explicit data found to support this am open to a redefinition of what "top" is).
  • However, most swipes are not matches, meaning that the variation of women's preferences outside of the top 10% end up leading to the relatively equitable distribution of messages for men above the 30th percentile in looks. The bottom 30% of men by looks get little to no attention in any stage.
  • "Chads" do exist, but only in the likes stage and end up becoming enforcers of a far more equal dating scene as opposed to perpetuating RP concepts such as the "cock carousel", "80/20", "AFBB", etc. "Stacys" play the same role to the benefit of their female counterparts.
  • Men's lower threshold for women means that only ~3% are "femcels", while women's much more stricter looks threshold makes about ~30% of men "incels" on online dating apps (the quoted terms do not have anything to do with ideology and are simply using the literal definitions of "involuntarily celibate").
  • Men from the 30th to 90th percentile will most likely still struggle to get solid amounts of female attention, but end up having a wide array of areas to improve on (looks, warmth of personality, earnings potential, etc.). It's a brutal game for them, but there is light at the end of the tunnel! For those below (as well as for women in general), focusing on looks is your best bet.