The Red and Blue Pills are Both Correct...Just for Different Groups

At the cornerstone of the RP ideology is the 80-20 rule, which stipulates that 80% of women only chase after 20% of men ("chads"). Some individuals have posited even lower percentages of chads (e.g. 10, 5, 1) based upon Tinder data. Examples include:

https://www.reddit.com/r/dataisbeautiful/comments/mbf6wg/oc_despite_being_far_more_selective_women_still/

https://www.reddit.com/r/dataisbeautiful/comments/cuazgf/oc_my_tinder_experience_as_a_24f_living_in/

One of Blue Pill's counterarguments to the OKCupid study and its successors "confirming" the 80/20 rule was that women tend to message down the middle in attractiveness, while men chase the most attractive women (all except the top ~5% that most feel are out of their league). I decided to explore the OKCupid data for myself to try and contemplate my own view on the matter. Specifically, I computed Gini coefficients (for the uninitiated, 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. Like the study, I solely used perceived attractiveness and messages received from the other gender as my two variables.

​

Findings:

• The dating market for men has a minimum Gini of 0.3284 (so about as equal as Japan's economy and slightly less equal than Germany's) while the dating market for women has a minimum Gini of 0.3960 (almost as unequal as the US economy). I use "minimum" because I assumed that all groups with the same perceived attractiveness got the same quantity of messages, which ends up dropping this number. Tldr; men care more than women about looks.
• If we exclude the ~26.5% of men ranked least attractive (0), the dating market for men has a minimum Gini of 0.2592 (about where formerly Communist Eastern Europe is today). Removing the ~57% of men given the second least attractive score (0.83) or less, we get a minimum Gini of 0.2141, more equal than any country today. In short, if you are in the top 70% of perceived attractiveness among men, the impact of looks on your dating life is quite small and you should focus on other attributes to up your game. Indeed, given that women only swipe right on ~2% of men on average yet find ~19% to be above average in appearance, it is clear that women value more than looks as long as your attractiveness is within the top 70%. For these men, it's about as close to an equal opportunity shot that they will ever experience in life.
• On the other hand, the bottom 26.5% of men are dealt an exceptionally brutal hand, with the lack of attention that women provide to them basically creating the inequality based on attractiveness that RPers repeat ad nauseum. Taking out people rated a 0.83/5 in attractiveness increases the male dating market's minimum Gini coefficient to 0.3612, 0.4303 if you take out men rated at 0.83 or 1.67, and a whopping 0.4493 if you take out men ranked 0.83, 1.67, or 2.5! The latter two numbers are essentially equal to the inequality in former colonial states in Africa, where dictators continue to monopolize all the country's wealth while letting their people suffer through massive amounts of extreme poverty. The same effect is not seen for those rated at 0.83/5 attractiveness, with no combination excluding those rated "least attractive" surpassing a minimum Gini of 0.29. For perspective, "20% of men getting 80% of the sex" corresponds to a Gini coefficient of 0.6000, which is not actually that far from the experience the bottom 30% of men looks-wise are getting from the dating scene (recall I have been calculating minimum Ginis). These people are ultimately getting an online dating market about as blatantly barbaric and grotesque to what RP proclaims, which likely explains why RP's message is so compelling to them (it is their lived truth).
• Despite the greater emphasis males have on the other gender's attractiveness (which women can rightly complain is shallow of them), the larger amount of messages they send in total largely makes up for this discrepancy. Depending on the methodology (one example here, although most studies I found used rather indirect measurements so I'm happy to change this number if someone finds a more direct study), men on average appear to message 3.6-6.2x more than women do. Thus, women rated "least attractive" by men (the bottom 6%) would be expected to get 0.6-1.0333x the amount of messages that a man would get on average, while men rated "least attractive" (the bottom 26.5%) and 0.83/5 (the next 30.5%) get just 0.415 and 0.754, respectively. Heck, the top 0.1% of men (think male models whose careers are 100% dependent on how they look) only end up getting more attention from the opposite gender than the bottom 30% of females by attractiveness. Ultimately, men do make up for the greater inequality they create by their focus on looks by sending a larger number of messages overall, decreasing the proportion of "femcels" to less than 3% (about one ninth of male incels). Note that I am defining incel here to strictly mean those neglected on online dating apps and not attributing any ideology whatsoever.

​

From what I've discovered, it appears that RP offers its explanatory power to the bottom 30% of men in terms of looks, while BP provides a lodestar for those in the top 70%. The larger inequality that men create for women in terms of looks understandably creates great frustration (see the Ultimatum Test) is offset enough by greater messaging frequency, which provides a bulwark protecting such neglected women from the profoundly painful effects of ostracism and neglect.

Unanswered Questions:

• How hard is it to go up from 0/5 to 0.83/5 for both men and women?
• Is the 30%/3% of "incel"/"femcel" proportions an iron rule in an OLD market? Or can it be lowered, as RP says, by those in this quagmire working to make themselves look better?
• What BP factors are most important for those in the top 70% of men, many of which still have to swipe thousands of ten of thousands of times for a match?

​

Here's the code I used:

#Female Numbers
y_pred = [11, 23, 26, 22.5, 13, 4, 0.5]
y_true = [26.5, 30.5, 23.8, 12.5, 4.9, 1.7, 0.1]

#Male Numbers
y_pred = [1, 4, 9.5, 16, 24, 27.5, 18]
y_true = [6, 15.5, 18.5, 20, 19.5, 15, 5.5]

#RP Baseline - 20% of Men Get 80% of the Sex
y_pred = [80, 20]
y_true = [20, 80]

import numpy as np
def gini(array):
"""Calculate the Gini coefficient of a numpy array."""
# All values are treated equally, arrays must be 1d:
    array = array.flatten()
if np.amin(array) < 0:
# Values cannot be negative:
        array -= np.amin(array)
# Values cannot be 0:
    array += 0.0000001
# Values must be sorted:
    array = np.sort(array)
# Index per array element:
    index = np.arange(1,array.shape[0]+1)
# Number of array elements:
    n = array.shape[0]
# Gini coefficient:
return ((np.sum((2 * index - n  - 1) * array)) / (n * np.sum(array)))

import math
a = []
for i in range(len(y_pred)):
    factor = math.floor(y_true[i])
for j in range(factor):
        a.append([y_pred[i] / (100 * factor)])
a = np.array(a)
gini(a)